{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":9,"total_is_capped":false,"direct_labels_cover":0,"predictions_cover":9,"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":"20522714e91c","filters":{"venue":"Findings of the Association for Computational Linguistics: ACL 2022"}},"results":[{"id":"W4285255856","doi":"10.18653/v1/2022.findings-acl.177","title":"ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Data Visualization and Analytics","field":"Computer Science","cited_by":246,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"York University","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Question answering; Benchmark (surveying); Vocabulary; Chart; Visual reasoning; Artificial intelligence; Qualitative reasoning; Variety (cybernetics); Natural language processing; Linguistics","authors":[{"name":"Ahmed Masry","is_ca":true},{"name":"Xuan Long","is_ca":true},{"name":"Jia Qing Tan","is_ca":false},{"name":"Shafiq Joty","is_ca":false},{"name":"Enamul Hoque","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.00984609113906177,"gpt":0.2772540827683285,"spread":0.2674079916292668,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004111089,0.003212562,0.001137991,0.006394983,0.00126995,0.003200818,0.00385166,0.003255888,0.01390989],"category_scores_gemma":[0.04093298,0.0005008758,0.002468993,0.005371743,0.001048974,0.006262931,0.003256219,0.00258332,0.006823266],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00278679,"about_ca_system_score_gemma":0.003654334,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0253598,"about_ca_topic_score_gemma":0.02904454,"domain_scores_codex":[0.9921536,0.002666268,0.001132578,0.001637871,0.002019465,0.0003902448],"domain_scores_gemma":[0.9761578,0.01578541,0.001055666,0.002476447,0.003621586,0.0009031366],"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":[0.00115254,0.0012425,0.01147461,0.009581684,0.0003436166,0.0006114886,0.00168374,0.03366618,0.008763603,0.01714311,0.6539646,0.2603723],"study_design_scores_gemma":[0.0008232821,0.0008586244,0.01740245,0.001327336,0.0002316712,0.001029331,0.002954575,0.3833861,0.02238224,0.06013995,0.5092036,0.0002607812],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.1032715,0.01339643,0.1933505,0.006407084,0.00144971,0.004203472,0.4923486,0.1456257,0.03994691],"genre_scores_gemma":[0.1403304,0.002152541,0.266473,0.001352935,0.0002584041,0.00224625,0.5780025,0.002694445,0.00648953],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0253598,"threshold_uncertainty_score":0.0504244,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4285155190","doi":"10.18653/v1/2022.findings-acl.168","title":"Question Generation for Reading Comprehension Assessment by Modeling How and What to Ask","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","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":"University of Alberta","funders":"","keywords":"Reading comprehension; Computer science; Comprehension; Reading (process); Focus (optics); Artificial intelligence; Natural language processing; Mathematics education; Linguistics; Psychology","authors":[{"name":"Bilal Ghanem","is_ca":true},{"name":"Lauren Lutz Coleman","is_ca":false},{"name":"Julia Rivard Dexter","is_ca":false},{"name":"Spencer von der Ohe","is_ca":true},{"name":"Alona Fyshe","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02239636680911099,"gpt":0.2829373594837503,"spread":0.2605409926746393,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003277139,0.001959847,0.0006173209,0.002861357,0.0006599366,0.001675362,0.002768022,0.003081611,0.005406877],"category_scores_gemma":[0.02201586,0.0004409969,0.001813869,0.001502872,0.0006542341,0.004015435,0.001844676,0.003114311,0.005135349],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00144906,"about_ca_system_score_gemma":0.001316939,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006337006,"about_ca_topic_score_gemma":0.01428678,"domain_scores_codex":[0.9968746,0.001666036,0.0002225404,0.000853925,0.0002914026,0.00009143002],"domain_scores_gemma":[0.9868009,0.009019402,0.0005819158,0.001919951,0.001373178,0.0003045892],"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":[0.001366534,0.002315366,0.06082597,0.003394434,0.0006029955,0.0006955085,0.003123555,0.07085049,0.02249245,0.009141075,0.1324114,0.6927803],"study_design