{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":2769,"total_is_capped":false,"direct_labels_cover":6,"predictions_cover":2769,"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":"e471de49da32","filters":{"topic":"Topic Modeling"}},"results":[{"id":"W1662133657","doi":"10.1613/jair.2934","title":"From Frequency to Meaning: Vector Space Models of Semantics","year":2010,"lang":"en","type":"article","venue":"Journal of Artificial Intelligence Research","topic":"Topic Modeling","field":"Computer Science","cited_by":2883,"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":"Semantics (computer science); Meaning (existential); Computer science; Context (archaeology); Space (punctuation); Perspective (graphical); Artificial intelligence; Programming language; Psychology","authors":[{"name":"Peter D. Turney","is_ca":true},{"name":"Patrick Pantel","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.2081680926678031,"gpt":0.4142808458992752,"spread":0.2061127532314721,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002809105,0.000881384,0.0009516493,0.004467224,0.0009208026,0.005835699,0.001745973,0.001375013,0.005364779],"category_scores_gemma":[0.01830118,0.0004164711,0.001536391,0.005264552,0.003830201,0.01662653,0.002092414,0.002227633,0.001012651],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001844048,"about_ca_system_score_gemma":0.001146197,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004393936,"about_ca_topic_score_gemma":0.002805263,"domain_scores_codex":[0.9975967,0.001343091,0.000144313,0.0003509118,0.0004302321,0.0001346202],"domain_scores_gemma":[0.9927769,0.00515273,0.0005381376,0.000708864,0.0006210369,0.0002022805],"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":[0.00006323949,0.00002575797,0.00116662,0.00009833342,0.0000396696,0.00004851053,0.0009852167,0.01037541,0.0001736005,0.9436175,0.002276377,0.04112974],"study_design_scores_gemma":[0.000007608036,0.00001382483,0.0001985163,0.00002451734,0.000008953666,0.00004504293,0.0001498627,0.0341371,0.00004400995,0.9621933,0.00316759,0.000009558034],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03187528,0.005143562,0.9355267,0.00529764,0.0002537654,0.0001121285,0.001066633,0.0004744394,0.02024987],"genre_scores_gemma":[0.7229849,0.005405884,0.258254,0.0008200248,0.00104879,0.0005380917,0.001479601,0.0002399448,0.009228783],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005835699,"threshold_uncertainty_score":0.01794696,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2020278455","doi":"10.1075/li.30.1.03nad","title":"A survey of named entity recognition and classification","year":2007,"lang":"en","type":"article","venue":"Lingvisticae Investigationes","topic":"Topic Modeling","field":"Computer Science","cited_by":2488,"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":"Natural Environment Research Council; Alfred P. Sloan Foundation","keywords":"Computer science; Natural language processing; Matching (statistics); Artificial intelligence; Field (mathematics); Information retrieval; Named entity; Key (lock); Word (group theory); Linguistics; Mathematics","authors":[{"name":"David R. Nadeau","is_ca":true},{"name":"Satoshi Sekine","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1391236913969597,"gpt":0.291564756940622,"spread":0.1524410655436623,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007817268,0.001834114,0.003062243,0.01431008,0.001421953,0.004739904,0.003972962,0.002092465,0.008388628],"category_scores_gemma":[0.01780604,0.0008817977,0.001803512,0.02716402,0.0008828376,0.01703105,0.002152723,0.00199369,0.009767558],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001537826,"about_ca_system_score_gemma":0.003004159,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003658784,"about_ca_topic_score_gemma":0.002882272,"domain_scores_codex":[0.9927514,0.001778281,0.000977919,0.001259781,0.002921331,0.0003113784],"domain_scores_gemma":[0.983254,0.008957142,0.000813954,0.001887972,0.004719707,0.0003672986],"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":[0.00005454073,0.0000934599,0.002592209,0.00291345,0.00009142201,0.0001106906,0.0001230192,0.001602677,0.0006050287,0.009097715,0.05034482,0.9323709],"study_design_scores_gemma":[0.00001528499,0.0001294233,0.004572304,0.001812747,0.0001558987,0.001274415,0.0005029165,0.01744683,0.003408942,0.01920094,0.9513692,0.0001112916],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.006312263,0.6061504,0.3328271,0.007809829,0.002659225,0.0005552437,0.006302957,0.005685422,0.03169764],"genre_scores_gemma":[0.03708992,0.679804,0.2334854,0.003396826,0.004006412,0.0005703187,0.02437643,0.0008212925,0.01644936],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.01431008,"threshold_uncertainty_score":0.04134214,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2963929190","doi":"10.18653/v1/k16-1028","title":"Abstractive Text Summarization using Sequence-to-sequence RNNs and Beyond","year":2016,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2209,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Automatic summarization; Sequence (biology); Computer science; Natural language processing; Artificial intelligence; Information retrieval; Biology; Genetics","authors":[{"name":"Ramesh Nallapati","is_ca":true},{"name":"Bowen Zhou","is_ca":true},{"name":"Cícero dos Santos","is_ca":false},{"name":"Çağlar Gülçehre","is_ca":true},{"name":"Bing