{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":19,"total_is_capped":false,"direct_labels_cover":0,"predictions_cover":19,"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":"b4e2a36aa5f9","filters":{"venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing"}},"results":[{"id":"W3152698349","doi":"10.18653/v1/2021.emnlp-main.230","title":"Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":182,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Word (group theory); Natural language processing; Artificial intelligence; Word order; Language model; Downstream (manufacturing); Order (exchange); Parametric statistics; Linguistics; Mathematics","authors":[{"name":"Koustuv Sinha","is_ca":true},{"name":"Robin Jia","is_ca":false},{"name":"Dieuwke Hupkes","is_ca":false},{"name":"Joëlle Pineau","is_ca":true},{"name":"Adina Williams","is_ca":false},{"name":"Douwe Kiela","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07636691541448742,"gpt":0.3862203416242639,"spread":0.3098534262097765,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003336436,0.001391239,0.0007572662,0.0004789316,0.000642718,0.001346939,0.001325684,0.00138786,0.00430266],"category_scores_gemma":[0.0177008,0.0006649107,0.0008272231,0.0005605142,0.0009945612,0.004688173,0.001508514,0.004304003,0.002496617],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000626173,"about_ca_system_score_gemma":0.00161977,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00397339,"about_ca_topic_score_gemma":0.008194104,"domain_scores_codex":[0.9984982,0.0006824813,0.00008020862,0.0005143941,0.0001157319,0.0001090177],"domain_scores_gemma":[0.9919951,0.005831776,0.0001936676,0.001395887,0.0003859006,0.0001977136],"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":[0.001910816,0.000808427,0.01988021,0.0006366516,0.0003517652,0.0007194346,0.001343457,0.1790687,0.09119251,0.0205491,0.02816353,0.6553755],"study_design_scores_gemma":[0.00007076973,0.0003268118,0.003286589,0.00006710485,0.00006628449,0.0002239117,0.0002630102,0.9231259,0.03162676,0.03601005,0.004878645,0.00005419835],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3295034,0.001240609,0.6500712,0.002922499,0.0005073823,0.0001780202,0.001062519,0.008094657,0.006419796],"genre_scores_gemma":[0.8300682,0.0002766097,0.1596263,0.001347635,0.0001163681,0.0002293684,0.003277213,0.0007923447,0.004265942],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00430266,"threshold_uncertainty_score":0.01764494,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3153046263","doi":"10.18653/v1/2021.emnlp-main.168","title":"Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":75,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Canadian Institute for Advanced Research; University of Alberta","funders":"Alberta Machine Intelligence Institute; Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Path (computing); Security token; Artificial neural network; Focus (optics); Artificial intelligence; Suite; Graph; Deep neural networks; Machine learning; Theoretical computer science; History","authors":[{"name":"Nouha Dziri","is_ca":true},{"name":"Andrea Madotto","is_ca":true},{"name":"Osmar R. Zai͏̈ane","is_ca":true},{"name":"Avishek Joey Bose","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0556578091105373,"gpt":0.3775607902916954,"spread":0.3219029811811581,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003050691,0.001702592,0.0007517153,0.00141425,0.0007260411,0.001289517,0.002568517,0.001844539,0.003045583],"category_scores_gemma":[0.01408667,0.0004242909,0.0008198223,0.0007865705,0.001359373,0.004369046,0.004166329,0.002297516,0.001258944],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007952512,"about_ca_system_score_gemma":0.001119439,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00399889,"about_ca_topic_score_gemma":0.008544529,"domain_scores_codex":[0.9981126,0.0007978775,0.00009269069,0.0005743362,0.000297312,0.0001251837],"domain_scores_gemma":[0.9921811,0.005217609,0.0003784903,0.001447833,0.0005767373,0.000198346],"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":[0.002052036,0.0007388141,0.01466902,0.001742594,0.0003411685,0.0009658912,0.00316566,0.1310563,0.02720343,0.01077154,0.04462298,0.7626706],"study_design_scores_gemma":[0.0002187347,0.0005134633,0.002225341,0.00007450075,0.0001195321,0.0003752639,0.0009493792,0.9294496,0.01907319,0.03334948,0.01358694,0.00006450953],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.2943598,0.003279842,0.6520181,0.001757376,0.0003331092,0.0006339395,0.005272846,0.03603615,0.00630895],"genre_scores_gemma":[0.6997132,0.0004140603,0.2824313,0.0005114538,0.00009473303,0.0003399759,0.01135629,0.0006843264,0.004454535],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00399889,"threshold_uncertainty_score":0.01613379,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3214455632","doi":"10.18653/v1/2021.emnlp-main.77","title":"Contextualized Query Embeddings for Conversational Search","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund","keywords":"Computer science; Leverage (statistics); Inference; Security token; Relevance (law); Pipeline (software); Information retrieval; Query expansion; Query language; Artificial intelligence","authors":[{"name":"Sheng-Chieh Lin","is_ca":false},{"name":"Jheng-Hong Yang","is_ca":false},{"name":"Jimmy