{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":52,"total_is_capped":false,"direct_labels_cover":0,"predictions_cover":52,"direct_label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline (scores rank; they never assert a category)","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12","author_layer_release":"2026-06-26"},"query_hash":"34b188fc4f0f","filters":{"venue":"Transactions of the Association for Computational Linguistics"}},"results":[{"id":"W2911227954","doi":"10.1162/tacl_a_00041","title":"Data Statements for Natural Language Processing: Toward Mitigating System Bias and Enabling Better Science","year":2018,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":808,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Macquarie University; York University; University of Washington; University of California, San Diego; National Science Foundation","keywords":"Embarrassment; Computer science; Natural (archaeology); Data science; Field (mathematics); Natural language; Lead (geology); Style (visual arts); Engineering ethics; Natural language processing; Psychology; Social psychology","authors":[{"name":"Emily M. Bender","is_ca":false},{"name":"Batya Friedman","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06724238139102369,"gpt":0.3486267136322955,"spread":0.2813843322412718,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.2543903,0.001285266,0.001519134,0.003926265,0.004139254,0.01144222,0.005334035,0.004844823,0.005221077],"category_scores_gemma":[0.4120362,0.002009786,0.001494599,0.003731263,0.01334146,0.03602486,0.0161058,0.01231708,0.002840397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003059554,"about_ca_system_score_gemma":0.01102469,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001372118,"about_ca_topic_score_gemma":0.0009106182,"domain_scores_codex":[0.7011495,0.2327128,0.01973961,0.01344035,0.0308386,0.002119228],"domain_scores_gemma":[0.3909179,0.3800997,0.03255863,0.1290555,0.06091979,0.006448459],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0009271836,0.0003164919,0.009593546,0.002263339,0.0001574796,0.0004584835,0.0173786,0.004005092,0.008293481,0.7733111,0.02601784,0.1572773],"study_design_scores_gemma":[0.0003561995,0.0004765967,0.001526683,0.001588329,0.0001688441,0.0004716024,0.003556426,0.03460449,0.02161915,0.6860071,0.2493735,0.0002509995],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006834033,0.0001822515,0.9646278,0.01934265,0.0004004842,0.0009162328,0.0003886867,0.002939006,0.004368837],"genre_scores_gemma":[0.1102897,0.0002772553,0.871415,0.00985027,0.0004542265,0.003254656,0.001099766,0.001282863,0.002076316],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.7456096,"threshold_uncertainty_score":0.9194695,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3151929433","doi":"10.1162/tacl_a_00360","title":"KEPLER: A Unified Model for Knowledge Embedding and Pre-trained Language Representation","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":602,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Canadian Institute for Advanced Research; HEC Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Kepler; Embedding; Benchmark (surveying); Language model; Representation (politics); Natural language processing; Construct (python library); ENCODE; Artificial intelligence; Programming language","authors":[{"name":"Xiaozhi Wang","is_ca":false},{"name":"Tianyu Gao","is_ca":false},{"name":"Zhaocheng Zhu","is_ca":true},{"name":"Zhengyan Zhang","is_ca":false},{"name":"Zhiyuan Liu","is_ca":false},{"name":"Juanzi Li","is_ca":false},{"name":"Jian Tang","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03147363115435537,"gpt":0.3240863137469095,"spread":0.2926126825925541,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001619438,0.001638541,0.001056161,0.001952822,0.0005481688,0.001807887,0.00348899,0.001979381,0.003844525],"category_scores_gemma":[0.008467794,0.0008509309,0.001598667,0.001977759,0.000810282,0.006360384,0.002909289,0.003685546,0.002657779],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001297045,"about_ca_system_score_gemma":0.001651878,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007042446,"about_ca_topic_score_gemma":0.0103277,"domain_scores_codex":[0.9989231,0.0003148703,0.00008170235,0.0004336472,0.0001429554,0.0001037442],"domain_scores_gemma":[0.9974995,0.001312425,0.0001670459,0.0005458388,0.0003765058,0.00009871783],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000286008,0.0003052568,0.00212704,0.0003351754,0.000237346,0.0002602575,0.0002888322,0.4736743,0.004539481,0.02031758,0.02366561,0.473963],"study_design_scores_gemma":[0.00001191673,0.0000245536,0.0001192538,0.00002077657,0.00002178006,0.00003280412,0.00001972442,0.9861498,0.001178379,0.01047352,0.001935419,0.00001213431],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01734339,0.001019423,0.9703542,0.0007331039,0.0001410118,0.0001401233,0.001671095,0.006764164,0.001833591],"genre_scores_gemma":[0.4433835,0.001413751,0.5243282,0.001041009,0.0002326722,0.0008719896,0.01626695,0.0009986747,0.01146322],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007042446,"threshold_uncertainty_score":0.01400292,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2906152891","doi":"10.1162/tacl_a_00254","title":"Analysis Methods in Neural Language Processing: A Survey","year":2019,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":486,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Categorization; Artificial neural network; Field (mathematics); Feature (linguistics); Artificial intelligence; Point (geometry); Data science; Natural language processing; Machine learning; Linguistics","authors":[{"name":"Yonatan Belinkov","is_ca":true},{"name":"James Glass","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02963772496761502,"gpt":0.3494461539490562,"spread":0.3198084289814412,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005247262,0.001175333,0.001503811,0.006815145,0.0005551925,0.003327871,0.002117141,0.001381191,0.003338611],"category_scores_gemma":[0.01358226,0.0006397034,0.001398879,0.008449441,0.001151046,0.004993781,0.001264353,0.001974873,0.00211179],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001394909,"about_ca_system_score_gemma":0.001737928,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002679691,"about_ca_topic_score_gemma":0.001500108,"domain_scores_codex":[0.9962387,0.001196243,0.0004216431,0.0006630212,0.001378633,0.0001018874],"domain_scores_gemma":[0.989958,0.00711563,0.0003745974,0.0005820197,0.001865727,0.0001040289],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00008815835,0.000117788,0.002904224,0.00364882,0.0002575865,0.00007869989,0.0002293752,0.005593345,0.001227578,0.04944742,0.01628727,0.9201197],"study_design_scores_gemma":[0.00007161051,0.0002138682,0.008603497,0.00473148,0.0004157707,0.001157758,0.0006612132,0.2498015,0.007050446,0.3166951,0.4103913,0.0002064527],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.004142226,0.507405,0.475009,0.003425075,0.0008518541,0.0001608496,0.0005316283,0.0009414137,0.007532984],"genre_scores_gemma":[0.09496226,0.5497673,0.3416665,0.001463459,0.004195713,0.0005342282,0.001602017,0.0006495239,0.005159032],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.006815145,"threshold_uncertainty_score":0.02775055,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2531207078","doi":"10.1162/tacl_a_00067","title":"Fully Character-Level Neural Machine Translation without Explicit Segmentation","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":415,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"York University; Samsung Advanced Institute of Technology; Samsung; Nvidia","keywords":"Computer science; Machine translation; Character (mathematics); Pooling; Encoder; Convolutional neural network; Artificial intelligence; Translation (biology); Natural language processing; Segmentation; Speech recognition; Task (project management); Language model; Representation (politics)","authors":[{"name":"Jason D. Lee","is_ca":false},{"name":"Kyunghyun Cho","is_ca":false},{"name":"Thomas Hofmann","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0388812594394734,"gpt":0.3152326302466266,"spread":0.2763513708071532,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004019013,0.0009828565,0.0006612442,0.0004598483,0.0003852232,0.0008666583,0.001083708,0.0008740707,0.006123744],"category_scores_gemma":[0.002153294,0.0003439613,0.0005900062,0.001021107,0.0003785697,0.001810606,0.0009074261,0.001024423,0.004576148],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005063899,"about_ca_system_score_gemma":0.00116801,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003681072,"about_ca_topic_score_gemma":0.007645729,"domain_scores_codex":[0.9995779,0.00008596748,0.00003321502,0.0001669619,0.00008662008,0.00004926735],"domain_scores_gemma":[0.9991353,0.0002345918,0.0000637335,0.0003126864,0.0002257725,0.00002789007],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005324047,0.0003176903,0.001353153,0.000491166,0.0001610785,0.0004859472,0.0001808849,0.1845673,0.09449791,0.0183189,0.02080486,0.6782887],"study_design_scores_gemma":[0.00002489029,0.0001186049,0.0005879584,0.00002126631,0.00003288747,0.000222429,0.00002794045,0.9396688,0.03822535,0.01116515,0.009879828,0.0000249355],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06779244,0.0007854792,0.9006131,0.0003423279,0.0002369238,0.0001127313,0.001358714,0.01466913,0.014089],"genre_scores_gemma":[0.6053543,0.0005226865,0.3657224,0.0003910598,0.000104567,0.0002041542,0.007608874,0.0009344073,0.01915754],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006123744,"threshold_uncertainty_score":0.020486,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3207937903","doi":"10.1162/tacl_a_00416","title":"MasakhaNER: Named Entity Recognition for African Languages","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":235,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo; Google (Canada)","funders":"","keywords":"Computer science; Named-entity recognition; Variety (cybernetics); Natural language processing; Artificial intelligence; Representation (politics); Code (set theory); Quality (philosophy); Data science; Information retrieval; Programming language; Task (project management); Political science","authors":[{"name":"David Ifeoluwa Adelani","is_ca":false},{"name":"Jade Abbott","is_ca":false},{"name":"Graham Neubig","is_ca":false},{"name":"Daniel D’souza","is_ca":false},{"name":"Julia Kreutzer","is_ca":true},{"name":"Constantine Lignos","is_ca":false},{"name":"Chester Palen-Michel","is_ca":false},{"name":"Happy Buzaaba","is_ca":false},{"name":"Shruti Rijhwani","is_ca":false},{"name":"Sebastian Ruder","is_ca":false},{"name":"Stephen Mayhew","is_ca":false},{"name":"Israel Abebe Azime","is_ca":false},{"name":"Shamsuddeen Hassan Muhammad","is_ca":false},{"name":"Chris Chinenye Emezue","is_ca":false},{"name":"Joyce Nakatumba‐Nabende","is_ca":false},{"name":"Perez Ogayo","is_ca":false},{"name":"Anuoluwapo Aremu","is_ca":false},{"name":"Catherine Gitau","is_ca":false},{"name":"Derguène Mbaye","is_ca":false},{"name":"Jesujoba O. Alabi","is_ca":false},{"name":"Seid Muhie Yimam","is_ca":false},{"name":"Tajuddeen Gwadabe","is_ca":false},{"name":"Ignatius Ezeani","is_ca":false},{"name":"Rubungo Andre Niyongabo","is_ca":false},{"name":"Jonathan Mukiibi","is_ca":false},{"name":"Verrah Otiende","is_ca":false},{"name":"Iroro Orife","is_ca":false},{"name":"Davis David","is_ca":false},{"name":"Samba Ngom","is_ca":false},{"name":"Tosin Adewumi","is_ca":false},{"name":"Paul Rayson","is_ca":false},{"name":"Mofetoluwa Adeyemi","is_ca":false},{"name":"Gerald Muriuki","is_ca":false},{"name":"Emmanuel Anebi","is_ca":false},{"name":"Chiamaka Chukwuneke","is_ca":false},{"name":"Nkiruka Odu","is_ca":false},{"name":"Eric Peter Wairagala","is_ca":false},{"name":"Samuel Oyerinde","is_ca":false},{"name":"Clemencia Siro","is_ca":false},{"name":"Tobius Saul Bateesa","is_ca":false},{"name":"Temilola Oloyede","is_ca":false},{"name":"Yvonne Wambui","is_ca":false},{"name":"Victor Akinode","is_ca":false},{"name":"Deborah Nabagereka","is_ca":false},{"name":"Maurice Katusiime","is_ca":false},{"name":"Ayodele Awokoya","is_ca":false},{"name":"Mouhamadane Mboup","is_ca":false},{"name":"Dibora Gebreyohannes","is_ca":false},{"name":"Henok Tilaye","is_ca":false},{"name":"Kelechi Nwaike","is_ca":false},{"name":"Degaga Wolde","is_ca":false},{"name":"Abdoulaye Faye","is_ca":false},{"name":"Blessing Sibanda","is_ca":false},{"name":"Orevaoghene Ahia","is_ca":false},{"name":"Bonaventure F. P. Dossou","is_ca":false},{"name":"Kelechi Ogueji","is_ca":true},{"name":"Thierno Ibrahima Diop","is_ca":false},{"name":"Abdoulaye Baniré Diallo","is_ca":false},{"name":"Adewale Akinfaderin","is_ca":false},{"name":"Tendai Marengereke","is_ca":false},{"name":"Salomey Osei","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0247748537854371,"gpt":0.274549278675741,"spread":0.2497744248903039,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004316772,0.002208157,0.001396557,0.004716551,0.002167372,0.002711142,0.002802291,0.001976445,0.01204532],"category_scores_gemma":[0.01106898,0.0007505238,0.001626501,0.003952262,0.0008489212,0.007309927,0.005194385,0.003049025,0.01528019],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001111183,"about_ca_system_score_gemma":0.002569595,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01183061,"about_ca_topic_score_gemma":0.01377255,"domain_scores_codex":[0.9968129,0.001133605,0.0003197939,0.0008801511,0.0005432417,0.0003103293],"domain_scores_gemma":[0.9954313,0.001547778,0.0003677019,0.001680911,0.0006103562,0.0003618619],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001796957,0.0008152204,0.01334305,0.002174138,0.0004458876,0.000940068,0.0009502239,0.01420509,0.01275374,0.01050303,0.7293999,0.2126728],"study_design_scores_gemma":[0.0009669065,0.0006499147,0.03058103,0.000705547,0.0004005299,0.002177175,0.001907332,0.26418,0.05402829,0.02406138,0.6198926,0.0004492187],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"methods","genre_scores_codex":[0.1273443,0.005116911,0.1005722,0.005284137,0.001912115,0.00180481,0.5555237,0.1842263,0.01821544],"genre_scores_gemma":[0.1003961,0.0009504109,0.1063696,0.0005847203,0.0002207571,0.001337104,0.7819177,0.001841533,0.00638212],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01204532,"threshold_uncertainty_score":0.0402956,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4312516176","doi":"10.1162/tacl_a_00511","title":"Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":194,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Columbia College","funders":"","keywords":"Causal inference; Computer science; Interpretability; Inference; Artificial intelligence; Causality (physics); Natural language processing; Robustness (evolution); Machine learning; Interpretation (philosophy); Data science; Econometrics","authors":[{"name":"Amir Feder","is_ca":false},{"name":"Katherine A. Keith","is_ca":false},{"name":"Emaad Manzoor","is_ca":false},{"name":"Reid Pryzant","is_ca":false},{"name":"Dhanya