Xiang","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07923942989030704,"gpt":0.3157808564587726,"spread":0.2365414265684656,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009583443,0.0014455,0.0008342246,0.001075477,0.000376924,0.001061412,0.00126108,0.0007744436,0.002281908],"category_scores_gemma":[0.003561313,0.0003564379,0.0008216841,0.0009254709,0.0003832125,0.002838042,0.0006351213,0.001340243,0.001615747],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006875195,"about_ca_system_score_gemma":0.0006075146,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004354774,"about_ca_topic_score_gemma":0.008747865,"domain_scores_codex":[0.9994697,0.0001607968,0.00004156701,0.0001879063,0.0001045379,0.00003549405],"domain_scores_gemma":[0.9985865,0.0005841265,0.000194377,0.0002270598,0.0003617035,0.00004626776],"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":[0.0004891588,0.0002220232,0.001141418,0.0007007819,0.0002871132,0.000245781,0.0003977656,0.3099568,0.06058573,0.01448652,0.01162537,0.5998616],"study_design_scores_gemma":[0.00001769523,0.0001340037,0.0004162074,0.00002579191,0.00006249711,0.0000509472,0.00004031743,0.9725658,0.01243722,0.008286667,0.005942193,0.00002061941],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02518982,0.002243538,0.9627937,0.0004432987,0.0002180939,0.0001396979,0.0007864041,0.005255697,0.002929612],"genre_scores_gemma":[0.397887,0.002118885,0.5819323,0.0004437651,0.0005012755,0.0002952498,0.00539711,0.0007423952,0.01068205],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004354774,"threshold_uncertainty_score":0.008658826,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2158139315","doi":"","title":"Word Representations: A Simple and General Method for Semi-Supervised Learning","year":2010,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1948,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Chunking (psychology); Computer science; Word (group theory); Natural language processing; Artificial intelligence; Simple (philosophy); Word embedding; Speech recognition; Linguistics","authors":[{"name":"Joseph Turian","is_ca":true},{"name":"Lev-Arie Ratinov","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03624813891496347,"gpt":0.3391485912100758,"spread":0.3029004522951123,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004092969,0.002760967,0.001530944,0.004293656,0.001087256,0.001931362,0.003724567,0.002411787,0.01011096],"category_scores_gemma":[0.01462329,0.001222705,0.002029267,0.003368736,0.0008834621,0.004780564,0.003272954,0.003751128,0.01631995],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006516128,"about_ca_system_score_gemma":0.001746582,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002612574,"about_ca_topic_score_gemma":0.005788106,"domain_scores_codex":[0.9957266,0.001378553,0.0004770391,0.001316453,0.0009151065,0.0001863342],"domain_scores_gemma":[0.9938969,0.002058394,0.0002887571,0.002408299,0.001213247,0.000134426],"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":[0.0001869539,0.000245567,0.001194687,0.000387638,0.0003165409,0.00008162599,0.0002685951,0.01689076,0.009251505,0.006789735,0.04343348,0.920953],"study_design_scores_gemma":[0.0001649326,0.0001659653,0.001404058,0.0001184231,0.0001621195,0.0004137868,0.0001967962,0.8556893,0.03292648,0.06581739,0.04279032,0.0001504212],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00154596,0.00009429794,0.979254,0.00006847882,0.00008381196,0.000260254,0.001133043,0.01664924,0.000910831],"genre_scores_gemma":[0.03739706,0.0001157867,0.9502493,0.0001315143,0.0001201234,0.001119256,0.004825043,0.002002099,0.004039853],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01011096,"threshold_uncertainty_score":0.03382456,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2962883855","doi":"10.1609/aaai.v30i1.9883","title":"Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models","year":2016,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1725,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"McGill University; Université de Montréal","funders":"Compute Canada; Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Canadian Institute for Advanced Research","keywords":"Generative grammar; Computer science; Bootstrapping (finance); Word (group theory); Artificial intelligence; End-to-end principle; Artificial neural network; Language model; Domain (mathematical analysis); Encoder; Recurrent neural network; Task (project management); Natural language processing; Generative model; Linguistics","authors":[{"name":"Iulian Vlad Serban","is_ca":true},{"name":"Alessandro Sordoni","is_ca":true},{"name":"Yoshua Bengio","is_ca":true},{"name":"Aaron Courville","is_ca":true},{"name":"Joëlle Pineau","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06428255094117762,"gpt":0.2761510849675109,"spread":0.2118685340263333,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001737216,0.0009312491,0.0008740866,0.0004200005,0.0005862525,0.001221811,0.001634221,0.001399191,0.002955623],"category_scores_gemma":[0.006055079,0.000776016,0.00103853,0.0003194733,0.0007866192,0.002655101,0.002053864,0.002058663,0.002163582],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007809123,"about_ca_system_score_gemma":0.0008228096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003029289,"about_ca_topic_score_gemma":0.005048819,"domain_scores_codex":[0.998995,0.0004078019,0.00004285455,0.0003666259,0.0001151818,0.00007258062],"domain_scores_gemma":[0.9977677,0.001519832,0.00009576759,0.0002274425,0.0002686506,0.0001205511],"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":[0.0004030473,0.0003926896,0.001549704,0.0003407688,0.0002402596,0.0004695529,0.001325044,0.7318091,0.03646794,0.02085556,0.004662604,0.2014837],"study_design_scores_gemma":[0.000008770001,0.00002183372,0.00005987901,0.000005056487,0.00001056829,0.00001855818,0.00002492895,0.9920222,0.002482798,0.004779813,0.0005575623,0.00000810927],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03063626,0.00015956,0.9621503,0.0001659702,0.00005901654,0.00009642297,0.0001307365,0.004947059,0.001654687],"genre_scores_gemma":[0.6042902,0.0001669144,0.3884238,0.0002743186,0.00005645283,0.0003704077,0.0009096847,0.000541444,0.004966798],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003029289,"threshold_uncertainty_score":0.009887516,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2889787757","doi":"10.18653/v1/d18-1259","title":"HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering","year":2018,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1582,"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":"Office