Lin","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.092333595524727,"gpt":0.4395395092037929,"spread":0.3472059136790658,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009613278,0.0008798506,0.0007908522,0.0009664522,0.0004169227,0.00109613,0.001613103,0.001319279,0.004236063],"category_scores_gemma":[0.006962941,0.0004818079,0.0008611562,0.001128411,0.0006807409,0.004279525,0.001462866,0.001848169,0.001862314],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001176532,"about_ca_system_score_gemma":0.001007725,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008016057,"about_ca_topic_score_gemma":0.01051199,"domain_scores_codex":[0.9992386,0.0002661147,0.00005120913,0.0002358174,0.0001262876,0.00008193574],"domain_scores_gemma":[0.9986927,0.0006620416,0.0001084261,0.0002835164,0.0001979134,0.00005539605],"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":[0.0006578758,0.0004651977,0.003538317,0.0005986327,0.0001858383,0.0003147981,0.00107127,0.438065,0.01801806,0.1217575,0.02413343,0.391194],"study_design_scores_gemma":[0.00001444508,0.00005380439,0.000244796,0.00001363975,0.00002099567,0.00006621602,0.00004721359,0.9640874,0.001342468,0.0311892,0.002904789,0.00001513669],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02703959,0.001297144,0.9645652,0.0005952728,0.00007895151,0.0001178203,0.001227289,0.002586653,0.002492253],"genre_scores_gemma":[0.7218541,0.0009861687,0.2639742,0.0004650004,0.0001956876,0.0004344167,0.004041268,0.0004286384,0.007620572],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008016057,"threshold_uncertainty_score":0.01593876,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3203338521","doi":"10.18653/v1/2021.emnlp-main.328","title":"Language-Aligned Waypoint (LAW) Supervision for Vision-and-Language Navigation in Continuous Environments","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Multimodal Machine Learning Applications","field":"Computer Science","cited_by":39,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Simon Fraser University","funders":"Western Canada Research Grid; Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Waypoint; Computer science; Task (project management); Human–computer interaction; Embodied cognition; Path (computing); Natural language; Work (physics); Measure (data warehouse); Metric (unit); Shortest path problem; Artificial intelligence; Multimedia; Programming language; Real-time computing; Engineering; Theoretical computer science; Database","authors":[{"name":"Sonia Raychaudhuri","is_ca":true},{"name":"Saim Wani","is_ca":false},{"name":"Shivansh Patel","is_ca":false},{"name":"Unnat Jain","is_ca":false},{"name":"Anne Lynn S. Chang","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02622123682191114,"gpt":0.3976567197951307,"spread":0.3714354829732195,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001853253,0.0007541559,0.0008261462,0.0006477664,0.0005494103,0.0006954458,0.001807616,0.001091132,0.001586684],"category_scores_gemma":[0.01388704,0.0004009423,0.0004857743,0.0005527799,0.0011935,0.003452491,0.001917273,0.001770964,0.0004299392],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007616959,"about_ca_system_score_gemma":0.001946118,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01074691,"about_ca_topic_score_gemma":0.01477339,"domain_scores_codex":[0.9986266,0.0004061881,0.00008607811,0.00043145,0.0003394982,0.0001102074],"domain_scores_gemma":[0.9945878,0.002776843,0.0006443874,0.0007009289,0.001010547,0.0002795683],"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":[0.001054516,0.0005193311,0.008880778,0.0005877467,0.0001393872,0.0002730511,0.001058723,0.2967217,0.02281029,0.01572783,0.007154617,0.6450721],"study_design_scores_gemma":[0.00004525778,0.0003080465,0.001792931,0.00002693252,0.00002323527,0.00008933361,0.00009969287,0.973577,0.006775008,0.0158027,0.00142949,0.0000303678],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05719918,0.0005740678,0.9377365,0.0001603027,0.00006926459,0.00008649895,0.0001815202,0.002998153,0.0009946027],"genre_scores_gemma":[0.8040998,0.0002497977,0.1928853,0.0001363511,0.00006686397,0.0001500189,0.0007980163,0.0002617998,0.001351966],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01074691,"threshold_uncertainty_score":0.02136868,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4231122779","doi":"10.18653/v1/2021.emnlp-main.288","title":"Automated Generation of Accurate & Fluent Medical X-ray Reports","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","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":"University of Alberta","funders":"","keywords":"Computer science; Fluency; Embedding; Generator (circuit theory); Transformer; Natural language processing; Artificial intelligence; Information retrieval; Linguistics","authors":[{"name":"Hoang Nga Nguyen","is_ca":true},{"name":"Dong Nie","is_ca":true},{"name":"Taivanbat Badamdorj","is_ca":false},{"name":"Yujie Liu","is_ca":true},{"name":"Yingying Zhu","is_ca":true},{"name":"Jason Truong","is_ca":true},{"name":"Cheng