Sridhar","is_ca":true},{"name":"Zach Wood-Doughty","is_ca":false},{"name":"Jacob Eisenstein","is_ca":false},{"name":"Justin Grimmer","is_ca":false},{"name":"Roi Reichart","is_ca":false},{"name":"Margaret E. Roberts","is_ca":false},{"name":"Brandon Stewart","is_ca":false},{"name":"Victor Veitch","is_ca":false},{"name":"Diyi Yang","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.008751666553274558,"gpt":0.2654126314121237,"spread":0.2566609648588491,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.05218536,0.001321264,0.00256068,0.005578053,0.00187628,0.009268486,0.00361209,0.003157155,0.005739526],"category_scores_gemma":[0.2225595,0.001356072,0.002159739,0.005388177,0.01131175,0.01310184,0.005131652,0.007863645,0.0009160257],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003556392,"about_ca_system_score_gemma":0.003908806,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006428292,"about_ca_topic_score_gemma":0.003429423,"domain_scores_codex":[0.9587908,0.0330407,0.001319385,0.003592415,0.002836696,0.000419992],"domain_scores_gemma":[0.6363848,0.3413191,0.007009892,0.009663693,0.004891913,0.000730554],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.000126952,0.0000830975,0.007146084,0.0009523456,0.000466533,0.0002491954,0.001044122,0.03859647,0.0003530164,0.8501629,0.006194159,0.09462497],"study_design_scores_gemma":[0.00001336968,0.00000948745,0.0005197817,0.000228905,0.00003448012,0.00004119903,0.00006605686,0.0632639,0.0001482387,0.9326429,0.003011555,0.00002015631],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006548898,0.008914301,0.9654197,0.0140366,0.0002423743,0.00009624548,0.0004610413,0.0004408295,0.003840042],"genre_scores_gemma":[0.6168602,0.01481231,0.3547273,0.004995914,0.003607524,0.0006780988,0.001372011,0.0004345402,0.002512107],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.05218536,"threshold_uncertainty_score":0.2759859,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3137010024","doi":"10.1162/tacl_a_00447","title":"Quality at a Glance: An Audit of Web-Crawled Multilingual Datasets","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":167,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo; Google (Canada)","funders":"Agence Nationale de la Recherche","keywords":"Computer science; USable; Audit; Natural language processing; Quality (philosophy); Artificial intelligence; Information retrieval; World Wide Web; Accounting","authors":[{"name":"Julia Kreutzer","is_ca":true},{"name":"Isaac Caswell","is_ca":false},{"name":"Lisa Wang","is_ca":false},{"name":"Ahsan Wahab","is_ca":false},{"name":"Daan van Esch","is_ca":false},{"name":"Nasanbayar Ulzii-Orshikh","is_ca":false},{"name":"Allahsera Auguste Tapo","is_ca":false},{"name":"Nishant Subramani","is_ca":false},{"name":"Artem Sokolov","is_ca":false},{"name":"Claytone Sikasote","is_ca":false},{"name":"Monang Setyawan","is_ca":false},{"name":"Supheakmungkol Sarin","is_ca":false},{"name":"Sokhar Samb","is_ca":false},{"name":"Benoît Sagot","is_ca":false},{"name":"Clara E. Rivera","is_ca":false},{"name":"Annette Rios","is_ca":false},{"name":"Isabel Papadimitriou","is_ca":false},{"name":"Salomey Osei","is_ca":false},{"name":"Pedro Ortiz Suárez","is_ca":false},{"name":"Iroro Orife","is_ca":false},{"name":"Kelechi Ogueji","is_ca":true},{"name":"Andre Niyongabo Rubungo","is_ca":false},{"name":"Toan Nguyen","is_ca":false},{"name":"Mathias Müller","is_ca":false},{"name":"André Müller","is_ca":false},{"name":"Shamsuddeen Hassan Muhammad","is_ca":false},{"name":"N Sakib Muhammad","is_ca":false},{"name":"Ayanda Mnyakeni","is_ca":false},{"name":"Jamshidbek Mirzakhalov","is_ca":false},{"name":"Tapiwanashe Matangira","is_ca":true},{"name":"Colin Leong","is_ca":false},{"name":"Nze Lawson","is_ca":false},{"name":"Sneha Kudugunta","is_ca":false},{"name":"Yacine Jernite","is_ca":false},{"name":"Mathias Jenny","is_ca":false},{"name":"Orhan Fırat","is_ca":false},{"name":"Bonaventure F. P. Dossou","is_ca":false},{"name":"Sakhile Dlamini","is_ca":false},{"name":"Nisansa de Silva","is_ca":false},{"name":"Stella Biderman","is_ca":false},{"name":"Alessia Battisti","is_ca":false},{"name":"Ahmed Baruwa","is_ca":false},{"name":"Ankur Bapna","is_ca":false},{"name":"Pallavi Baljekar","is_ca":true},{"name":"Israel Abebe Azime","is_ca":false},{"name":"Ayodele Awokoya","is_ca":false},{"name":"Duygu Ataman","is_ca":false},{"name":"Orevaoghene Ahia","is_ca":false},{"name":"Oghenefego Ahia","is_ca":false},{"name":"Sweta Agrawal","is_ca":false},{"name":"Mofetoluwa Adeyemi","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02197055172116473,"gpt":0.3306852318074935,"spread":0.3087146800863287,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.06712028,0.001397105,0.00132719,0.01715647,0.00343512,0.007021653,0.003319533,0.001607887,0.001370947],"category_scores_gemma":[0.2108094,0.001615657,0.001202456,0.01735238,0.003590734,0.005904365,0.007036,0.002938509,0.003048406],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002321183,"about_ca_system_score_gemma":0.005634047,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02001247,"about_ca_topic_score_gemma":0.02837876,"domain_scores_codex":[0.8996423,0.02861887,0.01683081,0.01018274,0.04248993,0.002235285],"domain_scores_gemma":[0.4999964,0.1270439,0.03055844,0.1397169,0.1957293,0.006955071],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0021886,0.00153723,0.2703964,0.004656477,0.001347916,0.002693572,0.01417526,0.006669717,0.03967949,0.004891357,0.302007,0.349757],"study_design_scores_gemma":[0.0004033604,0.0008357639,0.4684419,0.002261519,0.0007260713,0.002622361,0.00593911,0.05185007,0.08352879,0.004778695,0.3778376,0.0007747139],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7020456,0.005848777,0.05948382,0.009382596,0.001744382,0.002667416,0.1168523,0.0827925,0.0191827],"genre_scores_gemma":[0.5110751,0.001643284,0.1286201,0.002577777,0.0003854511,0.00200854,0.3251124,0.01962509,0.00895226],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9328797,"threshold_uncertainty_score":0.3549702,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1794039122","doi":"10.1162/tacl_a_00080","title":"Learning to Understand Phrases by Embedding the Dictionary","year":2016,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":161,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Compute Canada","keywords":"Computer science; Natural language processing; Word embedding; Artificial intelligence; Bridging (networking); Embedding; Lexical semantics; Semantics (computer science); Word (group theory); Task (project management); Linguistics; Lexical item","authors":[{"name":"Felix Hill","is_ca":false},{"name":"Kyunghyun Cho","is_ca":false},{"name":"Anna Korhonen","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01137076964757506,"gpt":0.2784432967756493,"spread":0.2670725271280742,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005401591,0.0007588982,0.0003066454,0.0006497494,0.0002257549,0.0009960396,0.0008028085,0.0008797898,0.005467405],"category_scores_gemma":[0.003906179,0.0004331626,0.0006741954,0.000615159,0.0006714675,0.005583747,0.001356014,0.002013062,0.002377883],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003700841,"about_ca_system_score_gemma":0.0004691446,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001027856,"about_ca_topic_score_gemma":0.002093297,"domain_scores_codex":[0.9997166,0.00009447097,0.00001793562,0.0001195687,0.00003314924,0.00001832422],"domain_scores_gemma":[0.9989995,0.0006065253,0.00008055145,0.0001768279,0.0001063673,0.00003030793],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0001902101,0.0001740433,0.004908606,0.0006160474,0.000144282,0.0003283462,0.001191937,0.08137792,0.02711996,0.08898593,0.01514849,0.7798142],"study_design_scores_gemma":[0.00002974928,0.0001241093,0.001132598,0.00009771817,0.00005069407,0.0002995711,0.0003863598,0.807824,0.009063927,0.1654742,0.01548609,0.00003097106],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04986843,0.0004195559,0.9422755,0.0008810904,0.0001262456,0.00009159806,0.0006573438,0.001532428,0.004147825],"genre_scores_gemma":[0.5329585,0.0012617,0.4515588,0.0005519855,0.00012451,0.000252118,0.00347111,0.00041413,0.009407132],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005467405,"threshold_uncertainty_score":0.01829034,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2179519966","doi":"10.1162/tacl_a_00104","title":"Named Entity Recognition with Bidirectional LSTM-CNNs","year":2016,"lang":"en","type":"preprint","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":115,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"","keywords":"Computer science; Feature engineering; Feature (linguistics); Word (group theory); Artificial intelligence; Lexicon; Named-entity recognition; Task (project management); Natural language processing; Encoding (memory); State (computer science); Architecture; Entity linking; Deep learning; Knowledge base","authors":[{"name":"Jason Chiu","is_ca":true},{"name":"Eric Nichols","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02717742762268828,"gpt":0.2607877003144704,"spread":0.2336102726917821,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007392808,0.001031329,0.0005387194,0.001214123,0.0003586304,0.001149773,0.001563573,0.0008704101,0.00379048],"category_scores_gemma":[0.002085323,0.0004530503,0.0005700003,0.001883477,0.0002839684,0.004150379,0.001397214,0.0009036603,0.003478843],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008834726,"about_ca_system_score_gemma":0.0006941946,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007222301,"about_ca_topic_score_gemma":0.01262419,"domain_scores_codex":[0.9994825,0.0000797071,0.00004429719,0.0002094878,0.000106628,0.00007728513],"domain_scores_gemma":[0.9992834,0.0002051736,0.00009350174,0.000204279,0.0001887356,0.0000250026],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003268193,0.0001941055,0.002698951,0.0002794128,0.0001866385,0.0002933634,0.000131864,0.07590875,0.02977807,0.00873275,0.02900412,0.8524651],"study_design_scores_gemma":[0.00001935061,0.00005665546,0.001130266,0.00002878932,0.0000628322,0.0001226439,0.00005329175,0.9504978,0.02069166,0.01453636,0.01277192,0.00002844016],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07416549,0.002048611,0.8748081,0.0008700798,0.0004367475,0.0001637488,0.004744695,0.02781254,0.01495004],"genre_scores_gemma":[0.6360032,0.001156278,0.3277495,0.0006165208,0.0001905492,0.0002480962,0.01691094,0.000487703,0.01663726],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007222301,"threshold_uncertainty_score":0.01436049,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2570431255","doi":"10.1162/tacl_a_00077","title":"Aspect-augmented Adversarial Networks for Domain Adaptation","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":93,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"Army Research Office","keywords":"Computer science; Adversarial system; Domain adaptation; Classifier (UML); Artificial intelligence; Transfer of learning; Sentence; Training set; Domain (mathematical analysis); Relevance (law); Natural language processing; Machine learning; Invariant (physics)","authors":[{"name":"Yuan Zhang","is_ca":true},{"name":"Regina Barzilay","is_ca":true},{"name":"Tommi Jaakkola","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02543341488488058,"gpt":0.2719049047599099,"spread":0.2464714898750293,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001155514,0.001174908,0.0007708182,0.0005491431,0.0002576981,0.0007745214,0.001383874,0.000879649,0.002353759],"category_scores_gemma":[0.003794324,0.0003735103,0.0007593727,0.0007037246,0.0007265564,0.001420522,0.001951448,0.002241588,0.001126001],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005888284,"about_ca_system_score_gemma":0.000507691,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001275426,"about_ca_topic_score_gemma":0.001758937,"domain_scores_codex":[0.9994262,0.000221692,0.00002692244,0.0001440107,0.0001298454,0.00005128136],"domain_scores_gemma":[0.9990277,0.0004648296,0.00009732901,0.0002330008,0.0001323918,0.00004471513],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000171961,0.0001144857,0.001572256,0.0001372483,0.0001350797,0.0001764332,0.0001745366,0.7355163,0.008301347,0.03332645,0.009190292,0.2111837],"study_design_scores_gemma":[0.000005125274,0.00001887774,0.000114613,0.000006876458,0.000007505732,0.00003130525,0.000006539952,0.9847269,0.0009373505,0.01257219,0.001566447,0.000006243788],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009802936,0.0004252828,0.9858505,0.0002148425,0.00007480339,0.0000467896,0.0001502762,0.0009782031,0.002456237],"genre_scores_gemma":[0.6915725,0.001018249,0.2935808,0.0007031672,0.0002706828,0.000406338,0.001500462,0.0003959498,0.0105519],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002353759,"threshold_uncertainty_score":0.007874131,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2185701500","doi":"10.1162/tacl_a_00239","title":"Measuring Machine Translation Errors in New Domains","year":2013,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":77,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"National Research Council Canada","funders":"Defense Advanced Research Projects Agency; National Science Foundation","keywords":"Computer science; Machine translation; Natural language processing; Domain (mathematical analysis); Phrase; Machine translation software usability; Artificial intelligence; Evaluation of machine translation; Porting; Example-based machine translation; Rule-based machine translation; Translation (biology); Programming language","authors":[{"name":"Ann Irvine","is_ca":false},{"name":"John Morgan","is_ca":false},{"name":"Marine Carpuat","is_ca":true},{"name":"Hal Daumé","is_ca":false},{"name":"Dragos Stefan Munteanu","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02489167998879905,"gpt":0.2681288994135456,"spread":0.2432372194247465,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009788002,0.0008871593,0.000966699,0.002999207,0.0008601692,0.002039157,0.0009784834,0.001367425,0.001199489],"category_scores_gemma":[0.0958329,0.0005322027,0.0005606752,0.003844236,0.001404963,0.004118084,0.002535453,0.00218275,0.0008006678],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009975034,"about_ca_system_score_gemma":0.0007121334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002149454,"about_ca_topic_score_gemma":0.002159377,"domain_scores_codex":[0.9780115,0.009939123,0.002473002,0.002427395,0.006657641,0.0004914083],"domain_scores_gemma":[0.880864,0.0798969,0.01104394,0.008937624,0.01810442,0.001153129],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002622331,0.001783848,0.2217986,0.001720604,0.001108728,0.0009711884,0.009714782,0.1574432,0.1490065,0.00530687,0.003571769,0.4449516],"study_design_scores_gemma":[0.0001830368,0.003114075,0.3306259,0.000182629,0.0004841894,0.001409266,0.004379227,0.3523328,0.2838315,0.01263303,0.01049017,0.0003341504],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9202732,0.0005929952,0.07176959,0.0003876248,0.0000814173,0.0002346296,0.0005695323,0.0007559647,0.005335064],"genre_scores_gemma":[0.936502,0.0002487844,0.05959171,0.0001785269,0.00006242957,0.0002757897,0.001298233,0.0003394027,0.00150309],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009788002,"threshold_uncertainty_score":0.05176455,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2799424953","doi":"10.1162/tacl_a_00018","title":"Questionable