of Naval Research; Defense Advanced Research Projects Agency; Université de Montréal; Nvidia; National Science Foundation","keywords":"Zhàng; Question answering; Computer science; Artificial intelligence; Natural language processing; Information retrieval; History; China; Archaeology","authors":[{"name":"Zhilin Yang","is_ca":true},{"name":"Peng Qi","is_ca":false},{"name":"Saizheng Zhang","is_ca":false},{"name":"Yoshua Bengio","is_ca":false},{"name":"William W. Cohen","is_ca":false},{"name":"Ruslan Salakhutdinov","is_ca":true},{"name":"Christopher D. Manning","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06108907148323083,"gpt":0.3121959901327288,"spread":0.2511069186494979,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001910017,0.003035023,0.00185651,0.004365229,0.001779899,0.001912627,0.004397795,0.005018116,0.01366162],"category_scores_gemma":[0.01088954,0.0006876175,0.002027706,0.003273329,0.0007104411,0.003933081,0.003996583,0.002343436,0.01112113],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001890202,"about_ca_system_score_gemma":0.002338488,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03047901,"about_ca_topic_score_gemma":0.06105387,"domain_scores_codex":[0.9976252,0.0006557292,0.0003007678,0.000658599,0.0005466761,0.0002130992],"domain_scores_gemma":[0.9956402,0.001926495,0.0002421895,0.0009267486,0.000837849,0.0004264489],"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":[0.0006742972,0.0004935014,0.005785217,0.002421837,0.0002927175,0.0005369638,0.0004589832,0.003229815,0.003928535,0.002575256,0.9368536,0.04274923],"study_design_scores_gemma":[0.001821499,0.0006766857,0.03553101,0.0008116472,0.0004916476,0.001493544,0.002508841,0.09806535,0.01014418,0.02200793,0.8260829,0.0003647891],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.03094041,0.004568587,0.01204815,0.002165876,0.0004489257,0.00085351,0.9264646,0.01579567,0.006714235],"genre_scores_gemma":[0.02339396,0.000307163,0.0136263,0.0004085373,0.00007070362,0.0005091894,0.9589603,0.0002326344,0.002491262],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03047901,"threshold_uncertainty_score":0.0606032,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2608787653","doi":"10.18653/v1/p17-1152","title":"Enhanced LSTM for Natural Language Inference","year":2017,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1196,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Ottawa; York University; National Research Council Canada","funders":"Fundamental Research Funds for the Central Universities; Chinese Academy of Sciences","keywords":"Inference; Computer science; Artificial intelligence; Parsing; Language model; Machine learning; Natural language; Artificial neural network; Natural language processing","authors":[{"name":"Qian Chen","is_ca":false},{"name":"Xiaodan Zhu","is_ca":true},{"name":"Zhen-Hua Ling","is_ca":false},{"name":"Si Wei","is_ca":false},{"name":"Hui Jiang","is_ca":true},{"name":"Diana Inkpen","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03843647516708471,"gpt":0.3355466537017409,"spread":0.2971101785346562,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009987364,0.001202806,0.0006858609,0.001016253,0.0003823774,0.001122316,0.001706151,0.001310482,0.00731934],"category_scores_gemma":[0.004881216,0.0004920878,0.0009851548,0.001606058,0.0004979695,0.004287284,0.001245187,0.002996072,0.003668611],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001233041,"about_ca_system_score_gemma":0.001412873,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007021696,"about_ca_topic_score_gemma":0.01308057,"domain_scores_codex":[0.9993805,0.0001500449,0.00003807979,0.0002377612,0.0001309463,0.00006266672],"domain_scores_gemma":[0.9988174,0.0006645861,0.00007009417,0.0002230773,0.0001898597,0.00003482414],"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":[0.0002956351,0.0002237013,0.001335976,0.0006265444,0.0002648855,0.0002562985,0.0002299345,0.2537316,0.02528692,0.03523801,0.03494487,0.6475657],"study_design_scores_gemma":[0.000009804585,0.00001995391,0.000198695,0.00002159179,0.00002986968,0.00004013455,0.00001341732,0.9536819,0.004368547,0.03698473,0.004620442,0.00001091546],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01874498,0.002402018,0.9557861,0.001189258,0.0003702113,0.00007172611,0.002907286,0.01348496,0.005043492],"genre_scores_gemma":[0.5102138,0.002065718,0.4638233,0.0008522755,0.0004815861,0.0002700006,0.00974422,0.0009083831,0.01164064],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.00731934,"threshold_uncertainty_score":0.02448559,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W196214544","doi":"","title":"Generating Text with Recurrent Neural Networks","year":2011,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1174,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto; University of New Brunswick","funders":"","keywords":"Recurrent neural network; Computer science; Language model; Sequence (biology); Artificial intelligence; Hessian matrix; Artificial neural network; Multiplicative function; Machine learning","authors":[{"name":"Ilya Sutskever","is_ca":true},{"name":"James Martens","is_ca":true},{"name":"Geoffrey