Li","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.09840819144765348,"gpt":0.4311545131288209,"spread":0.3327463216811675,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002156911,0.001502583,0.0007380606,0.001488018,0.0001871821,0.001144839,0.001704318,0.001013317,0.005209121],"category_scores_gemma":[0.01298163,0.0005293462,0.001158099,0.0006225577,0.0004265469,0.001269204,0.00169177,0.0008990154,0.003587427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004288535,"about_ca_system_score_gemma":0.0008585897,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0006545288,"about_ca_topic_score_gemma":0.0008831666,"domain_scores_codex":[0.9983838,0.0005927716,0.0001419191,0.0003954493,0.0004050018,0.00008110123],"domain_scores_gemma":[0.9920467,0.005440183,0.0005026309,0.0009599765,0.0008777683,0.0001728486],"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":[0.00121249,0.000375575,0.006043965,0.001572951,0.0002024853,0.00293525,0.000895316,0.09937468,0.06396499,0.009044558,0.04680929,0.7675684],"study_design_scores_gemma":[0.0002356734,0.000318624,0.002150985,0.0001321654,0.0001442515,0.001921802,0.0001877435,0.8355345,0.1178633,0.01776593,0.0236655,0.00007956575],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0452156,0.0006484048,0.9102278,0.0006249353,0.0002325203,0.0004003783,0.003748437,0.03680521,0.002096687],"genre_scores_gemma":[0.2848516,0.0005410219,0.6947877,0.0003286911,0.0002060666,0.0004142154,0.01250799,0.00266147,0.003701208],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005209121,"threshold_uncertainty_score":0.01742625,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3120832022","doi":"10.18653/v1/2021.emnlp-main.526","title":"Towards Zero-Shot Knowledge Distillation for Natural Language Processing","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":25,"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; Distillation; Task (project management); Benchmark (surveying); Artificial intelligence; Knowledge transfer; Natural language processing; Transfer of learning; Domain knowledge; Variety (cybernetics); Shot (pellet); Machine learning; Domain (mathematical analysis); Knowledge management; Engineering","authors":[{"name":"Ahmad Rashid","is_ca":false},{"name":"Vasileios Lioutas","is_ca":true},{"name":"Abbas Ghaddar","is_ca":false},{"name":"Mehdi Rezagholizadeh","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.08645918045858457,"gpt":0.4336183586249349,"spread":0.3471591781663503,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002804106,0.001473253,0.001283544,0.001031002,0.0008584849,0.001870214,0.002591586,0.002300761,0.003460127],"category_scores_gemma":[0.01026699,0.0005637097,0.0009698111,0.001060881,0.002212296,0.005875145,0.005291356,0.005015397,0.001858512],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001154235,"about_ca_system_score_gemma":0.001832829,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003193562,"about_ca_topic_score_gemma":0.005228842,"domain_scores_codex":[0.9982198,0.0006653931,0.00007379331,0.000416614,0.0004626081,0.0001617195],"domain_scores_gemma":[0.9959189,0.002680113,0.0001446402,0.000791066,0.0003164377,0.0001489747],"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":[0.0008926634,0.0005786886,0.001432345,0.0007288549,0.0001647401,0.0003145717,0.0007015468,0.4216012,0.01423633,0.09028196,0.02090286,0.4481641],"study_design_scores_gemma":[0.00003101061,0.00007478667,0.00009023704,0.00002461689,0.00001064329,0.00004874352,0.00005091702,0.9316118,0.005343916,0.06034838,0.002352605,0.0000123803],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04841625,0.001343383,0.9365972,0.00138038,0.0001463205,0.0001262248,0.000594325,0.006360498,0.005035567],"genre_scores_gemma":[0.563806,0.0007706717,0.4213525,0.001059048,0.0002138718,0.0003093253,0.002960196,0.0008664651,0.008661884],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003460127,"threshold_uncertainty_score":0.0148297,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3200130628","doi":"10.18653/v1/2021.emnlp-main.122","title":"Conditional probing: measuring usable information beyond a baseline","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":24,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Science Foundation","keywords":"Baseline (sea); USable; Representation (politics); Computer science; Word (group theory); Property (philosophy); Identity (music); Natural language processing; Artificial intelligence; Speech recognition; Mathematics","authors":[{"name":"John Hewitt","is_ca":false},{"name":"Kawin Ethayarajh","is_ca":false},{"name":"Percy Liang","is_ca":false},{"name":"Christopher D. Manning","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05992423051901477,"gpt":0.3690637550504681,"spread":0.3091395245314533,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.008933778,0.001324921,0.001264233,0.001337442,0.0006482222,0.002312578,0.001650814,0.002263468,0.002194627],"category_scores_gemma":[0.06834812,0.0007824252,0.0009894497,0.001445725,0.00306382,0.01035901,0.003767538,0.004318867,0.0003600494],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001101396,"about_ca_system_score_gemma":0.0007890836,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001113722,"about_ca_topic_score_gemma":0.000939293,"domain_scores_codex":[0.9949567,0.002229617,0.000307234,0.001368005,0.0008264302,0.0003120841],"domain_scores_gemma":[0.9248345,0.05753485,0.004006672,0.01053541,0.001919722,0.001168801],"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":[0.006123666,0.001388968,0.1078488,0.001821885,0.00166009,0.0006272247,0.002814818,0.3028664,0.1079394,0.1411985,0.00522641,0.3204839],"study_design_scores_gemma":[0.00009316794,0.001393226,0.02475498,0.0001407305,0.0003884749,0.0003913874,0.0003171033,0.6270015,0.03665104,0.306761,0.001921241,0.0001860263],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3439771,0.0006867602,0.6485376,0.0007606136,0.00005536628,0.0001451458,0.00100681,0.001461698,0.00336884],"genre_scores_gemma":[0.966808,0.0001368951,0.03143514,0.0001994227,0.00003521072,0.0001245323,0.0007592593,0.0001696582,0.000331786],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008933778,"threshold_uncertainty_score":0.04724693,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3197275000","doi":"10.18653/v1/2021.emnlp-main.181","title":"Unsupervised