Answers in Question Answering Research: Reproducibility and Variability of Published Results","year":2018,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Expert finding and Q&A systems","field":"Computer Science","cited_by":73,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo","funders":"","keywords":"Reproducibility; Computer science; Popularity; Reliability (semiconductor); Sample (material); Field (mathematics); Range (aeronautics); Code (set theory); Artificial intelligence; Data science; Statistics; Psychology; Mathematics; Social psychology","authors":[{"name":"Matt Crane","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03935335336173296,"gpt":0.3372124674052018,"spread":0.2978591140434689,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.5940362,0.001063335,0.002159154,0.009942214,0.00283769,0.01190427,0.006442023,0.004593777,0.002619948],"category_scores_gemma":[0.8792294,0.001750841,0.003239196,0.008855718,0.02007603,0.01066427,0.01139893,0.005461419,0.001026715],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.004891441,"about_ca_system_score_gemma":0.005183619,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00180553,"about_ca_topic_score_gemma":0.001863044,"domain_scores_codex":[0.2660926,0.5193769,0.06144322,0.04364005,0.1070568,0.002390423],"domain_scores_gemma":[0.03221424,0.8001816,0.03614707,0.103712,0.02684278,0.0009023547],"domain_codex":"methods","domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.003534528,0.0005050948,0.2064272,0.01683554,0.01279786,0.001555511,0.07247434,0.01113032,0.006307982,0.1407044,0.03080142,0.4969258],"study_design_scores_gemma":[0.0009503124,0.001053686,0.1446095,0.01601884,0.004067644,0.002833635,0.01259789,0.0252607,0.02486308,0.6349392,0.1318739,0.0009316446],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3148224,0.05308382,0.487966,0.08299384,0.006691295,0.002654075,0.004265961,0.002661117,0.04486143],"genre_scores_gemma":[0.9029801,0.003338563,0.07670691,0.009177021,0.001807487,0.001913457,0.001646637,0.001162835,0.001266972],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4059638,"threshold_uncertainty_score":0.5006257,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4226059645","doi":"10.1162/tacl_a_00471","title":"TopiOCQA: Open-domain Conversational Question Answering with Topic Switching","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":70,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":true},"ca_institutions":"Minnow Environmental (Canada); Research Canada; Microsoft (Canada); McGill University","funders":"Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Conversation; Computer science; Question answering; Open domain; Domain (mathematical analysis); Information retrieval; Interdependence; Natural language processing; Artificial intelligence; Relevance (law); Code (set theory); Linguistics; Set (abstract data type)","authors":[{"name":"Vaibhav Adlakha","is_ca":true},{"name":"Shehzaad Dhuliawala","is_ca":false},{"name":"Kaheer Suleman","is_ca":true},{"name":"Harm de Vries","is_ca":true},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01518758435261074,"gpt":0.2579240160992712,"spread":0.2427364317466605,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004193034,0.0024624,0.001371907,0.003432803,0.001971981,0.002659023,0.004410965,0.003063141,0.007699486],"category_scores_gemma":[0.02006466,0.0006903227,0.001921593,0.002437632,0.0009859382,0.005132341,0.005119776,0.003513007,0.006802582],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002440001,"about_ca_system_score_gemma":0.003103758,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03986457,"about_ca_topic_score_gemma":0.04586887,"domain_scores_codex":[0.9944724,0.002381542,0.0003820398,0.001635082,0.0007629418,0.0003658794],"domain_scores_gemma":[0.9894555,0.005302078,0.0004267942,0.002410132,0.001617444,0.0007880501],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00328068,0.001823583,0.01207427,0.004908916,0.0006358749,0.00101279,0.003071272,0.03275208,0.02015776,0.008756197,0.6344944,0.2770321],"study_design_scores_gemma":[0.001029072,0.0008591131,0.01538516,0.0004876065,0.0003217899,0.00119044,0.002260828,0.6801566,0.02359129,0.02129519,0.2530479,0.0003750591],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.2107721,0.01237772,0.2166013,0.005104836,0.001738594,0.005086524,0.3328103,0.1897411,0.02576763],"genre_scores_gemma":[0.291302,0.0008820947,0.2051602,0.001781717,0.000392421,0.003031041,0.4882387,0.001571522,0.007640294],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03986457,"threshold_uncertainty_score":0.07926506,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4398757454","doi":"10.1162/tacl_a_00667","title":"Evaluating Correctness and Faithfulness of Instruction-Following Models for Question Answering","year":2024,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":69,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true},"ca_institutions":"Canadian Institute for Advanced Research; McGill University; Mila - Quebec Artificial Intelligence Institute; Minnow Environmental (Canada); Research Canada","funders":"","keywords":"Correctness; Computer science; Question answering; Natural language processing; Artificial intelligence; Programming language","authors":[{"name":"Vaibhav Adlakha","is_ca":true},{"name":"Parishad BehnamGhader","is_ca":true},{"name":"Xing Han Lù","is_ca":true},{"name":"Nicholas Meade","is_ca":true},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03778902820660011,"gpt":0.3285510699647791,"spread":0.290762041758179,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.01909782,0.001115501,0.0009150107,0.001752439,0.0006676409,0.002926078,0.002302674,0.002493935,0.002085758],"category_scores_gemma":[0.1040018,0.0005551682,0.0008927653,0.0009017299,0.001645144,0.003426405,0.002425339,0.002598426,0.0009658962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002061284,"about_ca_system_score_gemma":0.001676693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008827999,"about_ca_topic_score_gemma":0.008541726,"domain_scores_codex":[0.9879096,0.007259812,0.0008078537,0.002161837,0.001418576,0.0004422038],"domain_scores_gemma":[0.8792539,0.09748355,0.004336442,0.01080103,0.006468599,0.001656582],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00448744,0.001165706,0.06192508,0.001209271,0.0006914596,0.00030912,0.00464162,0.5150812,0.02795876,0.009305319,0.01128441,0.3619407],"study_design_scores_gemma":[0.00004629346,0.0002981313,0.00293184,0.00004250773,0.00005636044,0.00006380582,0.0001859486,0.9819347,0.009088089,0.004488131,0.0008302592,0.0000339819],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6171094,0.001053758,0.3619538,0.001319619,0.0001916272,0.0006637375,0.001097423,0.01213438,0.004476221],"genre_scores_gemma":[0.9459374,0.00008363485,0.05154453,0.0001741629,0.00002712459,0.0001564039,0.00108523,0.0003100999,0.0006813122],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9809022,"threshold_uncertainty_score":0.1010001,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2769298630","doi":"10.1162/tacl_a_00011","title":"Modeling Past and Future for Neural Machine Translation","year":2018,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":54,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo","funders":"National Natural Science Foundation of China; National Science Foundation","keywords":"Machine translation; Computer science; Decoding methods; Translation (biology); German; Artificial intelligence; Natural language processing; Mechanism (biology); Speech recognition; Machine learning; Algorithm; Linguistics","authors":[{"name":"Zaixiang Zheng","is_ca":false},{"name":"Hao Zhou","is_ca":false},{"name":"Shujian Huang","is_ca":false},{"name":"Lili Mou","is_ca":true},{"name":"Xinyu Dai","is_ca":false},{"name":"Jiajun Chen","is_ca":false},{"name":"Zhaopeng Tu","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01634169044876612,"gpt":0.2815328401993656,"spread":0.2651911497505995,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009135964,0.000661927,0.000573656,0.000706746,0.0005064672,0.00103737,0.001139442,0.000997578,0.003497413],"category_scores_gemma":[0.003884787,0.0004396403,0.0007793237,0.0009936851,0.0006861368,0.003852327,0.0009964558,0.001496551,0.001196276],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007702792,"about_ca_system_score_gemma":0.0008198313,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005292531,"about_ca_topic_score_gemma":0.008616557,"domain_scores_codex":[0.9996451,0.000112571,0.00003088503,0.0001269891,0.00005607917,0.0000283491],"domain_scores_gemma":[0.9990958,0.0004801086,0.00009373979,0.0001722119,0.0001296467,0.00002842471],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006674564,0.0001398943,0.004002566,0.0006286479,0.0002283574,0.0006800281,0.0009058699,0.3766472,0.02355266,0.1370159,0.008923014,0.4466084],"study_design_scores_gemma":[0.00001538698,0.00004534801,0.000562121,0.00003487427,0.0000686663,0.0001542176,0.00003344014,0.9291243,0.003420334,0.05931395,0.007203102,0.00002426573],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04594436,0.004180486,0.9402043,0.001267262,0.0002635128,0.00003593666,0.0005661241,0.001776752,0.005761333],"genre_scores_gemma":[0.8307512,0.002509054,0.1570445,0.0003989066,0.0002552245,0.0001509475,0.00142869,0.0002499579,0.007211437],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005292531,"threshold_uncertainty_score":0.01170003,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4307680525","doi":"10.1162/tacl_a_00545","title":"Generative Spoken Dialogue Language Modeling","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":49,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Centre National de la Recherche Scientifique; Agence Nationale de la Recherche; École des Hautes Etudes en Sciences Sociales; Canadian Institute for Advanced Research","keywords":"Paralanguage; Computer science; Transformer; Spoken language; Generative grammar; Speech recognition; Natural language processing; Laughter; Language model; Artificial intelligence; Communication; Psychology","authors":[{"name":"Tu Anh Nguyen","is_ca":false},{"name":"Eugene Kharitonov","is_ca":false},{"name":"Jade Copet","is_ca":false},{"name":"Yossi Adi","is_ca":false},{"name":"Wei-Ning Hsu","is_ca":false},{"name":"Ali Elkahky","is_ca":false},{"name":"Paden Tomasello","is_ca":false},{"name":"Robin Algayres","is_ca":false},{"name":"Benoît Sagot","is_ca":false},{"name":"Abdelrahman Mohamed","is_ca":false},{"name":"Emmanuel Dupoux","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03106876500409547,"gpt":0.2777580017736145,"spread":0.246689236769519,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004964013,0.0007142774,0.0005202665,0.0004556214,0.0002579058,0.0008967351,0.001585908,0.001001605,0.006105737],"category_scores_gemma":[0.00187603,0.0004477058,0.001089799,0.0002956364,0.0005971885,0.0007330337,0.001227129,0.001312481,0.001485628],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006583724,"about_ca_system_score_gemma":0.0005910396,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003346221,"about_ca_topic_score_gemma":0.003912078,"domain_scores_codex":[0.9995884,0.0001656848,0.00001480852,0.0001215123,0.00007110782,0.00003851092],"domain_scores_gemma":[0.9994465,0.0003549307,0.00002593448,0.00006309454,0.00007718395,0.00003228414],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001182076,0.00006031333,0.0008217345,0.0001256185,0.00009771445,0.0002338466,0.0003604821,0.8972139,0.005610975,0.03670106,0.004247938,0.05440823],"study_design_scores_gemma":[0.000005398521,0.000006064155,0.00003698372,0.000003512245,0.000003794134,0.00001607884,0.00000701379,0.9925356,0.0004099288,0.006069771,0.0009023522,0.000003478873],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02300226,0.0004296591,0.9671104,0.000496818,0.0001210391,0.00006603313,0.0008372,0.002569695,0.00536694],"genre_scores_gemma":[0.8106447,0.0002849512,0.1717802,0.0003824884,0.0001365318,0.0002818432,0.00211138,0.0005517321,0.01382609],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006105737,"threshold_uncertainty_score":0.0204258,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4296711106","doi":"10.1162/tacl_a_00506","title":"Evaluating Attribution in Dialogue Systems: The BEGIN Benchmark","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":44,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Spurious relationship; Benchmark (surveying); Attribution; Artificial intelligence; Natural language processing; Grounded theory; Data science; Machine learning; Qualitative research","authors":[{"name":"Nouha Dziri","is_ca":true},{"name":"Hannah Rashkin","is_ca":false},{"name":"Tal Linzen","is_ca":false},{"name":"David Reitter","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04673618613088275,"gpt":0.3095691801184852,"spread":0.2628329939876024,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01285615,0.002883431,0.001314166,0.003055743,0.001204005,0.002874946,0.002531046,0.002646512,0.003888877],"category_scores_gemma":[0.05178178,0.0005784282,0.001092024,0.001841704,0.001409472,0.003660867,0.004786846,0.003014585,0.003333192],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001877344,"about_ca_system_score_gemma":0.001374885,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005043851,"about_ca_topic_score_gemma":0.007288284,"domain_scores_codex":[0.9777314,0.01477629,0.001103936,0.003178342,0.002483393,0.0007266432],"domain_scores_gemma":[0.9645017,0.02167398,0.001637895,0.005679882,0.004821747,0.001684825],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.007725429,0.004401443,0.05878584,0.0056298,0.001587345,0.000884786,0.005734972,0.2126558,0.0300513,0.008207325,0.1264062,0.5379299],"study_design_scores_gemma":[0.001029247,0.005047627,0.05493587,0.0007914133,0.0003916781,0.0007555475,0.004108924,0.7847366,0.05995833,0.02215948,0.06562322,0.000462079],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.73103,0.01103184,0.1562697,0.001589743,0.001599192,0.001922347,0.02971164,0.04240571,0.02443974],"genre_scores_gemma":[0.8645595,0.0006093174,0.06819958,0.000445379,0.0002618006,0.001152592,0.05792588,0.001590389,0.005255618],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01285615,"threshold_uncertainty_score":0.06799066,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4386488973","doi":"10.1162/tacl_a_00595","title":"<b>MIRACL</b>: A Multilingual Retrieval Dataset Covering 18 Diverse Languages","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Huawei Technologies (Canada); University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Annotation; Relevance (law); Natural language processing; Information retrieval; Process (computing); Quality (philosophy); Artificial intelligence; Resource (disambiguation); World