E. Hinton","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04898282774934054,"gpt":0.2321847635042877,"spread":0.1832019357549472,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007395638,0.0007678604,0.0005374533,0.0005697153,0.0002939698,0.0007323058,0.00107683,0.0007901358,0.003664342],"category_scores_gemma":[0.004543199,0.0003750346,0.0007033515,0.0006588419,0.0003418204,0.001637001,0.0008149799,0.0008455946,0.002323077],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000457573,"about_ca_system_score_gemma":0.0003748521,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002018858,"about_ca_topic_score_gemma":0.003544691,"domain_scores_codex":[0.9994387,0.0002137054,0.0000383244,0.0001516767,0.0001225991,0.00003497452],"domain_scores_gemma":[0.9988744,0.0006064516,0.00009012345,0.0002096824,0.0001887668,0.00003057973],"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":[0.0002156957,0.00008671484,0.0009359212,0.0002363031,0.00007503143,0.0003330244,0.000289258,0.4016665,0.02102427,0.01586892,0.009547125,0.5497212],"study_design_scores_gemma":[0.000008880988,0.00002392488,0.0001011334,0.00000553905,0.000008793952,0.00002877259,0.00001266534,0.987852,0.003691896,0.006871236,0.001388268,0.000006979734],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0264299,0.0002234719,0.9650934,0.0002506649,0.00008230384,0.00007752419,0.0003417023,0.005190057,0.002311026],"genre_scores_gemma":[0.3560074,0.0003475481,0.6329969,0.0001782395,0.0001279155,0.0002205999,0.00188013,0.0006732316,0.007568044],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003664342,"threshold_uncertainty_score":0.01225847,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2963903950","doi":"10.18653/v1/d16-1230","title":"How NOT To Evaluate Your Dialogue System: An Empirical Study of Unsupervised Evaluation Metrics for Dialogue Response Generation","year":2016,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":916,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; McGill University","funders":"","keywords":"Computer science; Empirical research; Artificial intelligence; Machine learning; Natural language processing; Statistics; Mathematics","authors":[{"name":"Chia‐Wei Liu","is_ca":true},{"name":"Ryan Lowe","is_ca":true},{"name":"Iulian Vlad Serban","is_ca":true},{"name":"Laurent Charlin","is_ca":true},{"name":"Joëlle Pineau","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.2550579332141903,"gpt":0.3930564024462317,"spread":0.1379984692320415,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.0751058,0.001401588,0.001135208,0.002512992,0.00169474,0.002574272,0.00162826,0.001556051,0.0009687002],"category_scores_gemma":[0.3490035,0.0004293195,0.0006807425,0.002088597,0.001969258,0.004241251,0.002724957,0.002848998,0.00101577],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002045282,"about_ca_system_score_gemma":0.001492706,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003303829,"about_ca_topic_score_gemma":0.003641384,"domain_scores_codex":[0.8514024,0.1207503,0.006361726,0.008328507,0.01186841,0.001288628],"domain_scores_gemma":[0.4495643,0.4645077,0.0219516,0.02744881,0.03308805,0.003439408],"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":[0.005577881,0.003631657,0.2433226,0.003833797,0.00165635,0.0003542115,0.02150789,0.03878148,0.02918665,0.007494257,0.0266714,0.6179819],"study_design_scores_gemma":[0.0006611232,0.01004224,0.3679879,0.001184213,0.0008384131,0.001690172,0.008597869,0.4965291,0.05874176,0.01984653,0.03302925,0.0008514684],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8622093,0.002277613,0.1209213,0.0007072926,0.0002733945,0.001157392,0.001408417,0.002561749,0.008483575],"genre_scores_gemma":[0.9556538,0.0001480679,0.03950766,0.0002346297,0.00007561212,0.0008589956,0.001730532,0.0006894632,0.001101123],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9248942,"threshold_uncertainty_score":0.3972022,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2247119764","doi":"10.1609/aaai.v25i1.7917","title":"Learning Structured Embeddings of Knowledge Bases","year":2011,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":861,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"WordNet; Computer science; Artificial intelligence; Process (computing); Natural language processing; Information retrieval; Knowledge extraction; Word (group theory); Annotation; Programming language","authors":[{"name":"Antoine Bordes","is_ca":false},{"name":"Jason Weston","is_ca":false},{"name":"Ronan Collobert","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03699051865105631,"gpt":0.2521480397486367,"spread":0.2151575210975804,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001180441,0.0008318274,0.0006526685,0.001773969,0.000350453,0.001589307,0.001307648,0.001184998,0.001847181],"category_scores_gemma":[0.01101155,0.0006022921,0.0007576907,0.001586935,0.000750072,0.005076902,0.001854219,0.001694949,0.0006711553],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007710074,"about_ca_system_score_gemma":0.0007003506,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002066429,"about_ca_topic_score_gemma":0.003636197,"domain_scores_codex":[0.9990456,0.0003387778,0.00008335969,0.0002857021,0.0001712978,0.00007533197],"domain_scores_gemma":[0.9956378,0.002693242,0.0003066057,0.0006713303,0.0005792411,0.0001118383],"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":[0.0003316166,0.0003002608,0.005434514,0.0004841206,0.0001907749,0.0002582125,0.0007969268,0.2902363,0.006045471,0.06519171,0.009112429,0.6216176],"study_design_scores_gemma":[0.00001685337,0.00006237337,0.0003591555,0.00004145847,0.00002384099,0.00003106814,0.0001043099,0.9319354,0.00162387,0.06378744,0.002000625,0.00001362212],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07634914,0.0005315537,0.9178073,0.0006459523,0.00008850132,0.00008713041,0.0009658724,0.001539409,0.001985071],"genre_scores_gemma":[0.6783291,0.0008036844,0.3104516,0.0002591806,0.0001175094,0.0002863142,0.005705477,0.0001988801,0.003848277],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.002066429,"threshold_uncertainty_score":0.006242812,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"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":"W2557764419","doi":"10.18653/v1/w17-2623","title":"NewsQA: A Machine Comprehension Dataset","year":2017,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":742,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Microsoft (Canada)","funders":"","keywords":"Textual entailment; Computer science; Comprehension; Natural language processing; Artificial intelligence; Matching (statistics); Set (abstract data type); Word (group theory); Process (computing); Exploratory analysis; Information retrieval; Machine learning; Logical consequence; Data science; Linguistics","authors":[{"name":"Adam Trischler","is_ca":true},{"name":"Tong Wang","is_ca":true},{"name":"Xingdi Yuan","is_ca":true},{"name":"Justin Harris","is_ca":true},{"name":"Alessandro Sordoni","is_ca":true},{"name":"Philip Bachman","is_ca":true},{"name":"Kaheer Suleman","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05848331174708454,"gpt":0.3064087311719761,"spread":0.2479254194248916,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001344247,0.001652982,0.0006319559,0.003437625,0.0008120087,0.001091593,0.001919042,0.002348515,0.01284014],"category_scores_gemma":[0.008463047,0.0002734374,0.0008662831,0.002464755,0.0004041905,0.001697431,0.001117218,0.00147741,0.01026854],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001092425,"about_ca_system_score_gemma":0.001147083,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008136968,"about_ca_topic_score_gemma":0.01808344,"domain_scores_codex":[0.9985902,0.0004121923,0.0001829454,0.0003540167,0.0003623795,0.00009827971],"domain_scores_gemma":[0.9955834,0.001979935,0.0003145472,0.0006147433,0.00120407,0.0003034726],"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":[0.0003956275,0.0007825175,0.008898137,0.00153974,0.0001448285,0.0004278564,0.0006551954,0.00272624,0.005304055,0.001945508,0.9295832,0.04759708],"study_design_scores_gemma":[0.0008791256,0.0006414271,0.04242887,0.0003069358,0.0001802876,0.0009968383,0.001350488,0.03284175,0.01283882,0.007178825,0.9001929,0.0001636532],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.05718521,0.001757914,0.006196069,0.001651546,0.0002454357,0.0006567295,0.9136712,0.006984905,0.01165111],"genre_scores_gemma":[0.03252353,0.0002303554,0.008884204,0.00042317,0.0001024901,0.0006223237,0.9535761,0.0002135454,0.003424373],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.01284014,"threshold_uncertainty_score":0.0429545,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3022187094","doi":"10.1609/aaai.v31i1.10983","title":"A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues","year":2017,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":702,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"HEC Montréal; McGill University; Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Samsung; Compute Canada; Samsung Advanced Institute of Technology; Canadian Institute for Advanced Research","keywords":"Computer science; Latent variable; Generative grammar; Generative model; Latent variable model; Artificial intelligence; Variable (mathematics); Artificial neural network; Encoder; Task (project management); Machine learning; Mathematics","authors":[{"name":"Iulian Vlad Serban","is_ca":true},{"name":"Alessandro Sordoni","is_ca":false},{"name":"Ryan Lowe","is_ca":true},{"name":"Laurent Charlin","is_ca":true},{"name":"Joëlle Pineau","is_ca":true},{"name":"Aaron Courville","is_ca":true},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.09665800747503062,"gpt":0.2927783749652687,"spread":0.1961203674902381,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001655264,0.0007402865,0.0005835815,0.0006155845,0.0003768189,0.000722288,0.001455978,0.001201599,0.003662721],"category_scores_gemma":[0.005099402,0.0005496513,0.000800672,0.000713076,0.0006123235,0.001315971,0.0009337185,0.001921377,0.001228312],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009188981,"about_ca_system_score_gemma":0.001137751,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005200295,"about_ca_topic_score_gemma":0.008891263,"domain_scores_codex":[0.9992504,0.0003807216,0.00003457821,0.0001819077,0.00009407,0.00005836106],"domain_scores_gemma":[0.9980733,0.001463153,0.00009773751,0.0001174688,0.0001821834,0.00006614991],"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":[0.0005125357,0.0001837368,0.001935252,0.0002458591,0.0001312384,0.0002708102,0.0007241999,0.7618315,0.01051327,0.06516249,0.005012923,0.1534762],"study_design_scores_gemma":[0.00001203923,0.0000165757,0.00007630081,0.000004822493,0.000009237792,0.00001988621,0.000007029505,0.9926433,0.0005092242,0.006288574,0.0004072414,0.00000578479],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0157297,0.0002584345,0.9806574,0.0003901287,0.0000564687,0.00006978428,0.00031398,0.001135359,0.00138869],"genre_scores_gemma":[0.6180487,0.0003620579,0.3707715,0.0002779547,0.0001221277,0.0005419916,0.001178263,0.0003103024,0.00838696],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005200295,"threshold_uncertainty_score":0.01225305,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2131876387","doi":"10.1145/2661829.2661935","title":"A Latent Semantic Model with Convolutional-Pooling Structure for Information Retrieval","year":2014,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":687,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Explicit