_scores_gemma":[0.0003295262,0.0007555957,0.02340589,0.00032725,0.0002600733,0.0006765441,0.0009786909,0.8290964,0.02397893,0.0310219,0.08902275,0.0001464381],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2221991,0.006494463,0.6018617,0.004943758,0.0007422758,0.003065375,0.08773612,0.05824472,0.01471258],"genre_scores_gemma":[0.3923664,0.0007003239,0.4157056,0.00116739,0.000195863,0.002426306,0.1812088,0.0008799477,0.005349403],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006337006,"threshold_uncertainty_score":0.01808786,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3206786886","doi":"10.18653/v1/2022.findings-acl.316","title":"Zero-Shot Dense Retrieval with Momentum Adversarial Domain Invariant Representations","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":19,"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; Invariant (physics); Classifier (UML); Source code; Embedding; Adversarial system; Artificial intelligence; Encoder; Autoencoder; Theoretical computer science; Pattern recognition (psychology); Algorithm; Deep learning; Mathematics","authors":[{"name":"Ji Xin","is_ca":true},{"name":"Chenyan Xiong","is_ca":false},{"name":"Ashwin Srinivasan","is_ca":false},{"name":"Ankita Sharma","is_ca":false},{"name":"Damien Jose","is_ca":false},{"name":"Paul N. Bennett","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01593641139419208,"gpt":0.2523305473785822,"spread":0.2363941359843901,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001718108,0.001094291,0.001365379,0.0006287016,0.000369049,0.0009073404,0.002249068,0.001440022,0.001910527],"category_scores_gemma":[0.004767039,0.0004048763,0.0007021881,0.0006744321,0.001096988,0.002965836,0.001835781,0.001861221,0.0009690588],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009377407,"about_ca_system_score_gemma":0.0007937246,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003816705,"about_ca_topic_score_gemma":0.004213269,"domain_scores_codex":[0.9991598,0.0002788957,0.00003578324,0.0002514094,0.0001779656,0.00009615626],"domain_scores_gemma":[0.9985297,0.0007499317,0.0001043781,0.0004278828,0.0001232251,0.00006493912],"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":[0.0003703822,0.000299211,0.0011855,0.0001935789,0.0001225153,0.0001749783,0.0001183179,0.7372962,0.008420791,0.02093991,0.009501789,0.2213767],"study_design_scores_gemma":[0.00001213437,0.00005767493,0.0001016836,0.00000511494,0.000007126003,0.0000409028,0.000008968065,0.9915775,0.001395103,0.006324028,0.0004615951,0.000008216227],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05794896,0.0009297201,0.9340972,0.0004616359,0.00008856666,0.0001472443,0.0003433411,0.003039991,0.002943286],"genre_scores_gemma":[0.8394619,0.0003885376,0.1493976,0.0006139449,0.0001313371,0.0001761514,0.001665701,0.0002188765,0.007946129],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003816705,"threshold_uncertainty_score":0.009086311,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3205231356","doi":"10.18653/v1/2022.findings-acl.135","title":"Morphosyntactic Tagging with Pre-trained Language Models for Arabic and its Dialects","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":13,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"York University; New York University Abu Dhabi","keywords":"Transformer; Arabic; Computer science; Modern Standard Arabic; Natural language processing; Artificial intelligence; Language model; Training set; Resource (disambiguation); Linguistics; Engineering; Voltage","authors":[{"name":"Go Inoue","is_ca":false},{"name":"Salam Khalifa","is_ca":false},{"name":"Nizar Habash","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.008924129542520672,"gpt":0.2522943333195564,"spread":0.2433702037770357,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001590058,0.002446197,0.0008345883,0.001559214,0.001009802,0.002342101,0.001837591,0.001221626,0.006922849],"category_scores_gemma":[0.006235406,0.0007230107,0.001233521,0.001292635,0.0006431009,0.003510089,0.001912848,0.002592636,0.009834396],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001289223,"about_ca_system_score_gemma":0.001387838,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01410134,"about_ca_topic_score_gemma":0.0216637,"domain_scores_codex":[0.998902,0.000272595,0.00008983855,0.0005129444