Wide Web","authors":[{"name":"Xinyu Zhang","is_ca":true},{"name":"Nandan Thakur","is_ca":true},{"name":"Odunayo Ogundepo","is_ca":true},{"name":"Ehsan Kamalloo","is_ca":true},{"name":"David Alfonso-Hermelo","is_ca":true},{"name":"Xiaoguang Li","is_ca":false},{"name":"Mehdi Rezagholizadeh","is_ca":true},{"name":"Jimmy Lin","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02744140968399567,"gpt":0.3310232120994008,"spread":0.3035818024154052,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002839488,0.002343049,0.001558488,0.008977263,0.002243018,0.003139624,0.002653157,0.002979386,0.02303038],"category_scores_gemma":[0.0118616,0.0007697366,0.001334622,0.006833417,0.0009814847,0.003448456,0.003641498,0.001993055,0.04389751],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001785306,"about_ca_system_score_gemma":0.00242796,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03527757,"about_ca_topic_score_gemma":0.05765631,"domain_scores_codex":[0.9957322,0.001067526,0.0007437337,0.0009734106,0.001069911,0.0004130991],"domain_scores_gemma":[0.9931892,0.001568347,0.0004701878,0.001318241,0.002697692,0.0007563895],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0005033853,0.0002596586,0.003980467,0.002816411,0.0001622191,0.0003284547,0.0004677125,0.0009957766,0.007957336,0.001219761,0.9553984,0.02591049],"study_design_scores_gemma":[0.0008147034,0.0003360078,0.02616134,0.0005403364,0.0001891932,0.001312417,0.001513417,0.01182299,0.01213676,0.002171007,0.9426841,0.0003178157],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.01904167,0.001393051,0.004018412,0.0005867735,0.0002587639,0.0004399314,0.9533213,0.01217182,0.008768313],"genre_scores_gemma":[0.009179716,0.00009346559,0.006551635,0.0001332041,0.00002856902,0.0002611907,0.9817632,0.0003384025,0.001650638],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.03527757,"threshold_uncertainty_score":0.07704425,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3196986263","doi":"10.1162/tacl_a_00519","title":"Neuron-level Interpretation of Deep NLP Models: A Survey","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Dalhousie University","funders":"","keywords":"Interpretability; Computer science; Representation (politics); Artificial intelligence; Domain (mathematical analysis); Adaptation (eye); Interpretation (philosophy); Artificial neural network; Domain adaptation; Machine learning; Natural language processing; Data science; Neuroscience","authors":[{"name":"Hassan Sajjad","is_ca":true},{"name":"Nadir Durrani","is_ca":false},{"name":"Fahim Dalvi","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05485884523086224,"gpt":0.2844312671001392,"spread":0.229572421869277,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002113833,0.0007394412,0.0008117456,0.002120973,0.0002968993,0.003687259,0.002609099,0.001281054,0.003743128],"category_scores_gemma":[0.006304968,0.0005866983,0.001087456,0.002164184,0.001510869,0.00472966,0.001697224,0.00194673,0.0009447109],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001718071,"about_ca_system_score_gemma":0.001137517,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002487748,"about_ca_topic_score_gemma":0.002314282,"domain_scores_codex":[0.9987226,0.0005291657,0.0001095022,0.0002046199,0.0003767159,0.0000573249],"domain_scores_gemma":[0.9975198,0.001553963,0.0001506256,0.0003933492,0.0003238124,0.00005838914],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001223528,0.0001273067,0.004502519,0.00178251,0.0002873027,0.0003932693,0.0009402689,0.1768086,0.003197445,0.3638027,0.008139981,0.4398957],"study_design_scores_gemma":[0.000009143918,0.00003840071,0.0008404394,0.0005002494,0.00004003998,0.0002091518,0.0002488512,0.5263513,0.002261468,0.4440272,0.02543934,0.00003438543],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.02029447,0.03293628,0.9260633,0.004829828,0.0001560254,0.00008171075,0.0008021987,0.001304092,0.01353209],"genre_scores_gemma":[0.6195332,0.05068413,0.319244,0.001039293,0.0004483449,0.0001920197,0.002043397,0.0005978137,0.006217812],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.003743128,"threshold_uncertainty_score":0.01252204,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4287120901","doi":"10.1162/tacl_a_00492","title":"Generate, Annotate, and Learn: NLP with Synthetic Text","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":37,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Google (Canada)","funders":"Defense Advanced Research Projects Agency","keywords":"Computer science; Artificial intelligence; Transformer; Natural language processing; Classifier (UML); Machine learning; Labeled data; Task (project management); Distillation; Language model","authors":[{"name":"Xuanli He","is_ca":false},{"name":"Islam Nassar","is_ca":false},{"name":"Jamie Kiros","is_ca":true},{"name":"Gholamreza Haffari","is_ca":false},{"name":"Mohammad Norouzi","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01282966128692613,"gpt":0.2278401156433895,"spread":0.2150104543564634,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002707636,0.0008602803,0.0004802741,0.0005363958,0.0005635999,0.001047708,0.001452629,0.001258601,0.004044543],"category_scores_gemma":[0.0154307,0.0003060373,0.0005009374,0.0005995124,0.001289586,0.00317167,0.001912209,0.001857437,0.00167907],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008224361,"about_ca_system_score_gemma":0.0006645097,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002102799,"about_ca_topic_score_gemma":0.003657861,"domain_scores_codex":[0.9982247,0.001002507,0.00005676873,0.0003669994,0.0002757785,0.00007326284],"domain_scores_gemma":[0.9900479,0.007523117,0.0002412275,0.00138134,0.0006321153,0.0001742586],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001051926,0.0006229797,0.003493272,0.0006655763,0.000131244,0.0005417097,0.0008018955,0.5767578,0.01292091,0.0633667,0.0351302,0.3045157],"study_design_scores_gemma":[0.00005055728,0.00005155817,0.0001548217,0.00001680769,0.000006946076,0.00004028418,0.00006872984,0.9605917,0.009361397,0.02536711,0.004277063,0.00001299745],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1763613,0.0007228936,0.7923977,0.002251224,0.0004031529,0.0002820592,0.003608165,0.01317655,0.01079697],"genre_scores_gemma":[0.6470042,0.0001651419,0.3397616,0.0004123927,0.00009916467,0.0003240173,0.00590084,0.0007953396,0.005537218],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004044543,"threshold_uncertainty_score":0.01431948,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3198711991","doi":"10.1162/tacl_a_00478","title":"It’s not Rocket Science: Interpreting Figurative Language in Narratives","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":35,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"","keywords":"Literal and figurative language; Computer science; Natural language processing; Principle of compositionality; Narrative; Generative grammar; Interpretation (philosophy); Artificial intelligence; Linguistics; Context (archaeology); Expression (computer science); Programming language; History","authors":[{"name":"Tuhin Chakrabarty","is_ca":false},{"name":"Yejin Choi","is_ca":false},{"name":"Vered Shwartz","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01209388290123788,"gpt":0.3034295858987213,"spread":0.2913357029974835,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001988821,0.0009794856,0.0002355631,0.001015928,0.000521685,0.00265385,0.001045999,0.001309478,0.003542212],"category_scores_gemma":[0.01511593,0.0003454076,0.0007225164,0.0006158512,0.001275472,0.004880017,0.001058235,0.001432217,0.001306471],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001142225,"about_ca_system_score_gemma":0.0006394879,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004974572,"about_ca_topic_score_gemma":0.008166929,"domain_scores_codex":[0.9988996,0.0007053089,0.00004643949,0.0002270734,0.00007719911,0.00004442382],"domain_scores_gemma":[0.9926323,0.005790607,0.0004418458,0.0007236254,0.0002859941,0.000125611],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001762758,0.0006217144,0.09854395,0.003207957,0.0004323576,0.004328845,0.05103011,0.1519861,0.03640646,0.06904928,0.0457736,0.5368568],"study_design_scores_gemma":[0.0001005287,0.0002337553,0.02136479,0.0007934897,0.0001329282,0.002029757,0.01119916,0.7836898,0.02977175,0.06796018,0.08251569,0.0002080661],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7994957,0.002372486,0.1623082,0.00400672,0.0002805963,0.0003472951,0.007301,0.004466206,0.01942169],"genre_scores_gemma":[0.9297994,0.0003717009,0.06238274,0.0002950877,0.00003712642,0.00006407498,0.005153553,0.0001575418,0.001738876],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004974572,"threshold_uncertainty_score":0.01184988,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4381686872","doi":"10.1162/tacl_a_00564","title":"Questions Are All You Need to Train a Dense Passage Retriever","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"DeepMind","keywords":"Computer science; Initialization; Code (set theory); Task (project management); Set (abstract data type); Encoder; Artificial intelligence; Information retrieval; Scheme (mathematics); Training set; Domain (mathematical analysis); Language model; Natural language processing; Machine learning","authors":[{"name":"Devendra Singh Sachan","is_ca":true},{"name":"Dani Yogatama","is_ca":false},{"name":"Luke Zettlemoyer","is_ca":false},{"name":"Joëlle Pineau","is_ca":true},{"name":"Manzil Zaheer","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03032712766975569,"gpt":0.2833717000899555,"spread":0.2530445724201998,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002263985,0.001100049,0.001148879,0.0007977537,0.0005332754,0.001715174,0.002097774,0.001820661,0.01485285],"category_scores_gemma":[0.009221409,0.0006025795,0.001326764,0.0005689844,0.0009781343,0.004097463,0.002457244,0.003081314,0.01077997],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007822199,"about_ca_system_score_gemma":0.001050461,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004415294,"about_ca_topic_score_gemma":0.004821027,"domain_scores_codex":[0.9987608,0.0003698583,0.00007610425,0.0004994872,0.0001957342,0.00009803634],"domain_scores_gemma":[0.9974956,0.001244298,0.00007598365,0.0007193187,0.0003563901,0.000108358],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008348469,0.0003371884,0.003057024,0.0008420313,0.0002482712,0.0005666047,0.000955815,0.07300904,0.05269521,0.01695954,0.06127658,0.7892178],"study_design_scores_gemma":[0.0001200869,0.0002791912,0.001112724,0.00009616559,0.000127176,0.0005483529,0.0002673999,0.8834192,0.04992708,0.02814982,0.03587086,0.0000818731],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03092643,0.0009200069,0.915548,0.001038949,0.0003632802,0.0003319547,0.002370875,0.04344815,0.005052343],"genre_scores_gemma":[0.3647412,0.0004812223,0.604005,0.001632871,0.0003314927,0.0005021151,0.01096152,0.003015699,0.01432883],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01485285,"threshold_uncertainty_score":0.04968774,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2605118633","doi":"10.1162/tacl_a_00047","title":"A Generative Model of Phonotactics","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":28,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"National Science Foundation","keywords":"Phonotactics; Computer science; Generative grammar; Artificial intelligence; Generative model; Feature (linguistics); Natural language processing; Set (abstract data type); Probabilistic logic; Phonology; Hierarchy; Linguistics; Programming language","authors":[{"name":"Richard Futrell","is_ca":false},{"name":"Adam Albright","is_ca":false},{"name":"Peter Graff","is_ca":false},{"name":"Timothy J. O’Donnell","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02717737201804962,"gpt":0.3050787880426549,"spread":0.2779014160246053,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0008487001,0.0006577195,0.0008159427,0.001215087,0.0006652,0.002076363,0.0019428,0.00151388,0.007475665],"category_scores_gemma":[0.00446942,0.0008622581,0.001635949,0.001217455,0.001403302,0.00320331,0.001540737,0.001966697,0.001697197],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001035,"about_ca_system_score_gemma":0.001033271,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007385026,"about_ca_topic_score_gemma":0.009196887,"domain_scores_codex":[0.9994575,0.0001703572,0.0000269048,0.0001762317,0.0001007379,0.00006814725],"domain_scores_gemma":[0.9986345,0.0008625222,0.0001064165,0.0002135088,0.0001224327,0.00006061276],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00011711,0.00007617051,0.005626055,0.0001356998,0.0001073439,0.0004244125,0.001027588,0.5389388,0.006493406,0.389137,0.006781249,0.05113518],"study_design_scores_gemma":[0.00001518102,0.00001631292,0.0005902626,0.00001902928,0.00002333888,0.0001842206,0.00003432056,0.8888471,0.0004636635,0.1063466,0.003438502,0.00002160227],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.04691969,0.0003855143,0.9378238,0.0009921156,0.00008644919,0.00007673486,0.001498654,0.001350321,0.01086671],"genre_scores_gemma":[0.8474216,0.0007016422,0.1324614,0.0005059533,0.0001538346,0.0003736386,0.002073537,0.0006821608,0.01562621],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007475665,"threshold_uncertainty_score":0.02500862,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4313459318","doi":"10.1162/tacl_a_00529","title":"<scp>FaithDial</scp>: A Faithful Benchmark for Information-Seeking Dialogue","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":27,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; University of Alberta","funders":"Alberta Machine Intelligence Institute; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Benchmark (surveying); Computer science; Utterance; Hallucinating; Natural language processing; Artificial intelligence; Crowdsourcing; Machine learning; World Wide Web","authors":[{"name":"Nouha Dziri","is_ca":true},{"name":"Ehsan Kamalloo","is_ca":true},{"name":"Sivan Milton","is_ca":true},{"name":"Osmar R. Zai͏̈ane","is_ca":true},{"name":"Mo Yu","is_ca":false},{"name":"Edoardo Maria Ponti","is_ca":false},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01280205941306425,"gpt":0.2332753737500383,"spread":0.220473314336974,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006931292,0.002594201,0.001037396,0.002702551,0.001745446,0.00275691,0.002896275,0.00326904,0.006216775],"category_scores_gemma":[0.02604859,0.0004618994,0.001275964,0.001293218,0.001696641,0.003148438,0.004144596,0.002948992,0.004460691],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001729801,"about_ca_system_score_gemma":0.001357193,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0092393,"about_ca_topic_score_gemma":0.01411737,"domain_scores_codex":[0.9907134,0.005456122,0.0004929999,0.001652681,0.001253155,0.0004316609],"domain_scores_gemma":[0.9827779,0.009158574,0.0006596957,0.003460013,0.002586874,0.001356835],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.006063689,0.004719162,0.01732304,0.005191254,0.0009193003,0.001523089,0.003971925,0.1258333,0.04382412,0.008591704,0.2475931,0.5344463],"study_design_scores_gemma":[0.0009337985,0.003904772,0.02005586,0.0004995764,0.0002050007,0.001377043,0.003203186,0.7813051,0.07424955,0.0169471,0.09688436,0.0004347437],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6336607,0.00894468,0.1675692,0.00377072,0.003110057,0.002551862,0.05512027,0.08624581,0.03902667],"genre_scores_gemma":[0.7580103,0.0006240549,0.1113841,0.001019298,0.0004117765,0.001138922,0.1132558,0.002625508,0.01153028],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0092393,"threshold_uncertainty_score":0.03665662,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2404429704","doi":"10.1162/tacl_a_00084","title":"Decoding