semantic analysis; Artificial intelligence; Natural language processing; Sentence; Word (group theory); Information retrieval; Context (archaeology); Probabilistic latent semantic analysis; Ranking (information retrieval); SemEval; Pooling; Semantic computing; Task (project management); Semantic Web; Semantic technology","authors":[{"name":"Yelong Shen","is_ca":false},{"name":"Xiaodong He","is_ca":false},{"name":"Jianfeng Gao","is_ca":false},{"name":"Li Deng","is_ca":false},{"name":"Grégoire Mesnil","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01307392977358029,"gpt":0.2120307665124632,"spread":0.1989568367388829,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009479622,0.0009199257,0.0009259533,0.001220985,0.0003034842,0.0009733903,0.001751665,0.001018625,0.001887561],"category_scores_gemma":[0.002198664,0.0004182531,0.001252157,0.001696749,0.0006762594,0.003418754,0.0007806181,0.001225545,0.00084992],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001666554,"about_ca_system_score_gemma":0.001301669,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01111124,"about_ca_topic_score_gemma":0.01183332,"domain_scores_codex":[0.9995865,0.000109218,0.00002727015,0.0001234644,0.0000895019,0.00006403626],"domain_scores_gemma":[0.9995794,0.0001769141,0.00006668597,0.00007660166,0.00007598135,0.00002437001],"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":[0.0005878149,0.0004095265,0.003555373,0.0004632919,0.0004675316,0.0003139733,0.0003150095,0.5191087,0.01779971,0.1201041,0.01347974,0.3233953],"study_design_scores_gemma":[0.0000104729,0.00003033202,0.0002007182,0.000007095259,0.00002885152,0.00002485108,0.000006947575,0.9781021,0.0009582904,0.01994591,0.0006743711,0.000009912858],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0242346,0.001219874,0.9701304,0.0006011374,0.0000638267,0.0000545525,0.0006239923,0.001494032,0.001577527],"genre_scores_gemma":[0.7926254,0.001424151,0.1944802,0.0004069175,0.0001539896,0.0002682151,0.002352222,0.0001453909,0.008143495],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01111124,"threshold_uncertainty_score":0.02209312,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W68132019","doi":"10.1007/s10994-013-5363-6","title":"A semantic matching energy function for learning with multi-relational data","year":2013,"lang":"en","type":"article","venue":"Machine Learning","topic":"Topic Modeling","field":"Computer Science","cited_by":687,"is_retracted":false,"has_abstract":false,"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; Agence Nationale de la Recherche; Compute Canada; Defense Advanced Research Projects Agency; Canadian Institute for Advanced Research","keywords":"Computer science; WordNet; Artificial intelligence; Statistical relational learning; Natural language processing; Sentence; Relational database; Relation (database); Semantic matching; Semantics (computer science); Parsing; Word (group theory); Context (archaeology); Information retrieval; Matching (statistics); Data mining","authors":[{"name":"Antoine Bordes","is_ca":false},{"name":"Xavier Glorot","is_ca":true},{"name":"Jason Weston","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03555843756672505,"gpt":0.2470883916188504,"spread":0.2115299540521254,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003498287,0.0005632806,0.001475651,0.002029199,0.0007871725,0.001934711,0.002505624,0.002531923,0.005321831],"category_scores_gemma":[0.009402822,0.0004965746,0.001609102,0.003234206,0.000885375,0.005068819,0.002869108,0.002297066,0.001514349],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001106068,"about_ca_system_score_gemma":0.001079065,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001519985,"about_ca_topic_score_gemma":0.001899035,"domain_scores_codex":[0.9982142,0.0005729099,0.0001640004,0.0003529068,0.0005693087,0.0001267424],"domain_scores_gemma":[0.9982937,0.000853479,0.00008196442,0.0004226259,0.0002750463,0.00007316673],"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":[0.0002806306,0.0004413804,0.001762119,0.0003799088,0.000263755,0.0001692141,0.0002355041,0.2019179,0.007652452,0.2326011,0.01328995,0.541006],"study_design_scores_gemma":[0.00001461601,0.00004744067,0.0004100374,0.00003027364,0.0000341292,0.00009620668,0.00004476045,0.8326504,0.001468844,0.1619386,0.003243565,0.00002106912],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006032728,0.0003473289,0.9916203,0.000274825,0.00006110873,0.00005359745,0.0001775814,0.0003534572,0.001079023],"genre_scores_gemma":[0.256939,0.0007897023,0.7314159,0.0004119966,0.0002138751,0.0005567508,0.001829553,0.0007069246,0.007136303],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005321831,"threshold_uncertainty_score":0.01850092,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2744909235","doi":"10.1016/j.neunet.2014.09.005","title":"Challenges in representation learning: A report on three machine learning contests","year":2014,"lang":"en","type":"article","venue":"Neural Networks","topic":"Topic Modeling","field":"Computer Science","cited_by":676,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo; University of Toronto; Université de Montréal","funders":"","keywords":"Computer science; Representation (politics); Artificial intelligence; Learning to learn; Feature learning; Machine learning; Mathematics education; Psychology","authors":[{"name":"Ian Goodfellow","is_ca":true},{"name":"Dumitru Erhan","is_ca":false},{"name":"Pierre Carrier","is_ca":true},{"name":"Aaron Courville","is_ca":true},{"name":"Mehdi Mirza","is_ca":true},{"name":"Ben Hamner","is_ca":false},{"name":"Will Cukierski","is_ca":false},{"name":"Yichuan Tang","is_ca":true},{"name":"David