,0.0001233097,0.00009931322],"domain_scores_gemma":[0.9966467,0.00165068,0.0001111057,0.0007681766,0.0007044837,0.0001188327],"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":[0.001080288,0.0005390086,0.01312763,0.001027456,0.000710432,0.0009175782,0.001414559,0.1035979,0.05795325,0.003359026,0.03956905,0.7767038],"study_design_scores_gemma":[0.000159145,0.0003236316,0.008412139,0.0002231477,0.0004567699,0.0008174787,0.001291554,0.8382786,0.09793703,0.01012375,0.04172909,0.0002476256],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.4244402,0.004453862,0.4374241,0.001193125,0.001477309,0.0004399508,0.01598232,0.09113926,0.02344993],"genre_scores_gemma":[0.6342845,0.0012259,0.3018605,0.0006494557,0.0001262914,0.0002666144,0.04681075,0.003675458,0.01110055],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01410134,"threshold_uncertainty_score":0.0280385,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4280645526","doi":"10.18653/v1/2022.findings-acl.164","title":"Richer Countries and Richer Representations","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":8,"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":"Embedding; Space (punctuation); Vocabulary; Inequality; Computer science; Word lists by frequency; Word (group theory); Power (physics); Work (physics); Natural language processing; Artificial intelligence; Econometrics; Linguistics; Economics; Mathematics; Engineering","authors":[{"name":"Kaitlyn Zhou","is_ca":false},{"name":"Kawin Ethayarajh","is_ca":false},{"name":"Dan Jurafsky","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01130151103574686,"gpt":0.2913571103954506,"spread":0.2800555993597038,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002420207,0.0005081706,0.0004743959,0.001880524,0.001212531,0.00400865,0.000467019,0.0006890437,0.01338013],"category_scores_gemma":[0.02420524,0.0003375814,0.0005194135,0.002404186,0.002002414,0.006975264,0.004294997,0.001118591,0.0009007286],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007200519,"about_ca_system_score_gemma":0.0004126468,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002906518,"about_ca_topic_score_gemma":0.003523169,"domain_scores_codex":[0.9973854,0.001268097,0.0001900163,0.0006002805,0.0002243815,0.0003318238],"domain_scores_gemma":[0.9862127,0.006931038,0.002134049,0.003596912,0.000657066,0.0004681486],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"observational","study_design_scores_codex":[0.001070304,0.0002614328,0.4152344,0.0006338989,0.0008294757,0.0009405969,0.02970107,0.02700936,0.01404026,0.241716,0.008277794,0.2602854],"study_design_scores_gemma":[0.0001170476,0.0005061375,0.2924188,0.0005584423,0.0005956396,0.001785957,0.03369395,0.0667417,0.009463291,0.5344805,0.05941068,0.0002277886],"study_design_candidate":"observational","study_design_consensus":"observational","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9346654,0.0004927738,0.03767307,0.002233279,0.0000542475,0.00002500419,0.001378608,0.000246747,0.02323092],"genre_scores_gemma":[0.9942263,0.0001081219,0.00406949,0.0001136412,0.00001449205,0.00001150918,0.0006308952,0.00004233015,0.0007832512],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01338013,"threshold_uncertainty_score":0.044761,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4285152239","doi":"10.18653/v1/2022.findings-acl.225","title":"Extracting Person Names from User Generated Text: Named-Entity Recognition for Combating Human Trafficking","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Authorship Attribution and Profiling","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"","funders":"Samsung; Institut de Valorisation des Données; Canadian Institute for Advanced Research","keywords":"Computer science; Named-entity recognition; Punctuation; Natural language processing; Domain (mathematical analysis); Task (project management); Artificial intelligence; Grammar; Named entity; Information retrieval; Linguistics","authors":[{"name":"Yifei Li","is_ca":false},{"name":"Pratheeksha Nair","is_ca":false},{"name":"Kellin Pelrine","is_ca":false},{"name":"Reihaneh