Anagrammed Texts Written in an Unknown Language and Script","year":2016,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Intelligence, Security, War Strategy","field":"Social Sciences","cited_by":24,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"Natural Sciences and Engineering Research Council of Canada; Alberta Innovates","keywords":"Decipherment; Computer science; Hebrew; Substitution (logic); Decoding methods; Natural language processing; Set (abstract data type); Cipher; Artificial intelligence; Identification (biology); Encryption; Speech recognition; Linguistics; Algorithm; Programming language","authors":[{"name":"Bradley Hauer","is_ca":true},{"name":"Grzegorz Kondrak","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02373837580185808,"gpt":0.3242809997761862,"spread":0.3005426239743281,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001298824,0.0006858622,0.0004989371,0.0009158149,0.0008486649,0.002051898,0.0007470007,0.0009831894,0.002467468],"category_scores_gemma":[0.01269244,0.0002685128,0.0003952844,0.0006017634,0.001532844,0.002536217,0.0009412519,0.001027574,0.003136323],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005298508,"about_ca_system_score_gemma":0.0009402023,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008829227,"about_ca_topic_score_gemma":0.00161049,"domain_scores_codex":[0.9985104,0.0006179916,0.0001389447,0.0004169859,0.000245739,0.00006996132],"domain_scores_gemma":[0.992821,0.004408506,0.0005197407,0.001240595,0.0009156528,0.00009444065],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005832792,0.0001325685,0.01082744,0.0005844165,0.0001379701,0.0008648733,0.003129459,0.03428008,0.1017564,0.04476913,0.00471914,0.7982153],"study_design_scores_gemma":[0.00003608543,0.0004056059,0.008052859,0.00023393,0.00007092295,0.00263954,0.003039446,0.5278667,0.3152809,0.1143297,0.02791671,0.0001275961],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2431383,0.0004370572,0.7456625,0.0008047718,0.0001291775,0.0001535559,0.0005731006,0.001875453,0.007226035],"genre_scores_gemma":[0.63114,0.0003748235,0.3604667,0.0001545292,0.0000820043,0.00006981606,0.001470837,0.0003039067,0.005937332],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002467468,"threshold_uncertainty_score":0.008254468,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4317897852","doi":"10.1162/tacl_a_00539","title":"Cross-Lingual Dialogue Dataset Creation via Outline-Based Generation","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":23,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Canadian Institute for Advanced Research","keywords":"Computer science; Naturalness; Natural language processing; Machine translation; Artificial intelligence; Modular design; Process (computing); Annotation; Benchmark (surveying); Information retrieval; Programming language","authors":[{"name":"Olga Majewska","is_ca":false},{"name":"Evgeniia Razumovskaia","is_ca":false},{"name":"Edoardo Maria Ponti","is_ca":false},{"name":"Ivan Vulić","is_ca":false},{"name":"Anna Korhonen","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03988226869889792,"gpt":0.3194351972184415,"spread":0.2795529285195436,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005768458,0.00156011,0.0008414801,0.003025856,0.00195977,0.002244485,0.002667047,0.001548953,0.005085057],"category_scores_gemma":[0.01840143,0.0004998351,0.001604014,0.001702635,0.001114367,0.002856352,0.006356001,0.002460239,0.005129466],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001149979,"about_ca_system_score_gemma":0.002050341,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005742458,"about_ca_topic_score_gemma":0.01121709,"domain_scores_codex":[0.9924836,0.003042683,0.0007109788,0.002334582,0.001106025,0.0003222126],"domain_scores_gemma":[0.9879822,0.003635942,0.0005749054,0.003953732,0.003168021,0.0006852178],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002467707,0.001741888,0.02037587,0.004620685,0.0005669213,0.001857074,0.0100999,0.02943067,0.06755533,0.01900468,0.3192295,0.5230498],"study_design_scores_gemma":[0.0005220053,0.0007840219,0.03279242,0.0008797183,0.0002811958,0.001466576,0.007717704,0.2236685,0.07518837,0.02057376,0.6355611,0.0005645612],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1761277,0.002719729,0.4918461,0.001997336,0.002347692,0.003719879,0.1904838,0.09455238,0.03620532],"genre_scores_gemma":[0.2140879,0.000299638,0.3621267,0.0005775226,0.0001696981,0.003712425,0.4091086,0.002900633,0.00701694],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005768458,"threshold_uncertainty_score":0.03050691,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3183195595","doi":"10.1162/tacl_a_00386","title":"Context-aware Adversarial Training for Name Regularity Bias in Named Entity Recognition","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":21,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; Huawei Technologies (Canada)","funders":"","keywords":"Adversarial system; Focus (optics); Testbed; Named-entity recognition; Training set; Noise (video); Training (meteorology)","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.0861111343750003,"gpt":0.2954547619652698,"spread":0.2093436275902695,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002914476,0.0009759403,0.0008158521,0.0003711851,0.0003697433,0.0005509856,0.001601281,0.00114702,0.001725665],"category_scores_gemma":[0.00915727,0.0004839032,0.0004993653,0.0003432179,0.001011129,0.001833241,0.001711026,0.00280716,0.0006564104],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005425994,"about_ca_system_score_gemma":0.0005648591,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002651697,"about_ca_topic_score_gemma":0.003688562,"domain_scores_codex":[0.9990292,0.0004761548,0.00004290468,0.0002603734,0.00009831541,0.00009307304],"domain_scores_gemma":[0.9948398,0.003737454,0.0002177967,0.0008696502,0.000230901,0.0001044074],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003749648,0.0001475016,0.002584224,0.0001265018,0.0001047659,0.0001300699,0.0001056794,0.8866991,0.01152151,0.003537005,0.003360393,0.09130825],"study_design_scores_gemma":[0.000006172966,0.00003882398,0.0002304635,0.00001052591,0.000008682974,0.00002245006,0.000008732326,0.9934961,0.004018182,0.001803408,0.0003491319,0.000007415173],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2717924,0.001704292,0.7144756,0.0008832896,0.0003390352,0.0000929185,0.0005567345,0.006004309,0.004151265],"genre_scores_gemma":[0.9379156,0.0001691864,0.05900348,0.0003001993,0.00006552102,0.0000501406,0.0006612039,0.0002472516,0.001587427],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002914476,"threshold_uncertainty_score":0.0154134,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4394647506","doi":"10.1162/tacl_a_00670","title":"Scope Ambiguities in Large Language Models","year":2024,"lang":"en","type":"preprint","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":20,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; National Research Council Canada; Mila - Quebec Artificial Intelligence Institute","funders":"Fonds de Recherche du Québec-Société et Culture","keywords":"Scope (computer science); Computer science; Linguistics; Cognitive science; Psychology; Programming language; Philosophy","authors":[{"name":"Gaurav Kamath","is_ca":true},{"name":"Sebastian Schuster","is_ca":false},{"name":"Sowmya Vajjala","is_ca":true},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02432477583018319,"gpt":0.2884349640007543,"spread":0.2641101881705711,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01214811,0.0009581585,0.00114541,0.002415678,0.0008754414,0.003200091,0.001176603,0.00117239,0.001467629],"category_scores_gemma":[0.05126545,0.0007042747,0.001264532,0.002281595,0.001747151,0.004489352,0.001616941,0.002625595,0.0003685952],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001528014,"about_ca_system_score_gemma":0.0009840496,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007378241,"about_ca_topic_score_gemma":0.00832256,"domain_scores_codex":[0.9911237,0.006212833,0.0003981973,0.001172444,0.0008971487,0.0001958029],"domain_scores_gemma":[0.931223,0.06177311,0.002961238,0.002255571,0.001380658,0.000406449],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0006892596,0.0001953489,0.02791137,0.0005172478,0.0004694706,0.001080486,0.004681053,0.7596234,0.004019022,0.1069712,0.007330769,0.08651127],"study_design_scores_gemma":[0.00002036531,0.00002158745,0.001482938,0.00003600601,0.00002620965,0.00007645391,0.0002319149,0.9092196,0.0005795248,0.08723713,0.001040671,0.00002761899],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.324225,0.001955675,0.6645086,0.001797179,0.00008947296,0.0001217236,0.001037984,0.001861909,0.0044025],"genre_scores_gemma":[0.9528204,0.0002983013,0.04485708,0.0001780687,0.00008096759,0.00006772448,0.001006534,0.0001398449,0.0005511562],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01214811,"threshold_uncertainty_score":0.06424612,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4382449327","doi":"10.1162/tacl_a_00556","title":"Aggretriever: A Simple Approach to Aggregate Textual Representations for Robust Dense Passage Retrieval","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund","keywords":"Computer science; Language model; Security token; Aggregate (composite); Exploit; Overhead (engineering); Encoder; Code (set theory); Artificial intelligence; Natural language processing; Set (abstract data type); Machine learning; Programming language","authors":[{"name":"Sheng-Chieh Lin","is_ca":true},{"name":"Minghan Li","is_ca":true},{"name":"Jimmy Lin","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04200530028915431,"gpt":0.2897842792160807,"spread":0.2477789789269264,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001846006,0.001847586,0.001556373,0.002408703,0.0006698607,0.001710632,0.003688904,0.001475402,0.01062227],"category_scores_gemma":[0.006045006,0.0007738376,0.001489315,0.001920859,0.001055231,0.004681627,0.003638043,0.002208554,0.007996951],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009156452,"about_ca_system_score_gemma":0.001689172,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007731715,"about_ca_topic_score_gemma":0.01311705,"domain_scores_codex":[0.998899,0.0002898141,0.0000851709,0.0003781806,0.0002155893,0.0001323288],"domain_scores_gemma":[0.9983304,0.0005434905,0.0001083593,0.0006124005,0.0003072886,0.00009797362],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004732493,0.0004307414,0.001346483,0.0003649699,0.0001548788,0.0003645771,0.0004173906,0.09544717,0.02470934,0.01588338,0.03080574,0.8296021],"study_design_scores_gemma":[0.00006738585,0.0002031749,0.0002706144,0.00002984087,0.00005821669,0.0001783079,0.000115613,0.9597753,0.01249564,0.0169566,0.009801117,0.0000482389],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02090739,0.0006843304,0.9396949,0.0004042812,0.0002389269,0.0003476829,0.001111925,0.03407427,0.002536218],"genre_scores_gemma":[0.2603454,0.0004505809,0.720184,0.0005412574,0.0002004548,0.0005298983,0.005105868,0.002249437,0.01039321],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01062227,"threshold_uncertainty_score":0.03553504,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3213458975","doi":"10.1162/tacl_a_00419","title":"<scp>ParsiNLU</scp>: A Suite of Language Understanding Challenges for Persian","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":18,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Microsoft (Canada); International Medias Data Services (Canada)","funders":"","keywords":"Computer science; Natural language understanding; Natural language processing; Suite; Benchmark (surveying); Artificial intelligence; Persian; Natural language; Linguistics","authors":[{"name":"Daniel Khashabi","is_ca":false},{"name":"Arman Cohan","is_ca":false},{"name":"Siamak Shakeri","is_ca":false},{"name":"Pedram Hosseini","is_ca":false},{"name":"Pouya Pezeshkpour","is_ca":false},{"name":"Malihe Alikhani","is_ca":false},{"name":"Moin Aminnaseri","is_ca":false},{"name":"Marzieh Bitaab","is_ca":false},{"name":"Faeze Brahman","is_ca":false},{"name":"Sarik Ghazarian","is_ca":false},{"name":"Mozhdeh Gheini","is_ca":false},{"name":"Arman Kabiri","is_ca":true},{"name":"Rabeeh Karimi Mahabagdi","is_ca":false},{"name":"Omid Memarrast","is_ca":false},{"name":"Ahmadreza Mosallanezhad","is_ca":false},{"name":"Erfan Noury","is_ca":false},{"name":"Shahab Raji","is_ca":false},{"name":"Mohammad Sadegh Rasooli","is_ca":false},{"name":"Sepideh Sadeghi","is_ca":false},{"name":"Erfan Sadeqi Azer","is_ca":false},{"name":"Niloofar Safi Samghabadi","is_ca":false},{"name":"Mahsa Shafaei","is_ca":false},{"name":"Saber Sheybani","is_ca":false},{"name":"Ali Tazarv","is_ca":false},{"name":"Yadollah Yaghoobzadeh","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05384418972214958,"gpt":0.281614772268062,"spread":0.2277705825459125,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002070852,0.00192997,0.0008625564,0.002915516,0.002256888,0.00218555,0.001647298,0.001675976,0.007875171],"category_scores_gemma":[0.009059928,0.0002945137,0.0006941662,0.002961193,0.001244059,0.003916285,0.002846814,0.001828892,0.005960732],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001636066,"about_ca_system_score_gemma":0.002414869,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02529729,"about_ca_topic_score_gemma":0.03280266,"domain_scores_codex":[0.9975258,0.001041478,0.0001929522,0.0006214657,0.0004382776,0.0001799329],"domain_scores_gemma":[0.9946214,0.002535433,0.0002010056,0.001035302,0.001267279,0.0003396248],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0007308775,0.0004417709,0.01457426,0.001732096,0.0001905504,0.003280389,0.003533613,0.01981629,0.00939104,0.00863805,0.6180795,0.3195915],"study_design_scores_gemma":[0.0004175819,0.000474011,0.05564914,0.0007659082,0.0001663215,0.005980846,0.01798126,0.1691344,0.05786688,0.05207434,0.6391106,0.0003787563],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6244645,0.01014314,0.04877105,0.01692512,0.002180739,0.0005722797,0.1891115,0.04811381,0.0597179],"genre_scores_gemma":[0.6013018,0.00115579,0.04344494,0.001726029,0.0003679951,0.0004199042,0.3403288,0.001735997,0.009518782],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02529729,"threshold_uncertainty_score":0.05030006,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2621376330","doi":"10.1162/tacl_a_00055","title":"Joint