S. Thaler","is_ca":false},{"name":"Dong‐Hyun Lee","is_ca":false},{"name":"Yingbo Zhou","is_ca":false},{"name":"Chetan Ramaiah","is_ca":false},{"name":"Fangxiang Feng","is_ca":false},{"name":"Ruifan Li","is_ca":false},{"name":"Xiaojie Wang","is_ca":false},{"name":"Dimitris Athanasakis","is_ca":false},{"name":"John Shawe‐Taylor","is_ca":false},{"name":"Maxim Milakov","is_ca":false},{"name":"Cheol Park","is_ca":false},{"name":"Radu Tudor Ionescu","is_ca":false},{"name":"Marius Popescu","is_ca":false},{"name":"Cristian Grozea","is_ca":false},{"name":"James Bergstra","is_ca":true},{"name":"Jingjing Xie","is_ca":false},{"name":"Łukasz Romaszko","is_ca":false},{"name":"Bing Xu","is_ca":false},{"name":"Chuang Zhang","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07473609516329448,"gpt":0.2917676883922016,"spread":0.2170315932289071,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.03071552,0.00147908,0.002487253,0.003230849,0.003719024,0.009159641,0.003235385,0.004486783,0.00670192],"category_scores_gemma":[0.06079143,0.0005474741,0.001880783,0.0056941,0.002537386,0.01145816,0.008992145,0.006565975,0.002939785],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.003620216,"about_ca_system_score_gemma":0.00574043,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009163261,"about_ca_topic_score_gemma":0.009796759,"domain_scores_codex":[0.9890695,0.003201785,0.0007537503,0.001320839,0.004321615,0.001332483],"domain_scores_gemma":[0.9398355,0.03492837,0.001437289,0.004708101,0.01415034,0.004940446],"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":[0.0005149687,0.0005983409,0.004013119,0.00124861,0.0001785725,0.0002457879,0.001570619,0.005629183,0.002702539,0.05432916,0.4350744,0.4938948],"study_design_scores_gemma":[0.0001814034,0.000369,0.009283717,0.0005703353,0.0001442103,0.000588384,0.004605902,0.03801476,0.006953949,0.1485045,0.7906026,0.0001812481],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"commentary","genre_gemma":"empirical","genre_scores_codex":[0.1633583,0.1613529,0.221599,0.346857,0.02382824,0.001006739,0.008055005,0.002876824,0.07106601],"genre_scores_gemma":[0.5876262,0.07032181,0.1869653,0.02348068,0.02621556,0.001266268,0.03466753,0.002569672,0.06688704],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.03071552,"threshold_uncertainty_score":0.1624412,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4392193048","doi":"10.1038/s41591-024-02855-5","title":"Adapted large language models can outperform medical experts in clinical text summarization","year":2024,"lang":"en","type":"article","venue":"Nature Medicine","topic":"Topic Modeling","field":"Computer Science","cited_by":664,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"National Institute of Biomedical Imaging and Bioengineering; Foundation for the National Institutes of Health; National Institute of Arthritis and Musculoskeletal and Skin Diseases; Agency for Healthcare Research and Quality; National Institutes of Health; National Heart, Lung, and Blood Institute","keywords":"Automatic summarization; Computer science; Natural language processing; Artificial intelligence; Information retrieval; Medicine","authors":[{"name":"Dave Van Veen","is_ca":true},{"name":"Cara Van Uden","is_ca":false},{"name":"Louis Blankemeier","is_ca":false},{"name":"Jean-Benoit Delbrouck","is_ca":false},{"name":"Asad Aali","is_ca":false},{"name":"Christian Bluethgen","is_ca":false},{"name":"Anuj Pareek","is_ca":false},{"name":"Malgorzata Polacin","is_ca":false},{"name":"Eduardo Pontes Reis","is_ca":false},{"name":"Anna Seehofnerová","is_ca":false},{"name":"Nidhi Rohatgi","is_ca":false},{"name":"Poonam Hosamani","is_ca":false},{"name":"William Collins","is_ca":false},{"name":"Neera Ahuja","is_ca":false},{"name":"Curtis P. Langlotz","is_ca":false},{"name":"Jason Hom","is_ca":false},{"name":"Sergios Gatidis","is_ca":false},{"name":"John M. Pauly","is_ca":false},{"name":"Akshay Chaudhari","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02209648554058223,"gpt":0.3445173154923743,"spread":0.3224208299517921,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005629766,0.001541467,0.001116136,0.002364814,0.0005290118,0.001794644,0.001003833,0.001835313,0.002782171],"category_scores_gemma":[0.02221637,0.0005136384,0.001443735,0.001298014,0.0003036874,0.002192709,0.001107723,0.002158841,0.003270875],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005169972,"about_ca_system_score_gemma":0.001357205,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003054273,"about_ca_topic_score_gemma":0.00635845,"domain_scores_codex":[0.9972128,0.001617581,0.0002470075,0.0005325134,0.0002732049,0.000116949],"domain_scores_gemma":[0.9832591,0.01379226,0.0005079204,0.0007023882,0.001445,0.000293426],"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":[0.003712418,0.0008791186,0.01062725,0.001380241,0.001566835,0.0004612286,0.0007401545,0.1522861,0.02495134,0.001775609,0.05325057,0.7483692],"study_design_scores_gemma":[0.0002382022,0.0004884746,0.003555301,0.00008113742,0.0006406442,0.0001714362,0.0001745096,0.9717281,0.008998472,0.007135444,0.006720969,0.0000672688],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2112013,0.01315701,0.7369706,0.006000019,0.001673517,0.0006040697,0.007028454,0.01721436,0.006150641],"genre_scores_gemma":[0.7999325,0.002700444,0.1680123,0.001730043,0.001757392,0.0003673864,0.01747366,0.000920566,0.007105789],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005629766,"threshold_uncertainty_score":0.02977341,"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":"W3096266342","doi":"10.1007/978-3-030-61377-8_28","title":"BERTimbau: Pretrained BERT Models for Brazilian Portuguese","year":2020,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Topic Modeling","field":"Computer