Conversation Disentanglement through Co-Training","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":20,"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":"Conversation; Computer science; Classifier (UML); Session (web analytics); Artificial intelligence; Machine learning; Reinforcement learning; Training set; Speech recognition; World Wide Web","authors":[{"name":"Hui Liu","is_ca":true},{"name":"Zhan Shi","is_ca":true},{"name":"Xiaodan Zhu","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1138268762368847,"gpt":0.4306470469350119,"spread":0.3168201706981271,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002365443,0.002356522,0.001427136,0.001635329,0.0008437277,0.001024576,0.002515421,0.001566844,0.003083415],"category_scores_gemma":[0.00864773,0.0007541566,0.001285241,0.001028,0.0008892327,0.003674195,0.003195519,0.004084158,0.002427146],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000758416,"about_ca_system_score_gemma":0.001531136,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00438437,"about_ca_topic_score_gemma":0.00933098,"domain_scores_codex":[0.9972361,0.0009276456,0.00009624664,0.001221073,0.0002821795,0.0002366774],"domain_scores_gemma":[0.9951775,0.002820145,0.0003651358,0.0006993425,0.0005706384,0.0003671747],"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":[0.001352426,0.0008854615,0.009288321,0.00040053,0.000271831,0.0002978882,0.001433833,0.08755701,0.03732668,0.003880849,0.01036651,0.8469387],"study_design_scores_gemma":[0.00002779297,0.0001135713,0.001378486,0.00002286004,0.00003908344,0.00009037356,0.0001542214,0.9798365,0.009430743,0.005638822,0.00323726,0.00003027279],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06670424,0.001080898,0.9213111,0.0003829105,0.0001565362,0.0002458704,0.0004833981,0.006586153,0.003048972],"genre_scores_gemma":[0.7221346,0.0003271153,0.2630728,0.0004192645,0.0002217147,0.0005777828,0.002810572,0.0006790292,0.009757007],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00438437,"threshold_uncertainty_score":0.01250976,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3213180921","doi":"10.18653/v1/2021.emnlp-main.603","title":"Universal-KD: Attention-based Output-Grounded Intermediate Layer Knowledge Distillation","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","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":"University of Waterloo","funders":"","keywords":"Computer science; Layer (electronics); Distillation; Matching (statistics); Interpretability; Projection (relational algebra); Base (topology); Space (punctuation); Architecture; Artificial intelligence; Deep learning; Computer architecture; Machine learning; Algorithm; Mathematics; Nanotechnology; Chromatography; Operating system; Materials science; Chemistry","authors":[{"name":"Yimeng Wu","is_ca":true},{"name":"Mehdi Rezagholizadeh","is_ca":true},{"name":"Abbas Ghaddar","is_ca":true},{"name":"Md Akmal Haidar","is_ca":true},{"name":"Ali Ghodsi","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07139261601228206,"gpt":0.3974343703737435,"spread":0.3260417543614614,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001106557,0.001486948,0.001243675,0.00120745,0.0007569246,0.001530648,0.003529372,0.001490115,0.007275992],"category_scores_gemma":[0.004092461,0.0006629612,0.001347269,0.001467243,0.001089838,0.005337355,0.004388689,0.003035701,0.002226051],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001197233,"about_ca_system_score_gemma":0.002089198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007131571,"about_ca_topic_score_gemma":0.01052419,"domain_scores_codex":[0.9991096,0.0001402738,0.00006651468,0.0003374857,0.0001781795,0.0001680753],"domain_scores_gemma":[0.9987871,0.0004130235,0.00008296138,0.0004182696,0.0002216544,0.00007696138],"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":[0.0003326776,0.0002347432,0.001060676,0.0003899106,0.0001364953,0.0002207704,0.0004677069,0.1124056,0.01539009,0.02642752,0.01294352,0.8299904],"study_design_scores_gemma":[0.00004073467,0.00008623925,0.0003244895,0.0000388218,0.00006153434,0.00009096831,0.00008379138,0.9381472,0.01706261,0.03694777,0.007083495,0.00003233843],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01735344,0.0005491757,0.9684529,0.0002849874,0.0001598253,0.0001003597,0.0003745075,0.009335288,0.003389508],"genre_scores_gemma":[0.5222674,0.0004082308,0.46205,0.0006167882,0.0001055348,0.0002622218,0.002481685,0.0007841099,0.01102404],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007275992,"threshold_uncertainty_score":0.02434063,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3209039755","doi":"10.18653/v1/2021.emnlp-main.821","title":"IndoNLI: A Natural Language Inference Dataset for Indonesian","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Universitas Indonesia; York University; Samsung; National Science Foundation","keywords":"Annotation; Indonesian; Sentence; Test set; Set (abstract data type); Inference; Data