Modeling of Topics, Citations, and Topical Authority in Academic Corpora","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Kootenay Association for Science & Technology","funders":"Institute for Information and Communications Technology Promotion; Ministry of Science, ICT and Future Planning","keywords":"Latent Dirichlet allocation; Computer science; Topic model; Citation; Search engine indexing; Information retrieval; Process (computing); Joint (building); Generative grammar; Data science; Artificial intelligence; World Wide Web","authors":[{"name":"Jooyeon Kim","is_ca":true},{"name":"Dongwoo Kim","is_ca":false},{"name":"Alice Oh","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07106014512581252,"gpt":0.318250592846475,"spread":0.2471904477206625,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["bibliometrics"],"consensus_categories":[],"category_scores_codex":[0.01039058,0.001115723,0.001551583,0.008389806,0.001634603,0.004208154,0.00207344,0.002187163,0.002434107],"category_scores_gemma":[0.04555575,0.000909063,0.002014219,0.01019488,0.001608447,0.00830467,0.002259714,0.002778608,0.001764792],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002386807,"about_ca_system_score_gemma":0.002723745,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01647594,"about_ca_topic_score_gemma":0.0224506,"domain_scores_codex":[0.9956542,0.002215741,0.0003409135,0.0009961798,0.0005692671,0.0002236654],"domain_scores_gemma":[0.9745771,0.01974169,0.001525031,0.001826122,0.001884864,0.0004451852],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008235031,0.0007243739,0.0446758,0.000972366,0.0006577469,0.0006739224,0.003662061,0.552632,0.005015152,0.08699914,0.02890758,0.2742563],"study_design_scores_gemma":[0.00005478408,0.00003389429,0.003231131,0.00004069344,0.0000602016,0.00008290669,0.0001318459,0.9552019,0.0006297481,0.03624069,0.004258548,0.00003360718],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3560072,0.007778679,0.6035477,0.003806041,0.0004533119,0.0005694352,0.008518483,0.005805213,0.01351399],"genre_scores_gemma":[0.8158143,0.002586906,0.1558871,0.000314413,0.001016593,0.0009815642,0.01500465,0.0006796211,0.007714904],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9916102,"threshold_uncertainty_score":0.05495131,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3036116903","doi":"10.1162/tacl_a_00316","title":"Learning Lexical Subspaces in a Distributional Vector Space","year":2020,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Distributional semantics; Linear subspace; Artificial intelligence; Natural language processing; Similarity (geometry); Vector space; Word (group theory); Space (punctuation); Relation (database); Semantics (computer science); Semantic similarity; Suite; Code (set theory); Linguistics; Programming language; Data mining; Mathematics","authors":[{"name":"Kushal Arora","is_ca":true},{"name":"Aishik Chakraborty","is_ca":true},{"name":"Jackie Chi Kit Cheung","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01387567187832345,"gpt":0.2681135972948576,"spread":0.2542379254165342,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001742265,0.001163298,0.00137806,0.002530649,0.0006995851,0.002564076,0.001369585,0.001162941,0.002647024],"category_scores_gemma":[0.005556089,0.0005251367,0.001159981,0.002495478,0.001075779,0.006074792,0.00312975,0.002018674,0.001638369],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008138457,"about_ca_system_score_gemma":0.0009682217,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00249283,"about_ca_topic_score_gemma":0.004119848,"domain_scores_codex":[0.9981315,0.0006467518,0.0001342734,0.0006417313,0.0003055115,0.0001403077],"domain_scores_gemma":[0.9980826,0.0008137062,0.000215071,0.0003656897,0.0003592963,0.0001635587],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0005051931,0.0005341108,0.01744426,0.0004795365,0.000311981,0.0003299372,0.001040102,0.1251475,0.009618644,0.1422143,0.01590299,0.6864715],"study_design_scores_gemma":[0.0000243788,0.0001350446,0.001144367,0.00005231711,0.00002130839,0.0001099053,0.0003257134,0.7769454,0.00122993,0.2157408,0.004236056,0.00003492108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05503278,0.0007160115,0.9394506,0.0004169782,0.00004822585,0.00007728327,0.0007656198,0.001693921,0.001798725],"genre_scores_gemma":[0.5963921,0.0008294194,0.3891634,0.0004788981,0.0001665555,0.0003450959,0.007380394,0.0004052473,0.004838897],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002647024,"threshold_uncertainty_score":0.009214044,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4390037801","doi":"10.1162/tacl_a_00627","title":"AfriSpeech-200: Pan-African Accented Speech Dataset for Clinical and General Domain ASR","year":2023,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Benchmark (surveying); Computer science; Speech recognition; Domain (mathematical analysis); Set (abstract data type); Productivity; Natural language processing; Test set; Test (biology); Artificial intelligence; Biology","authors":[{"name":"Tobi Olatunji","is_ca":false},{"name":"Tejumade Afonja","is_ca":false},{"name":"Aditya Yadavalli","is_ca":false},{"name":"Chris Chinenye Emezue","is_ca":true},{"name":"Sahib Singh","is_ca":false},{"name":"Bonaventure F. P. Dossou","is_ca":true},{"name":"Joanne Osuchukwu","is_ca":false},{"name":"Salomey Osei","is_ca":false},{"name":"Atnafu Lambebo Tonja","is_ca":false},{"name":"Naome A. Etori","is_ca":false},{"name":"Clinton Mbataku","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05812305070298272,"gpt":0.3487872494558333,"spread":0.2906641987528506,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001612317,0.00285175,0.001160138,0.002197735,0.001039433,0.00117596,0.001630493,0.002290001,0.01418737],"category_scores_gemma":[0.004128498,0.0003860275,0.001045057,0.001272286,0.0006244834,0.001120482,0.001889749,0.001374727,0.02479672],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007158857,"about_ca_system_score_gemma":0.00152424,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01612459,"about_ca_topic_score_gemma":0.02227729,"domain_scores_codex":[0.9983625,0.0003709759,0.0002008902,0.0003778908,0.0004509769,0.0002367595],"domain_scores_gemma":[0.9982704,0.0004125146,0.0000833612,0.0004571973,0.0005982194,0.0001783551],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002729665,0.0009639089,0.006284671,0.001952268,0.0003156969,0.002089001,0.0004693149,0.007290827,0.03705096,0.0007965968,0.7566945,0.1833625],"study_design_scores_gemma":[0.002106995,0.002700012,0.1275389,0.001002795,0.000568519,0.01101543,0.003192896,0.08682817,0.0729942,0.003732404,0.6874828,0.0008367677],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.1485341,0.003927663,0.01951498,0.001549177,0.00260002,0.001912362,0.7791665,0.02226699,0.02052819],"genre_scores_gemma":[0.05424718,0.0005261062,0.00978362,0.0002851176,0.0002632073,0.0009302304,0.9259215,0.0003962375,0.007646847],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01612459,"threshold_uncertainty_score":0.04746145,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4285138469","doi":"10.1162/tacl_a_00487","title":"Heterogeneous Supervised Topic Models","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Computational and Text Analysis Methods","field":"Social Sciences","cited_by":10,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Office of Naval Research; Simons Foundation; Alfred P. Sloan Foundation; National Science Foundation","keywords":"Computer science; Inference; Artificial intelligence; Outcome (game theory); Machine learning; Topic model; Bayes' theorem; Bayesian inference; Latent variable; Language model; Bayesian probability; Natural language processing; Probabilistic logic; Tone (literature); Linguistics","authors":[{"name":"Dhanya Sridhar","is_ca":true},{"name":"Hal Daumé","is_ca":false},{"name":"David M. Blei","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04311022111236518,"gpt":0.3237508931693172,"spread":0.280640672056952,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005525609,0.001037423,0.001583417,0.001808461,0.0006626355,0.001714584,0.002481788,0.001881503,0.003574236],"category_scores_gemma":[0.0151163,0.0006279444,0.001554629,0.001605756,0.001200673,0.00305963,0.001224058,0.002488877,0.001135734],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001232767,"about_ca_system_score_gemma":0.000775667,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004565104,"about_ca_topic_score_gemma":0.006379957,"domain_scores_codex":[0.9971926,0.001512287,0.0001257464,0.0007497307,0.0002580676,0.0001616425],"domain_scores_gemma":[0.9859291,0.01129106,0.0008556317,0.000882067,0.0008596858,0.0001824915],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0004469627,0.0002376429,0.007444682,0.0003188367,0.0003539177,0.0002290834,0.0005129277,0.8152267,0.001742747,0.05414791,0.008340361,0.1109982],"study_design_scores_gemma":[0.00001033277,0.000009427908,0.0002610422,0.00001041064,0.0000106459,0.00001260869,0.00001041339,0.9825963,0.0001522207,0.01663852,0.0002828065,0.000005456902],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09319498,0.001181091,0.8991059,0.001207904,0.0001569898,0.0001284487,0.001361429,0.001043334,0.002619868],"genre_scores_gemma":[0.9010714,0.0005177825,0.08842243,0.0003064465,0.0005397376,0.0002781544,0.003093706,0.0002259118,0.005544361],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005525609,"threshold_uncertainty_score":0.02922255,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3208625967","doi":"10.1162/tacl_a_00427","title":"Lexically Aware Semi-Supervised Learning for OCR Post-Correction","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Handwritten Text Recognition Techniques","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of British Columbia","funders":"Government of Canada; National Endowment for the Humanities; National Science Foundation","keywords":"Computer science; Decoding methods; Optical character recognition; Consistency (knowledge bases); Artificial intelligence; Natural language processing; Language model; Raw data; Vocabulary; Error detection and correction; Machine learning; Speech recognition; Image (mathematics); Algorithm","authors":[{"name":"Shruti Rijhwani","is_ca":false},{"name":"Daisy Rosenblum","is_ca":true},{"name":"Antonios Anastasopoulos","is_ca":false},{"name":"Graham Neubig","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01389359865853384,"gpt":0.2640636733909112,"spread":0.2501700747323774,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001381945,0.0009173825,0.000897059,0.0007010094,0.0006017464,0.0008160943,0.002388001,0.001092281,0.002492286],"category_scores_gemma":[0.008213647,0.0004547873,0.0006499813,0.0004708325,0.0009261576,0.001442339,0.001165547,0.001737345,0.001604583],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000827758,"about_ca_system_score_gemma":0.001632644,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004704421,"about_ca_topic_score_gemma":0.01034698,"domain_scores_codex":[0.9984806,0.0004780066,0.0001302597,0.0003770059,0.0004385783,0.00009555593],"domain_scores_gemma":[0.9925141,0.003310873,0.0006402131,0.001239196,0.002166445,0.0001291434],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000274528,0.000377386,0.001638641,0.000234335,0.00009145754,0.000132698,0.0002239425,0.2345565,0.05812885,0.002787087,0.002455893,0.6990987],"study_design_scores_gemma":[0.000005090315,0.00003846314,0.000203661,0.000006975134,0.000005896439,0.00002896942,0.00001529363,0.9803652,0.01789188,0.001065004,0.0003643981,0.000009274471],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03763921,0.0001525507,0.9543724,0.000113303,0.00005872252,0.00008650655,0.00008709478,0.006595835,0.0008944007],"genre_scores_gemma":[0.5950214,0.00008055998,0.3996906,0.0001624607,0.00005552913,0.0002435841,0.000510118,0.0005287086,0.00370698],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004704421,"threshold_uncertainty_score":0.009354055,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3127907679","doi":"10.1162/tacl_a_00378","title":"A Computational Framework for Slang Generation","year":2021,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Swearing, Euphemism, Multilingualism","field":"Social Sciences","cited_by":9,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Vector Institute; University of Toronto","funders":"","keywords":"Slang; Natural language; Word (group theory); Interpretation (philosophy); Construct (python library); Inference; Meaning (existential); Probabilistic logic","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.04744032429022425,"gpt":0.3581615381772568,"spread":0.3107212138870326,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002070243,0.0004684526,0.0004971514,0.001613047,0.001076389,0.001891135,0.002394989,0.001038469,0.006742869],"category_scores_gemma":[0.01085028,0.0004570083,0.001245437,0.0009586507,0.00305037,0.003348808,0.00256831,0.001947171,0.0007413518],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001751867,"about_ca_system_score_gemma":0.001459593,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006392587,"about_ca_topic_score_gemma":0.008025387,"domain_scores_codex":[0.9984717,0.0007283211,0.00007311733,0.0003343285,0.0002862149,0.0001063601],"domain_scores_gemma":[0.9957576,0.002985045,0.0002644724,0.0004929708,0.0003681168,0.0001317441],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00005065756,0.0001000334,0.001669178,0.0001018592,0.00003926679,0.0002237394,0.0004962777,0.3662158,0.001624614,0.5673882,0.002635028,0.05945547],"study_design_scores_gemma":[0.000005502819,0.000007631103,0.0001088031,0.000009450487,0.000002641069,0.00002283791,0.00003213946,0.8057445,0.0002566192,0.1927167,0.001085663,0.000007531029],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01093439,0.00005998857,0.9850853,0.000435527,0.0000277188,0.00004805227,0.0001395601,0.0004037917,0.002865748],"genre_scores_gemma":[0.5875412,0.00007923697,0.408675,0.0001915169,0.00006565655,0.0001961373,0.0005162198,0.0001871898,0.002547874],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006742869,"threshold_uncertainty_score":0.02255714,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4221053465","doi":"10.1162/tacl_a_00458","title":"Neuro-symbolic Natural Logic with Introspective Revision for Natural Language Inference","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Queen's University","funders":"","keywords":"Interpretability; Computer science; Artificial intelligence; Inference; Generalization; Introspection; Spurious relationship; Machine learning; Rule of inference; Natural language; Natural (archaeology); Overfitting; Artificial neural network; Cognitive psychology","authors":[{"name":"Yufei Feng","is_ca":true},{"name":"Xiaoyu Yang","is_ca":true},{"name":"Xiaodan Zhu","is_ca":true},{"name":"Michael