Rabbany","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04314776999712625,"gpt":0.2892943644149413,"spread":0.246146594417815,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001195485,0.0007048884,0.0004314801,0.001900272,0.0004867804,0.0007049469,0.000770372,0.001146048,0.001621314],"category_scores_gemma":[0.003513259,0.0001556598,0.0004675522,0.001226919,0.0005181276,0.002551274,0.0009705076,0.0007263892,0.003418523],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003101302,"about_ca_system_score_gemma":0.0004459386,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001148711,"about_ca_topic_score_gemma":0.001878166,"domain_scores_codex":[0.9989896,0.000358493,0.0001060096,0.000289632,0.0001985506,0.00005774136],"domain_scores_gemma":[0.9975261,0.001245781,0.0003123847,0.0004803909,0.0003699767,0.00006538915],"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":[0.0005261494,0.0003688666,0.01103248,0.0008043283,0.0001106126,0.001679569,0.0007570369,0.01711802,0.04668313,0.003431828,0.04797442,0.8695135],"study_design_scores_gemma":[0.00007034902,0.0003652453,0.02153688,0.000169986,0.0001838781,0.003633951,0.001630623,0.6872189,0.1754163,0.01143346,0.09819223,0.0001481315],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3759468,0.004301409,0.5652753,0.002622926,0.00114382,0.0005189169,0.01332812,0.02530907,0.01155362],"genre_scores_gemma":[0.608487,0.001702283,0.3499383,0.0006134751,0.0003343271,0.0001859225,0.02847566,0.0004663651,0.009796586],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.001900272,"threshold_uncertainty_score":0.006322443,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4285148079","doi":"10.18653/v1/2022.findings-acl.293","title":"Local Structure Matters Most: Perturbation Study in NLU","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","field":"Computer Science","cited_by":8,"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; McGill University; Polytechnique Montréal; Mila - Quebec Artificial Intelligence Institute","funders":"Canada First Research Excellence Fund; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Perturbation (astronomy); Word order; Phenomenon; Artificial neural network; Invariant (physics); Artificial intelligence; Natural language processing; Mathematics; Physics","authors":[{"name":"Louis Clouâtre","is_ca":true},{"name":"Prasanna Parthasarathi","is_ca":true},{"name":"Amal Zouaq","is_ca":true},{"name":"Sarath Chandar","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01015079981417298,"gpt":0.2441797445615391,"spread":0.2340289447473661,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002466983,0.0003408302,0.0005736956,0.0004403256,0.0004108349,0.0009341705,0.000609073,0.0007648727,0.001004198],"category_scores_gemma":[0.03353014,0.0002995041,0.0003069224,0.0004278127,0.001296472,0.002709759,0.0009393271,0.00161894,0.0001893349],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000642793,"about_ca_system_score_gemma":0.0002884997,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001541064,"about_ca_topic_score_gemma":0.001158738,"domain_scores_codex":[0.9986019,0.0009196848,0.00004789995,0.0002370875,0.000142276,0.00005131652],"domain_scores_gemma":[0.9796521,0.01679408,0.0008062698,0.001939623,0.000496909,0.0003110039],"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":[0.002732866,0.001064435,0.0444022,0.0005336584,0.0004686507,0.0006758727,0.004042729,0.6395189,0.145335,0.03391996,0.002252222,0.1250535],"study_design_scores_gemma":[0.00004116764,0.0002901046,0.01446959,0.0000219124,0.00004894717,0.000108375,0.000371936,0.9240426,0.02087468,0.03900556,0.0006899219,0.00003526151],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9346654,0.0004768881,0.06195209,0.0005671018,0.00002773028,0.00004684801,0.00007889848,0.0003223946,0.001862655],"genre_scores_gemma":[0.9949645,0.00006364839,0.004639853,0.0000497527,0.00001291066,0.00001963836,0.00005190942,0.00004494576,0.0001527405],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.002466983,"threshold_uncertainty_score":0.0130468,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4226145103","doi":"10.18653/v1/2022.findings-acl.326","title":"On the data requirements of probing","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","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":"Vector Institute; University of Toronto","funders":"","keywords":"Computer science; Reliability (semiconductor); Context (archaeology); Construct (python library); Data mining; Machine learning; Artificial intelligence; Power (physics)","authors":[{"name":"Zining Zhu","is_ca":true},{"name":"Jixuan Wang","is_ca":true},{"name":"Bai Li","is_ca":true},{"name":"Frank Rudzicz","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05416273860118753,"gpt":0.291028757242492,"spread":0.2368660186413045,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.1514632,0.002077524,0.005660926,0.003211724,0.004348049,0.01122486,0.01005458,0.01074506,0.01188878],"category_scores_gemma":[0.685493,0.003752362,0.003197118,0.007854061,0.01009687,0.03424552,0.01430576,0.01449659,0.004753011],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003703604,"about_ca_system_score_gemma":0.006071488,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004637831,"about_ca_topic_score_gemma":0.003509584,"domain_scores_codex":[0.7993628,0.1466337,0.01085994,0.01743936,0.02287018,0.002834147],"domain_scores_gemma":[0.149344,0.7665843,0.006776336,0.0593323,0.01448048,0.003482618],"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":[0.009351456,0.001184792,0.05405917,0.002786642,0.0007923673,0.001582256,0.006661666,0.06693953,0.008518392,0.4485721,0.1200505,0.2795013],"study_design_scores_gemma":[0.0008555682,0.0004845809,0.006930318,0.0005116955,0.0001913161,0.00124559,0.002498881,0.2380625,0.003625033,0.719126,0.02631887,0.0001496617],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1293585,0.007681459,0.6953404,0.118013,0.002287666,0.001941252,0.01294484,0.006339601,0.02609326],"genre_scores_gemma":[0.5066476,0.002337239,0.446109,0.01365018,0.00233689,0.004114285,0.01786789,0.002705138,0.004231726],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.1514632,"threshold_uncertainty_score":0.8010237,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4281263937","doi":"10.18653/v1/2022.findings-acl.117","title":"Two-Step Question Retrieval for Open-Domain QA","year":2022,"lang":"en","type":"article","venue":"Findings of the Association for Computational Linguistics: ACL 2022","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true},"ca_institutions":"","funders":"Ministry of Science and ICT, South Korea; National Research Foundation of Korea; National Research Foundation","keywords":"Search engine indexing; Inference; Computer science; Information retrieval; Labrador Retriever; Pipeline (software); Squid; Domain (mathematical analysis); Artificial intelligence; Open domain; Question answering; Mathematics; Programming language","authors":[{"name":"Yeon Seonwoo","is_ca":false},{"name":"Juhee Son","is_ca":false},{"name":"Jiho Jin","is_ca":false},{"name":"Sang‐Woo Lee","is_ca":false},{"name":"Ji‐Hoon Kim","is_ca":false},{"name":"Jung-Woo Ha","is_ca":false},{"name":"Alice Oh","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01875750815666395,"gpt":0.2925474942045004,"spread":0.2737899860478364,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003484777,0.000825689,0.001031021,0.001333484,0.0006761005,0.001198632,0.002568392,0.001790962,0.009398755],"category_scores_gemma":[0.01089002,0.0006647364,0.00122987,0.0009476984,0.0009235206,0.005374018,0.002892502,0.002800946,0.005467695],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001233987,"about_ca_system_score_gemma":0.002147533,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008352006,"about_ca_topic_score_gemma":0.01053769,"domain_scores_codex":[0.9985671,0.0005922961,0.00008777731,0.000401242,0.0002341387,0.0001175059],"domain_scores_gemma":[0.9948623,0.002682291,0.0001662064,0.001366412,0.0007102555,0.000212625],"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":[0.001002704,0.0008574103,0.005205947,0.0009548726,0.0001828752,0.0002082218,0.0008622739,0.07719907,0.02542557,0.02889133,0.03443011,0.8247796],"study_design_scores_gemma":[0.00008736369,0.0001965959,0.0008150071,0.00002186545,0.00003672318,0.0001967949,0.00007878431,0.9617288,0.01028136,0.02015907,0.006366363,0.0000312849],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02775576,0.001041913,0.9518597,0.0008017536,0.0001089852,0.0003724749,0.0006811127,0.01422962,0.003148724],"genre_scores_gemma":[0.4850205,0.0004853074,0.5036844,0.0005585844,0.0001637943,0.0003594253,0.003324152,0.000459717,0.005944135],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.009398755,"threshold_uncertainty_score":0.03144199,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}