set","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.06648941278896932,"gpt":0.43358885394361,"spread":0.3670994411546407,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00136268,0.001464061,0.0008390729,0.002736604,0.001273521,0.001234693,0.001847341,0.001658905,0.01081394],"category_scores_gemma":[0.005044593,0.0004802059,0.0007408656,0.002232438,0.0006360148,0.002140397,0.00179399,0.002141649,0.01137615],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001744611,"about_ca_system_score_gemma":0.002205404,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01364511,"about_ca_topic_score_gemma":0.03218947,"domain_scores_codex":[0.9981651,0.0004857442,0.0002714191,0.0006153373,0.0003463194,0.0001161526],"domain_scores_gemma":[0.9973979,0.001006218,0.0002762184,0.0006135142,0.0005093798,0.0001968104],"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":[0.0007707928,0.0009670305,0.01749924,0.005246401,0.000241115,0.002156754,0.002256936,0.006330294,0.01786698,0.006519645,0.795597,0.1445479],"study_design_scores_gemma":[0.0003497424,0.0002511561,0.08519321,0.0006117635,0.0001963447,0.0026465,0.002662491,0.05062395,0.02261049,0.009092948,0.8254596,0.0003017908],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.05711932,0.002061597,0.02139697,0.00120106,0.0004040064,0.0008615474,0.8796455,0.0140655,0.02324462],"genre_scores_gemma":[0.03894811,0.0002585581,0.0232452,0.0003263016,0.0000501041,0.0008208487,0.9327601,0.0003824104,0.003208442],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01364511,"threshold_uncertainty_score":0.03617626,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3200285169","doi":"10.18653/v1/2021.emnlp-main.779","title":"PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","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":"McGill University; University of Toronto","funders":"","keywords":"Computer science; Decoding methods; Parsing; Language model; Natural language processing; Artificial intelligence; Code (set theory); Programming language; Compiler; Algorithm; Set (abstract data type)","authors":[{"name":"Torsten Scholak","is_ca":true},{"name":"Nathan Schucher","is_ca":false},{"name":"Dzmitry Bahdanau","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07064381815737095,"gpt":0.4144700036335565,"spread":0.3438261854761855,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001926959,0.003057839,0.001384508,0.001231691,0.0007987178,0.002664651,0.00312672,0.002208459,0.01455098],"category_scores_gemma":[0.01530036,0.001935573,0.0020474,0.001049302,0.001256367,0.003214001,0.003295782,0.004347242,0.008778794],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001160997,"about_ca_system_score_gemma":0.003515273,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009828807,"about_ca_topic_score_gemma":0.02609289,"domain_scores_codex":[0.9982396,0.0005936691,0.0001151244,0.0005737827,0.0003400421,0.0001377341],"domain_scores_gemma":[0.9947523,0.003542662,0.0001432899,0.0009580164,0.0004985082,0.0001052411],"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":[0.0006691882,0.0002255013,0.002398109,0.0008250527,0.0004597921,0.0006857183,0.000682593,0.2430716,0.02887361,0.02716466,0.07495188,0.6199923],"study_design_scores_gemma":[0.00005583789,0.00004607507,0.0001768212,0.00003722253,0.00004491465,0.0001352851,0.00005608327,0.9584664,0.01325303,0.01973841,0.007943274,0.00004669802],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007871318,0.0003333483,0.899148,0.0002669955,0.0001430389,0.000136701,0.00114753,0.08873478,0.002218243],"genre_scores_gemma":[0.1428921,0.0003146261,0.827105,0.0006312913,0.0001262331,0.000502458,0.00692154,0.0158948,0.005611944],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01455098,"threshold_uncertainty_score":0.04867786,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3093808828","doi":"10.18653/v1/2021.emnlp-main.740","title":"BARThez: a Skilled Pretrained French Sequence-to-Sequence Model","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Polytechnique Montréal","funders":"Centre National de la Recherche Scientifique","keywords":"Automatic summarization; Discriminative model; Computer science; Generative grammar; Benchmark (surveying); Transfer of learning; Artificial intelligence; Natural language processing; Sequence (biology); Code (set theory); Generative model; Field (mathematics); Language model; Machine learning; Programming language; Cartography","authors":[{"name":"Moussa Kamal Eddine","is_ca":false},{"name":"Antoine J.‐P. Tixier","is_ca":false},{"name":"Michalis Vazirgiannis","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.08687262609204593,"gpt":0.4277476868274459,"spread":0.3408750607354,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0006625992,0.001713466,0.0006130204,0.0008025641,0.0006907166,0.001179443,0.002435459,0.001802896,0.0134195],"category_scores_gemma":[0.002268163,0.0006697215,0.001342518,0.0006744096,0.0005022141,0.001528335,0.0009989153,0.002735292,0.006609989],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001344532,"about_ca_system_score_gemma":0.00187336,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.04578343,"about_ca_topic_score_gemma":0.08969801,"domain_scores_codex":[0.9996749,0.00007127652,0.00001215134,0.0001398103,0.00004663186,0.0000552225],"domain_scores_gemma":[0.9994627,0.0002758549,0.00002096438,0.00008421552,0.0001256527,0.00003058707],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006318222,0.0002601196,0.003288117,0.0003355122,0.0002898411,0.0005313192,0.0002664932,0.5248369,0.01381398,0.01605581,0.139776,0.2999141],"study_design_scores_gemma":[0.00004173888,0.00007220711,0.0004632197,0.00002193506,0.00002582498,0.00009028849,0.00004054637,0.9770685,0.002833025,0.004732363,0.01458263,0.00002770729],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09347856,0.00272059,0.7354111,0.002608416,0.001249976,0.0005385333,0.02318166,0.1077211,0.03309003],"genre_scores_gemma":[0.5475124,0.001001092,0.3122304,0.002443979,0.0004040922,0.001121124,0.0688137,0.005630249,0.06084294],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.04578343,"threshold_uncertainty_score":0.09103388,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3211384195","doi":"10.18653/v1/2021.emnlp-main.130","title":"Translation-based