Greenspan","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01236504266629197,"gpt":0.2721111208052126,"spread":0.2597460781389206,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001843182,0.0005528256,0.000633723,0.0008508851,0.0004576654,0.001403981,0.001792758,0.0006745504,0.003133187],"category_scores_gemma":[0.006642176,0.0003335029,0.001132038,0.0005914749,0.001791209,0.002718461,0.001574409,0.002096638,0.0004030195],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001661842,"about_ca_system_score_gemma":0.002015925,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005793418,"about_ca_topic_score_gemma":0.008246122,"domain_scores_codex":[0.9987454,0.0005189743,0.00007245454,0.0002496143,0.000320226,0.00009310909],"domain_scores_gemma":[0.9969094,0.001712593,0.0002893679,0.0006043874,0.0003711364,0.0001130994],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002057715,0.0001885293,0.001793911,0.0002010745,0.0001403731,0.0003122473,0.0003263401,0.6272089,0.004886189,0.2020287,0.002742216,0.1599657],"study_design_scores_gemma":[0.00001037496,0.00001530027,0.00009419651,0.000008019105,0.000009638651,0.0000228229,0.000007046537,0.9390288,0.0006276523,0.05962204,0.0005472131,0.000006912875],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02058485,0.0002788542,0.9748824,0.0004582441,0.00003620764,0.0000611287,0.0001401086,0.001287956,0.0022702],"genre_scores_gemma":[0.7641208,0.0002218944,0.2333304,0.0002086576,0.00005914608,0.000157332,0.0002678611,0.0001130113,0.001520731],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005793418,"threshold_uncertainty_score":0.01205754,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4394773742","doi":"10.1162/tacl_a_00645","title":"To Diverge or Not to Diverge: A Morphosyntactic Perspective on Machine Translation vs Human Translation","year":2024,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Google (Canada)","funders":"","keywords":"Divergence (linguistics); Machine translation; Perspective (graphical); Computer science; Translation (biology); Artificial intelligence; Natural language processing; Diversity (politics); Linguistics; Sociology; Biology","authors":[{"name":"Jiaming Luo","is_ca":true},{"name":"Colin Cherry","is_ca":true},{"name":"George Foster","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0292461547114884,"gpt":0.3409946650393249,"spread":0.3117485103278365,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00486376,0.0003684484,0.0006066866,0.002959391,0.0008985989,0.003278647,0.0004603377,0.0007479464,0.002682977],"category_scores_gemma":[0.01867468,0.0001833446,0.0002930729,0.002809058,0.003564146,0.003750035,0.001592248,0.001740747,0.0004213634],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009523283,"about_ca_system_score_gemma":0.0005157721,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001183395,"about_ca_topic_score_gemma":0.002042489,"domain_scores_codex":[0.996273,0.002330605,0.000156055,0.0004869862,0.0006082986,0.0001449911],"domain_scores_gemma":[0.9837124,0.01282236,0.001146975,0.0009854997,0.0009758917,0.0003568158],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.002960713,0.0003920102,0.1973631,0.001280062,0.0009728507,0.002259983,0.02738713,0.01921145,0.1590488,0.2539157,0.006001375,0.3292069],"study_design_scores_gemma":[0.0001295136,0.0009818107,0.4808874,0.0003692632,0.0003854569,0.00267598,0.01850969,0.06027025,0.0351783,0.3768546,0.02352371,0.0002340234],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9199578,0.004843834,0.0498286,0.003851873,0.00008855452,0.00003456031,0.0005873243,0.0002292325,0.02057835],"genre_scores_gemma":[0.9933867,0.0003315083,0.005630956,0.0001780343,0.0000367843,0.0000108734,0.0001585995,0.0000646499,0.0002018292],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00486376,"threshold_uncertainty_score":0.02572232,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2115008472","doi":"10.1162/tacl_a_00233","title":"Distributional Semantics Beyond Words: Supervised Learning of Analogy and Paraphrase","year":2013,"lang":"en","type":"preprint","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"National Research Council Canada","funders":"","keywords":"Paraphrase; Natural language processing; Similarity (geometry); Artificial intelligence; Computer science; Analogy; Distributional semantics; Word (group theory); Tuple; Pairwise comparison; Noun; Function (biology); Noun phrase; Semantic similarity; Linguistics; Mathematics","authors":[{"name":"Peter D. Turney","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01650727852168584,"gpt":0.2569243560769647,"spread":0.2404170775552789,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003284114,0.0009171499,0.001438515,0.00318852,0.0008811745,0.001762636,0.002155639,0.001708845,0.00166352],"category_scores_gemma":[0.01882578,0.0003911899,0.00114053,0.002863998,0.001494516,0.006444142,0.002777406,0.002502385,0.0006786152],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007890144,"about_ca_system_score_gemma":0.0007073039,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00108449,"about_ca_topic_score_gemma":0.001584563,"domain_scores_codex":[0.9959468,0.002221786,0.0002515734,0.0009380958,0.0005056971,0.0001361034],"domain_scores_gemma":[0.98967,0.006453741,0.0009402772,0.001705217,0.0008697627,0.0003610117],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008186728,0.001408063,0.02906203,0.0006459027,0.0005556328,0.0003996466,0.001826454,0.09484866,0.005910423,0.05233319,0.01242904,0.7997622],"study_design_scores_gemma":[0.00007330934,0.00021574,0.004882552,0.00005279358,0.0000552694,0.0001935253,0.0003051455,0.7836914,0.001694779,0.2065454,0.002249451,0.00004063411],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2442998,0.002368551,0.7426266,0.001175991,0.0001229232,0.0002315655,0.0007629723,0.001634515,0.006776982],"genre_scores_gemma":[0.9132507,0.0003367523,0.08240594,0.0002727341,0.0001812423,0.0002035874,0.001935793,0.0001031072,0.00131002],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003284114,"threshold_uncertainty_score":0.01736826,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4400678062","doi":"10.1162/tacl_a_00678","title":"Can Authorship Attribution Models Distinguish Speakers in Speech Transcripts?","year":2024,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Computer science; Conversation; Punctuation; Natural language processing; Benchmark (surveying); Attribution; Spurious relationship; Task (project management); Speech recognition; Artificial intelligence; Transcription (linguistics); Suite; Construct (python library); Linguistics; Psychology; Machine learning; Communication","authors":[{"name":"Cristina Aggazzotti","is_ca":false},{"name":"Nicholas Andrews","is_ca":false},{"name":"Elizabeth A. T. Smith","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04126170943162795,"gpt":0.2890382353474506,"spread":0.2477765259158227,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006152193,0.0009932863,0.0007322036,0.001132002,0.0006508416,0.002166151,0.001194782,0.001494875,0.002472329],"category_scores_gemma":[0.03260499,0.0003020056,0.0005922641,0.0007048139,0.0006858785,0.00263888,0.001331428,0.002200357,0.00221538],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006568084,"about_ca_system_score_gemma":0.0007589965,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002441399,"about_ca_topic_score_gemma":0.002510987,"domain_scores_codex":[0.996847,0.001860488,0.0001394487,0.0007048766,0.0002360795,0.000212151],"domain_scores_gemma":[0.9815583,0.01297966,0.001382747,0.002080228,0.001488712,0.00051039],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002778819,0.001015205,0.06006737,0.0005329744,0.0004197591,0.0004158845,0.002083062,0.1514935,0.02882894,0.002285995,0.008552332,0.7415262],"study_design_scores_gemma":[0.0000404229,0.0001781193,0.009033632,0.00006784711,0.00006738426,0.0001092492,0.000562207,0.9691664,0.01312855,0.006433697,0.001168805,0.00004370304],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8448912,0.001660529,0.139593,0.001731259,0.0006534742,0.0001889277,0.001261804,0.003272704,0.006747118],"genre_scores_gemma":[0.9809574,0.0001452163,0.01593953,0.00008231399,0.00009870767,0.00004732312,0.001023944,0.00006876824,0.00163692],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006152193,"threshold_uncertainty_score":0.03253633,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2774582842","doi":"10.1162/tacl_a_00076","title":"Joint Prediction of Word Alignment with Alignment Types","year":2017,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Simon Fraser University","funders":"","keywords":"Computer science; Word (group theory); Artificial intelligence; Task (project management); Probabilistic logic; Natural language processing; Joint (building); Generative grammar; Pattern recognition (psychology)","authors":[{"name":"Anahita Mansouri Bigvand","is_ca":true},{"name":"Te Bu","is_ca":true},{"name":"Anoop Sarkar","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01894919867233834,"gpt":0.2665170272751402,"spread":0.2475678286028018,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003824954,0.001203847,0.001556491,0.002589863,0.0007296239,0.001767745,0.00169081,0.001941835,0.002451143],"category_scores_gemma":[0.01983449,0.000893784,0.001553261,0.002398066,0.001078021,0.006804264,0.002117852,0.003623861,0.003123739],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007521151,"about_ca_system_score_gemma":0.001371414,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002035827,"about_ca_topic_score_gemma":0.003891905,"domain_scores_codex":[0.9956328,0.001608225,0.0002634276,0.001644999,0.0006061528,0.0002443634],"domain_scores_gemma":[0.9849518,0.009473177,0.001760977,0.001729292,0.001638588,0.0004461689],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.002125115,0.0008137728,0.06747568,0.00056098,0.0005576672,0.0004373915,0.0006687481,0.2585334,0.03039533,0.01900526,0.01279981,0.6066269],"study_design_scores_gemma":[0.0000236341,0.00006127921,0.003342645,0.00001612283,0.0000353315,0.000082364,0.00005275379,0.96353,0.003791545,0.02821947,0.0008162929,0.00002843694],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1645198,0.000686596,0.8256328,0.0006548646,0.00021491,0.0001396633,0.001214854,0.004391225,0.002545387],"genre_scores_gemma":[0.824299,0.0002244329,0.1691843,0.0002194831,0.0001571711,0.0001671017,0.002799293,0.0005205815,0.002428737],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003824954,"threshold_uncertainty_score":0.02022851,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W7086984966","doi":"10.1162/tacl.a.38","title":"Elements of World Knowledge (<scp>EWoK</scp>): A Cognition-Inspired Framework for Evaluating Basic World Knowledge in Language Models","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Schizophrenia research and treatment","field":"Medicine","cited_by":3,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Vector Institute","funders":"Quest for Intelligence, Massachusetts Institute of Technology; Yuhan","keywords":"Situated; Language model; Conceptual model; Domain knowledge; Simple (philosophy); Conceptual framework; Knowledge-based systems","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.04625059357637484,"gpt":0.3867130782903953,"spread":0.3404624847140204,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01025439,0.00185831,0.0009416936,0.008193426,0.001080158,0.005221235,0.002839415,0.002955503,0.005328209],"category_scores_gemma":[0.05874512,0.0005915812,0.002021406,0.004879477,0.001860841,0.006650415,0.005277615,0.003134143,0.001584124],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002005639,"about_ca_system_score_gemma":0.001672002,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01032883,"about_ca_topic_score_gemma":0.01636918,"domain_scores_codex":[0.9910358,0.004647004,0.001010058,0.001404489,0.001651868,0.0002507439],"domain_scores_gemma":[0.9636072,0.02502816,0.002160604,0.00600261,0.002425677,0.0007757291],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001653832,0.001669878,0.06574433,0.003428713,0.001601136,0.0007277629,0.001853402,0.2641517,0.005503545,0.06933048,0.1005255,0.4838098],"study_design_scores_gemma":[0.0001458486,0.0003421513,0.01136979,0.0003997107,0.000137939,0.0002646047,0.0005839518,0.8748009,0.004959216,0.0806859,0.02615811,0.0001518009],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1754205,0.004024954,0.6953224,0.003018302,0.0005002106,0.001885175,0.08051805,0.01807152,0.02123893],"genre_scores_gemma":[0.5527543,0.0006855269,0.3659405,0.0005863734,0.0001261809,0.001309829,0.07644338,0.0007126877,0.001441185],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01032883,"threshold_uncertainty_score":0.05423111,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4417275930","doi":"10.1162/tacl.a.54","title":"The Impact of Automatic Speech Transcription on Speaker Attribution","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université TÉLUQ; Université du Québec à Montréal","funders":"","keywords":"Attribution; Transcription (linguistics); Task (project management); Speaker diarisation; Phonetic transcription; Speech processing","authors":[{"name":"Cristina Aggazzotti","is_ca":false},{"name":"Matthew Wiesner","is_ca":false},{"name":"Elizabeth A. T. Smith","is_ca":true},{"name":"Nicholas Andrews","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01855258785467077,"gpt":0.296895449186917,"spread":0.2783428613322462,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0181842,0.001088014,0.0008494884,0.001029287,0.001006703,0.002615976,0.001023237,0.001236728,0.002140767],"category_scores_gemma":[0.2278911,0.0004284653,0.0003897981,0.001119119,0.001408916,0.002495368,0.002209859,0.001575474,0.002050024],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009321132,"about_ca_system_score_gemma":0.0009023989,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002745805,"about_ca_topic_score_gemma":0.001612927,"domain_scores_codex":[0.9617702,0.02243109,0.002972779,0.004602919,0.007254244,0.0009686448],"domain_scores_gemma":[0.6311768,0.2919121,0.02159804,0.03164111,0.02134322,0.00232874],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.008938152,0.0008559735,0.1188267,0.001486371,0.0007077683,0.001296658,0.006582418,0.1274363,0.1767625,0.001933371,0.007291877,0.5478819],"study_design_scores_gemma":[0.0002151317,0.004442165,0.2277292,0.0006374141,0.0006670598,0.002714825,0.004561025,0.3941679,0.3484959,0.009045025,0.006651521,0.0006729754],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.922733,0.001250912,0.06596328,0.0007841224,0.000633007,0.0001494675,0.0006722307,0.003159028,0.004655025],"genre_scores_gemma":[0.9883636,0.0001773955,0.009109929,0.0001530546,0.000104016,0.00003343148,0.0007677997,0.0004441745,0.0008467015],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.0181842,"threshold_uncertainty_score":0.0961684,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4399426318","doi":"10.1162/tacl_a_00669","title":"Source-Free Domain Adaptation for Question Answering with Masked Self-training","year":2024,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Western University","funders":"","keywords":"Computer science; Domain adaptation; Adaptation (eye); Domain (mathematical analysis); Question answering; Training (meteorology); Training set; Natural language