Supervision for Policy Generation in Simultaneous Neural Machine Translation","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":6,"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; Machine translation; Artificial intelligence; Translation (biology); Security token; Oracle; Reinforcement learning; Heuristic; Action (physics); Inference; Machine learning; Quality (philosophy); Sentence; Natural language processing; Programming language","authors":[{"name":"Ashkan Alinejad","is_ca":true},{"name":"Hassan S. Shavarani","is_ca":true},{"name":"Anoop Sarkar","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07715591203049524,"gpt":0.4211824962246504,"spread":0.3440265841941551,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003407454,0.0008492752,0.001037722,0.0004194127,0.0006573363,0.0006639799,0.001502561,0.001144116,0.002449974],"category_scores_gemma":[0.01604421,0.0006600831,0.0004260174,0.0004715534,0.001569547,0.002022476,0.001377004,0.002343032,0.0007340631],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001014568,"about_ca_system_score_gemma":0.002282703,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002956854,"about_ca_topic_score_gemma":0.004953816,"domain_scores_codex":[0.9982182,0.0008587032,0.0001027464,0.0004389664,0.0002599096,0.000121516],"domain_scores_gemma":[0.9900966,0.006831525,0.0007197004,0.001213961,0.0008919974,0.0002462267],"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":[0.0006749986,0.0004635644,0.002164782,0.0002573835,0.00006969392,0.0001924375,0.0004565527,0.6885932,0.01189706,0.01671385,0.002988741,0.2755277],"study_design_scores_gemma":[0.0000273636,0.00006393868,0.000107576,0.000008456806,0.000006267212,0.00002236248,0.00001338698,0.9889154,0.002819014,0.00761265,0.0003963343,0.000007221902],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04783933,0.0002076218,0.947091,0.0002823155,0.00005969371,0.0001071638,0.00006522608,0.002522047,0.0018255],"genre_scores_gemma":[0.8067012,0.00007836626,0.1909518,0.0002200304,0.00004903259,0.0002673903,0.000206745,0.0002248858,0.001300484],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003407454,"threshold_uncertainty_score":0.01802051,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3200853751","doi":"10.18653/v1/2021.emnlp-main.71","title":"Predicting emergent linguistic compositions through time: Syntactic frame extension via multimodal chaining","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Language, Metaphor, and Cognition","field":"Psychology","cited_by":6,"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 Toronto","funders":"Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; University of Toronto","keywords":"Computer science; Chaining; Natural language processing; Artificial intelligence; Linguistics; Noun; Psychology","authors":[{"name":"Lei Yu","is_ca":true},{"name":"Yang Xu","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05792580951332454,"gpt":0.4257146922206785,"spread":0.3677888827073539,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002143623,0.0004337032,0.0003639641,0.002107068,0.0005214307,0.001721784,0.0006400653,0.0007413772,0.00322005],"category_scores_gemma":[0.01571918,0.0004843728,0.0008919575,0.001315937,0.001115234,0.006860902,0.001557182,0.001173808,0.0006050062],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007737384,"about_ca_system_score_gemma":0.0003623268,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004532787,"about_ca_topic_score_gemma":0.005116552,"domain_scores_codex":[0.9995276,0.0002070285,0.00002004362,0.0001743651,0.00003895749,0.00003204932],"domain_scores_gemma":[0.9940592,0.004198589,0.0006123793,0.0005937714,0.0004109825,0.0001251226],"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":[0.0009616443,0.0003683374,0.258062,0.0005854857,0.0004847172,0.001413619,0.01145694,0.09678438,0.03635152,0.08688787,0.005609671,0.5010338],"study_design_scores_gemma":[0.0000385739,0.000149709,0.06725992,0.0001136852,0.0001388823,0.0005034218,0.001456854,0.7304212,0.00766693,0.1874772,0.004653309,0.0001202618],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7503597,0.0008136466,0.2405599,0.0005652127,0.00002832539,0.00005883464,0.0009864772,0.0006138855,0.006013935],"genre_scores_gemma":[0.9644827,0.0001969517,0.03374525,0.00004380051,0.00001320945,0.00007063334,0.0007790297,0.00008305103,0.0005853652],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.004532787,"threshold_uncertainty_score":0.01133668,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3201341200","doi":"10.18653/v1/2021.emnlp-main.42","title":"Mitigating Language-Dependent Ethnic Bias in BERT","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","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":"Kootenay Association for Science & Technology","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; National Research Foundation","keywords":"Categorical variable; Computer science; Ethnic group; German; Turkish; Natural language processing; Metric (unit); Language model; Linguistics; Arabic; Gender bias; Word (group theory); Artificial intelligence; Psychology; Machine learning; Sociology; Social psychology","authors":[{"name":"Jaimeen Ahn","is_ca":false},{"name":"Alice Oh","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1389738437343006,"gpt":0.44294323317028,"spread":0.3039693894359795,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004272232,0.0008389969,0.0006941632,0.001094268,0.0008362728,0.001104427,0.0007463333,0.0006599454,0.001094152],"category_scores_gemma":[0.01623391,0.0003104157,0.0005035033,0.001146101,0.000544026,0.002454872,0.001749516,0.001099458,0.0005672492],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000648275,"about_ca_system_score_gemma":0.0009455655,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007853147,"about_ca_topic_score_gemma":0.017672,"domain_scores_codex":[0.9982111,0.0009891189,0.00009777729,0.0002643817,0.0002802784,0.0001573593],"domain_scores_gemma":[0.9901862,0.006344861,0.0008828929,0.00113197,0.001232534,0.0002215069],"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":[0.00105989,0.0003117787,0.1883346,0.0003801563,0.0004619137,0.0006083252,0.003674763,0.3355414,0.01955374,0.02452218,0.007216659,0.4183345],"study_design_scores_gemma":[0.00002800053,0.00009348388,0.02014456,0.00003303359,0.0000610635,0.0002474686,0.001220544,0.9478813,0.007470477,0.01879464,0.003973505,0.00005197468],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7023168,0.0005701412,0.2899924,0.0005063235,0.00006307322,0.00007768157,0.0004201418,0.001561986,0.004491512],"genre_scores_gemma":[0.9588619,0.00009810258,0.0386928,0.00008421125,0.00003480532,0.00003776191,0.0007932257,0.0002068275,0.001190367],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007853147,"threshold_uncertainty_score":0.02259403,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3200172683","doi":"10.18653/v1/2021.emnlp-main.516","title":"Mind the Context: The Impact of Contextualization in Neural Module Networks for Grounding Visual Referring Expressions","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Multimodal Machine Learning Applications","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":"McGill University","funders":"","keywords":"Contextualization; Computer science; Generalization; Context (archaeology); Set (abstract data type); Exploit; Parameterized complexity; Artificial intelligence; Test set; Embedding; Cube (algebra); Machine learning; Algorithm; Mathematics; Programming language","authors":[{"name":"Arjun Akula","is_ca":false},{"name":"Spandana Gella","is_ca":false},{"name":"Keze Wang","is_ca":false},{"name":"Song‐Chun Zhu","is_ca":false},{"name":"Siva Reddy","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06160955148695172,"gpt":0.4607618939277148,"spread":0.399152342440763,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001078833,0.001486927,0.0005789296,0.0003838533,0.0004857542,0.001122078,0.001997448,0.00135476,0.003407358],"category_scores_gemma":[0.004440737,0.0005083755,0.0009924402,0.0004748066,0.0006960927,0.003650448,0.001688073,0.002028863,0.0009200707],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001024971,"about_ca_system_score_gemma":0.0006630484,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005161128,"about_ca_topic_score_gemma":0.0113629,"domain_scores_codex":[0.9993153,0.000257003,0.00002572054,0.000287773,0.00004824617,0.00006603645],"domain_scores_gemma":[0.9992201,0.000393972,0.00007566831,0.0001666888,0.00009838092,0.00004518438],"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":[0.0009555078,0.0003321617,0.008352417,0.0004603096,0.0003508991,0.0005309218,0.0007072735,0.427656,0.03855117,0.02106779,0.01364951,0.487386],"study_design_scores_gemma":[0.00003362792,0.0001242596,0.001189869,0.00005144695,0.0001031306,0.0001027571,0.00007210169,0.9757529,0.006327809,0.01197114,0.00424542,0.0000255478],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3448872,0.00407355,0.6210233,0.001968483,0.0003332584,0.0001859816,0.0009781845,0.01246224,0.01408789],"genre_scores_gemma":[0.882308,0.0005631006,0.1103881,0.0007483577,0.0001130486,0.0001387635,0.001403192,0.000444527,0.003892906],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005161128,"threshold_uncertainty_score":0.01139873,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3198457396","doi":"10.18653/v1/2021.emnlp-main.259","title":"Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"Simon Fraser University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Benchmark (surveying); Focus (optics); Noise (video); Code (set theory); Artificial intelligence; Training set; Machine learning; Spoken language; Resource (disambiguation); Noise reduction; Natural language processing; Speech recognition; Image (mathematics); Set (abstract data type)","authors":[{"name":"Yingmei Guo","is_ca":false},{"name":"Linjun Shou","is_ca":false},{"name":"Jian Pei","is_ca":true},{"name":"Ming Gong","is_ca":false},{"name":"Mingxing Xu","is_ca":false},{"name":"Zhiyong Wu","is_ca":false},{"name":"Daxin