processing; Artificial intelligence; Information retrieval; Speech recognition; Psychology","authors":[{"name":"M. Yin","is_ca":true},{"name":"Boyu Wang","is_ca":true},{"name":"Yue Dong","is_ca":false},{"name":"Charles X. Ling","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01946249239711238,"gpt":0.2532761076822381,"spread":0.2338136152851257,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002699584,0.00105239,0.0008186845,0.0007698624,0.0004545663,0.0007461662,0.001947975,0.001391542,0.00185134],"category_scores_gemma":[0.006169381,0.0005001897,0.0009997393,0.0006632974,0.0008909043,0.002378234,0.002549531,0.002318628,0.001271883],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006379754,"about_ca_system_score_gemma":0.0007149871,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002303574,"about_ca_topic_score_gemma":0.002567192,"domain_scores_codex":[0.9987387,0.0005694845,0.00005777819,0.0004301883,0.0001288874,0.00007501158],"domain_scores_gemma":[0.9966581,0.001716739,0.0001578899,0.000950499,0.000397498,0.0001191824],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007723626,0.0007603513,0.006416548,0.0003856136,0.0003211002,0.000262804,0.0009438813,0.2688917,0.05941187,0.007608148,0.01329819,0.6409274],"study_design_scores_gemma":[0.00001800668,0.00007086327,0.0006737737,0.00001336011,0.00002234712,0.00006229043,0.00004506152,0.9837277,0.008378778,0.005113182,0.001859654,0.00001496666],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06854468,0.0009427209,0.9213489,0.0003480919,0.0001267196,0.0001478116,0.0002962575,0.006584733,0.001660161],"genre_scores_gemma":[0.7512888,0.0002728077,0.2420896,0.0007692216,0.0001195058,0.000326255,0.001659821,0.0004271389,0.003046776],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002699584,"threshold_uncertainty_score":0.01427692,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3206313044","doi":"10.1162/tacl_a_00441","title":"Quantifying Cognitive Factors in Lexical Decline","year":2021,"lang":"en","type":"preprint","venue":"Transactions of the Association for Computational Linguistics","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"Natural Sciences and Engineering Research Council of Canada; University of Toronto","keywords":"German; Lexical diversity; Variety (cybernetics); Affect (linguistics); Cognition; Linguistics; Set (abstract data type); Logistic regression; Ecological niche; Psychology; Cognitive psychology; Computer science; Artificial intelligence; Ecology; Biology; Vocabulary; Communication","authors":[{"name":"David M. Francis","is_ca":true},{"name":"Ella Rabinovich","is_ca":true},{"name":"Farhan Samir","is_ca":true},{"name":"David R. Mortensen","is_ca":false},{"name":"Suzanne Stevenson","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06521040049384863,"gpt":0.3677425165379125,"spread":0.3025321160440638,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002770694,0.0004264251,0.000435402,0.003307078,0.0004895668,0.002139625,0.0005237554,0.000746868,0.002592406],"category_scores_gemma":[0.03317388,0.0002198582,0.0003709355,0.002344891,0.001533237,0.002200787,0.001167028,0.0006514678,0.0003476976],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007309626,"about_ca_system_score_gemma":0.0003340903,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005009676,"about_ca_topic_score_gemma":0.00408427,"domain_scores_codex":[0.9988953,0.0003130881,0.000109829,0.0003227507,0.0002368653,0.0001221781],"domain_scores_gemma":[0.9766021,0.01472096,0.004494153,0.001215413,0.002161161,0.0008062581],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.0002523672,0.0001133748,0.9749157,0.00006987963,0.00008791062,0.0001541559,0.001570836,0.003043929,0.003155445,0.001222322,0.0001328763,0.0152812],"study_design_scores_gemma":[0.000005289704,0.0000812061,0.9886615,0.00001396463,0.00002526628,0.00013151,0.001162256,0.005076732,0.0007808185,0.003664874,0.0003744626,0.00002211904],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9960663,0.00010438,0.001686658,0.00003921348,0.000002432315,0.00001419774,0.0001944661,0.00001637184,0.001876049],"genre_scores_gemma":[0.9993461,0.00001742662,0.0003499872,0.000007779392,0.000003459678,0.000008502585,0.0001703422,0.000005512755,0.00009087753],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005009676,"threshold_uncertainty_score":0.01465297,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4411934141","doi":"10.1162/tacl_a_00759","title":"A Comparative Approach for Auditing Multilingual Phonetic Transcript Archives","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Speech Recognition and Synthesis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Vector Institute; University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Audit; Natural language processing; Linguistics; Speech recognition; Data science; Artificial intelligence; World Wide Web; Accounting; Business","authors":[{"name":"Farhan Samir","is_ca":true},{"name":"Emily P. Ahn","is_ca":false},{"name":"Shreya Prakash","is_ca":false},{"name":"Márton Sóskuthy","is_ca":true},{"name":"Vered Shwartz","is_ca":true},{"name":"Jian Zhu","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03049783272393018,"gpt":0.2948096777967044,"spread":0.2643118450727742,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008241626,0.0008935209,0.0007599272,0.007489355,0.002016329,0.003396112,0.00252374,0.001333731,0.007476762],"category_scores_gemma":[0.0327965,0.0006214657,0.000655517,0.00462818,0.001108615,0.003320325,0.004956577,0.001255126,0.003299051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001287349,"about_ca_system_score_gemma":0.003037369,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005928961,"about_ca_topic_score_gemma":0.01548088,"domain_scores_codex":[0.9881666,0.004668763,0.0009603114,0.00195156,0.003811687,0.0004409437],"domain_scores_gemma":[0.96839,0.007780415,0.001823026,0.007962522,0.01306484,0.0009791758],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001232138,0.000418661,0.03264702,0.0009832545,0.0002872203,0.001355582,0.007050565,0.009237997,0.1090874,0.009038255,0.01117677,0.817485],"study_design_scores_gemma":[0.0003046301,0.0016339,0.1267012,0.0009056668,0.0007534023,0.005274907,0.01965096,0.368171,0.2681974,0.03018671,0.17757,0.0006501582],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.1700148,0.001585816,0.7810168,0.001330037,0.000449969,0.001110275,0.005237091,0.01466807,0.02458709],"genre_scores_gemma":[0.3551533,0.0003789694,0.6346353,0.0002273929,0.0001461762,0.0005043009,0.003885173,0.0009503952,0.004119077],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008241626,"threshold_uncertainty_score":0.04358637,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W7114916806","doi":"10.1162/tacl.a.62","title":"Towards More Realistic Extraction Attacks: An Adversarial Perspective","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Adversarial system; Adversary; Perspective (graphical); Training set; Memorization; Attack model","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.01430245061515127,"gpt":0.3394118539592187,"spread":0.3251094033440675,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006008617,0.0009761075,0.0008322574,0.0008660495,0.0009449301,0.002254807,0.001404915,0.002285636,0.003243368],"category_scores_gemma":[0.03709186,0.0005844443,0.0009360754,0.0005581822,0.003194119,0.005162231,0.004250068,0.004898091,0.0008256504],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001150795,"about_ca_system_score_gemma":0.0008848191,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007787325,"about_ca_topic_score_gemma":0.0006660346,"domain_scores_codex":[0.9940867,0.003210005,0.0002383044,0.000770865,0.001232906,0.0004611528],"domain_scores_gemma":[0.9582527,0.03105998,0.0017232,0.007544335,0.001053268,0.0003666408],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003671406,0.0002042413,0.004027377,0.0001784363,0.0001313189,0.0006959915,0.000453261,0.7845139,0.01245065,0.1559333,0.005948925,0.0350955],"study_design_scores_gemma":[0.00001936473,0.00006508721,0.0002884915,0.00003936003,0.00001490737,0.0002390634,0.00006456278,0.9081494,0.005793771,0.08284262,0.002458101,0.00002514296],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1009529,0.0004885882,0.8818971,0.004155521,0.0001829155,0.0002420525,0.0004384346,0.001341145,0.01030138],"genre_scores_gemma":[0.9277842,0.0002472544,0.06723172,0.0008242801,0.00009896348,0.0001281966,0.0002241364,0.0001877965,0.003273391],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006008617,"threshold_uncertainty_score":0.03177696,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4416372001","doi":"10.1162/tacl.a.50","title":"Objectifying the Subjective: Cognitive Biases in Topic Interpretations","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Victoria","funders":"","keywords":"Representativeness heuristic; Interpretation (philosophy); Salient; Cognitive bias; Quality (philosophy); Heuristics; Context (archaeology); Coherence (philosophical gambling strategy); Cognition","authors":[{"name":"Swapnil Hingmire","is_ca":false},{"name":"Ze Shi Li","is_ca":false},{"name":"Shiyu Zeng","is_ca":true},{"name":"Ahmed Musa Awon","is_ca":true},{"name":"Neil Ernst","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02274245253839106,"gpt":0.3057547969801711,"spread":0.28301234444178,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.08430993,0.0009341402,0.0007544581,0.003676996,0.00237473,0.01286434,0.001587045,0.001810002,0.002526541],"category_scores_gemma":[0.3301892,0.0008402878,0.0009277177,0.002216997,0.01297129,0.01352137,0.005814709,0.003369295,0.0004530333],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002794985,"about_ca_system_score_gemma":0.001844392,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002110547,"about_ca_topic_score_gemma":0.001490829,"domain_scores_codex":[0.8841001,0.09107905,0.004213532,0.005254555,0.01417117,0.001181616],"domain_scores_gemma":[0.6028208,0.323261,0.02563188,0.02662153,0.01890775,0.002757146],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0009672212,0.0002347413,0.07207017,0.001727758,0.0004894328,0.0005118925,0.3134026,0.007408341,0.01479159,0.3351765,0.006761776,0.246458],"study_design_scores_gemma":[0.0001790398,0.0004729037,0.04222086,0.001567089,0.0003717869,0.0008689147,0.07332376,0.06034706,0.01397128,0.7644504,0.04163784,0.0005890975],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4345993,0.002171145,0.5183964,0.007792767,0.0002877057,0.0005416407,0.0001967992,0.0008541177,0.03516022],"genre_scores_gemma":[0.9612249,0.0002642223,0.0368055,0.000507564,0.0000874319,0.0002031612,0.00006774176,0.0001784945,0.0006609756],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.08430993,"threshold_uncertainty_score":0.4458789,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4313459314","doi":"10.1162/tacl_a_00524","title":"The Emergence of Argument Structure in Artificial Languages","year":2022,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Computer science; Word order; Transitive relation; Predicate (mathematical logic); Natural language processing; Contrast (vision); Artificial intelligence; Argument (complex analysis); Sentence; Natural language; Relation (database); Object (grammar); Verb; Mathematics; Programming language","authors":[{"name":"Tom Bosc","is_ca":true},{"name":"Pascal Vincent","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0125030818935014,"gpt":0.2988419155317233,"spread":0.2863388336382219,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002984737,0.0002593906,0.0005622936,0.001344941,0.0008277565,0.002695756,0.0006119349,0.0009699428,0.002063726],"category_scores_gemma":[0.02654661,0.0005428841,0.0005022924,0.0009241175,0.004185729,0.004273042,0.002344498,0.001451628,0.0001822482],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001274572,"about_ca_system_score_gemma":0.0004747305,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005208153,"about_ca_topic_score_gemma":0.0004719925,"domain_scores_codex":[0.9964063,0.002167649,0.0001970197,0.0004400001,0.0006176312,0.0001713691],"domain_scores_gemma":[0.9662848,0.02448543,0.003901284,0.002612093,0.001794997,0.000921459],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001024553,0.0004547041,0.06496318,0.0006978383,0.0002465708,0.002427265,0.01321663,0.08834636,0.1220248,0.635683,0.002039217,0.06887592],"study_design_scores_gemma":[0.0001241998,0.0004433143,0.03930751,0.0000859986,0.00006962167,0.0010682,0.002406183,0.3526511,0.02540421,0.5718907,0.006442546,0.0001063887],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8894517,0.0001551011,0.1032481,0.00075732,0.00001941103,0.00004348811,0.0001523338,0.0004257832,0.005746733],"genre_scores_gemma":[0.9854184,0.00002603599,0.0140605,0.00003203074,0.000009695964,0.00003047494,0.00008185134,0.00005373038,0.0002871615],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002984737,"threshold_uncertainty_score":0.01578498,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W7128225249","doi":"10.1162/tacl.a.27","title":"Investigating Adversarial Trigger Transfer in Large Language Models","year":2025,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Canadian Institute for Advanced Research","funders":"","keywords":"Adversarial system; Security token; Offensive; Robustness (evolution); Language model; Threat model; Language understanding","authors":[{"name":"Nicholas Meade","is_ca":false},{"name":"Arkil Patel","is_ca":false},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01217664085012492,"gpt":0.2804505190115679,"spread":0.2682738781614429,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006812424,0.0009049687,0.0008230606,0.000564665,0.0006063094,0.001132913,0.001453933,0.001389797,0.002972712],"category_scores_gemma":[0.03289236,0.0004792129,0.000870406,0.0003432957,0.002058381,0.002598836,0.002271235,0.002882926,0.000415513],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001504141,"about_ca_system_score_gemma":0.001141129,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00229536,"about_ca_topic_score_gemma":0.002084655,"domain_scores_codex":[0.9955223,0.00282904,0.0001380014,0.0005897134,0.0005597537,0.0003611546],"domain_scores_gemma":[0.9692957,0.02544806,0.001536172,0.002569046,0.0006957865,0.0004551501],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001705114,0.00008950493,0.00194149,0.00006111289,0.00005557678,0.0001165916,0.0001637801,0.9713169,0.002078064,0.01299321,0.0006445194,0.01036881],"study_design_scores_gemma":[0.00001025248,0.00005567039,0.0001595768,0.000005990378,0.000006206842,0.00001193281,0.00002184283,0.9839468,0.0009916908,0.01463066,0.0001516493,0.000007609715],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5215399,0.0003827667,0.4698577,0.001233797,0.00008825446,0.0001730071,0.0003251647,0.002183037,0.004216474],"genre_scores_gemma":[0.9794787,0.00005598449,0.01915249,0.0001463198,0.00001855554,0.00008373985,0.000183802,0.0001486424,0.000731702],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006812424,"threshold_uncertainty_score":0.03602797,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}