Jiang","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1489856121254087,"gpt":0.4560963868956432,"spread":0.3071107747702345,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0042676,0.002701546,0.002076249,0.001730398,0.001056099,0.003122623,0.002706836,0.002644282,0.004371716],"category_scores_gemma":[0.0153662,0.001053637,0.002489953,0.001855541,0.001406558,0.00583303,0.006328544,0.005297908,0.005076002],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007332758,"about_ca_system_score_gemma":0.001883215,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005940411,"about_ca_topic_score_gemma":0.01034889,"domain_scores_codex":[0.9955645,0.001855276,0.0002619963,0.001500295,0.0005838185,0.0002341198],"domain_scores_gemma":[0.991971,0.003786085,0.00030542,0.002572512,0.001196744,0.0001681702],"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":[0.0005967362,0.0007786389,0.005311682,0.0007123644,0.0005924959,0.0005149361,0.001732161,0.1053595,0.04347101,0.005847394,0.01966492,0.8154182],"study_design_scores_gemma":[0.00006635761,0.0002950373,0.002328047,0.0001310441,0.0001269569,0.0002544672,0.0009743865,0.9354633,0.02544989,0.01962884,0.0151581,0.0001235968],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0616038,0.001603002,0.9151431,0.0007920721,0.0004010904,0.0001948661,0.002226178,0.0145724,0.003463434],"genre_scores_gemma":[0.3517685,0.0006649341,0.6231773,0.0006669614,0.0001914008,0.0005908122,0.01607111,0.001329834,0.005539086],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005940411,"threshold_uncertainty_score":0.02256954,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3152747197","doi":"10.18653/v1/2021.emnlp-main.234","title":"Linguistic Dependencies and Statistical Dependence","year":2021,"lang":"en","type":"preprint","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University","funders":"","keywords":"Dependency (UML); Pointwise mutual information; Pointwise; Computer science; Context (archaeology); Natural language processing; ENCODE; Artificial intelligence; Simple (philosophy); Linguistics; Mathematics; Mutual information; Geography","authors":[{"name":"Jacob Louis Hoover","is_ca":true},{"name":"Wenyu Du","is_ca":false},{"name":"Alessandro Sordoni","is_ca":false},{"name":"Timothy J. O’Donnell","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06719722973920086,"gpt":0.4482117086445764,"spread":0.3810144789053755,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00415345,0.0004757527,0.0006136427,0.002332915,0.001136994,0.00169256,0.000953738,0.001121175,0.005549612],"category_scores_gemma":[0.0467152,0.0006208778,0.0006850646,0.002306164,0.003131602,0.004162072,0.002171518,0.002714469,0.001193465],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008119974,"about_ca_system_score_gemma":0.0006676134,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00251379,"about_ca_topic_score_gemma":0.00365825,"domain_scores_codex":[0.9960517,0.001548746,0.0002435591,0.001258538,0.0006652073,0.0002322959],"domain_scores_gemma":[0.9482195,0.04031291,0.004312425,0.004259543,0.002130572,0.0007649394],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001061146,0.0002889983,0.3667333,0.001168744,0.001109616,0.001496183,0.005972538,0.05847919,0.02531352,0.17418,0.01410685,0.3500899],"study_design_scores_gemma":[0.00004369789,0.0002224504,0.2739777,0.0002455754,0.0003335632,0.002032478,0.00137424,0.1769174,0.006789246,0.5224704,0.01538067,0.0002125792],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6530908,0.002601594,0.3124496,0.002844695,0.00013992,0.00009939467,0.002363202,0.000889124,0.02552169],"genre_scores_gemma":[0.9819409,0.0003767543,0.01472658,0.0002948958,0.0001100636,0.00008272255,0.001106868,0.0001673242,0.00119398],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005549612,"threshold_uncertainty_score":0.0219658,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3211439810","doi":"10.18653/v1/2021.emnlp-main.835","title":"Types of Out-of-Distribution Texts and How to Detect Them","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"York University","keywords":"Computer science; Categorization; Artificial intelligence; Calibration; Natural language processing; Data mining; Statistics; Mathematics","authors":[{"name":"Udit Arora","is_ca":false},{"name":"William C. Huang","is_ca":false},{"name":"He He","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0736749441821645,"gpt":0.3899331424867225,"spread":0.316258198304558,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005784863,0.0008816297,0.0009364073,0.006984505,0.001090183,0.004401062,0.001602334,0.002203989,0.002861118],"category_scores_gemma":[0.07817785,0.00055587,0.0007307596,0.003228913,0.001324851,0.007493149,0.002227371,0.001968752,0.002795136],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007571501,"about_ca_system_score_gemma":0.0007216017,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001408438,"about_ca_topic_score_gemma":0.002275611,"domain_scores_codex":[0.9901248,0.003549391,0.001108926,0.002215523,0.002642371,0.000358915],"domain_scores_gemma":[0.938959,0.04427333,0.005402951,0.005192767,0.005186823,0.0009851063],"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":[0.001073732,0.0003381698,0.3041928,0.004581575,0.000417866,0.002502373,0.008854931,0.007849592,0.02504239,0.02395971,0.06861564,0.5525713],"study_design_scores_gemma":[0.000212204,0.0003581676,0.2458381,0.002462381,0.0004628376,0.01605955,0.01956075,0.3659904,0.07120726,0.09008551,0.1872862,0.0004766865],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6693535,0.008470272,0.269127,0.007282251,0.001058917,0.0008162445,0.01598664,0.007240902,0.02066438],"genre_scores_gemma":[0.8251305,0.001683383,0.1519521,0.000688033,0.0003925987,0.0003822305,0.01446866,0.0009002696,0.004402319],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006984505,"threshold_uncertainty_score":0.03059369,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}