Science","cited_by":598,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Natural language processing; Artificial intelligence; Transformer; Language model; Sentence; Encoder; Portuguese; Transfer of learning; Textual entailment; Named-entity recognition; Logical consequence; Linguistics","authors":[{"name":"Fábio Souza","is_ca":false},{"name":"Rodrigo Nogueira","is_ca":true},{"name":"Roberto Lotufo","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02926096844476913,"gpt":0.252910922927005,"spread":0.2236499544822358,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009027429,0.001211053,0.0008981635,0.0005203742,0.0008505856,0.002034405,0.001774372,0.001501129,0.01519735],"category_scores_gemma":[0.004401272,0.0006696854,0.001250112,0.00081598,0.0004539629,0.002201563,0.001067516,0.001811847,0.002444646],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001978866,"about_ca_system_score_gemma":0.001749722,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.07336289,"about_ca_topic_score_gemma":0.08831627,"domain_scores_codex":[0.9997281,0.00009838288,0.00001407802,0.000058155,0.00004917988,0.00005199624],"domain_scores_gemma":[0.9993741,0.0003527912,0.0000390357,0.00008598504,0.0001022637,0.00004588922],"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":[0.0001779828,0.00006336613,0.001049537,0.0001358946,0.00005235831,0.0001755508,0.0002243216,0.7455327,0.0006086121,0.1708688,0.02407515,0.05703569],"study_design_scores_gemma":[0.00001191836,0.000007856795,0.0002106223,0.00002518172,0.00001329347,0.00002270986,0.00003092013,0.9444594,0.0002106398,0.04464785,0.01034836,0.00001131834],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08035384,0.003286482,0.7756815,0.00461344,0.0008435063,0.0001623093,0.01007311,0.004093669,0.1208921],"genre_scores_gemma":[0.7494774,0.00198554,0.1289776,0.0004105178,0.0002864029,0.0002636437,0.007433998,0.001927074,0.1092377],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.07336289,"threshold_uncertainty_score":0.1458718,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4400949264","doi":"10.1038/s41586-024-07566-y","title":"AI models collapse when trained on recursively generated data","year":2024,"lang":"en","type":"article","venue":"Nature","topic":"Topic Modeling","field":"Computer Science","cited_by":583,"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":"Generative grammar; Generative model; Intuition; Computer science; Variety (cybernetics); The Internet; Artificial intelligence; Data science; Psychology; Cognitive science; World Wide Web","authors":[{"name":"Ilia Shumailov","is_ca":false},{"name":"Zakhar Shumaylov","is_ca":false},{"name":"Yiren Zhao","is_ca":false},{"name":"Nicolas Papernot","is_ca":true},{"name":"Ross Anderson","is_ca":false},{"name":"Yarin Gal","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0509041440660831,"gpt":0.3005333002021868,"spread":0.2496291561361037,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.002722175,0.0008498321,0.0006725438,0.0007067595,0.0004652006,0.00128953,0.001320143,0.001493062,0.003209789],"category_scores_gemma":[0.01098684,0.0006960903,0.001163478,0.0006535237,0.0007109308,0.001979758,0.001246835,0.003530083,0.002546047],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001491201,"about_ca_system_score_gemma":0.001264034,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01773535,"about_ca_topic_score_gemma":0.02312865,"domain_scores_codex":[0.9992014,0.0002949761,0.00004611061,0.000282748,0.0001086898,0.00006620877],"domain_scores_gemma":[0.996143,0.002368566,0.0001219655,0.0008072818,0.0004578155,0.0001013999],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000181842,0.0001825316,0.007565845,0.0001867833,0.0003035087,0.00012224,0.0003487478,0.7034908,0.004747043,0.01621577,0.01024293,0.2564119],"study_design_scores_gemma":[0.000005852139,0.00002878201,0.0005506502,0.0000123686,0.0000116936,0.00002292409,0.00001828031,0.9916589,0.001391678,0.004964153,0.001327402,0.000007424414],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2919036,0.003011813,0.668986,0.002879228,0.000740539,0.0002357117,0.001571375,0.01091777,0.01975393],"genre_scores_gemma":[0.8662288,0.0005348307,0.119042,0.0007439231,0.0001159214,0.0001681255,0.002561962,0.0005013198,0.01010317],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9972778,"threshold_uncertainty_score":0.03526425,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2091812280","doi":"10.1145/1273496.1273577","title":"Three new graphical models for statistical language modelling","year":2007,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":582,"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","keywords":"Computer science; Probabilistic logic; Language model; Representation (politics); Word (group theory); n-gram; Binary number; Set (abstract data type); Parametric statistics; Artificial intelligence; Natural language processing; Statistical model; Sequence (biology); Theoretical computer science; Probability distribution; Mathematics; Programming language","authors":[{"name":"Andriy Mnih","is_ca":true},{"name":"Geoffrey E. Hinton","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05588757867207705,"gpt":0.2949680272604291,"spread":0.2390804485883521,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003715558,0.001353445,0.001255672,0.002196432,0.0006279988,0.003053521,0.003442994,0.002417107,0.004534267],"category_scores_gemma":[0.01641673,0.0007977593,0.002641621,0.002253524,0.001710114,0.004695405,0.003379181,0.003308084,0.001942136],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001575413,"about_ca_system_score_gemma":0.001218959,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004173832,"about_ca_topic_score_gemma":0.00560596,"domain_scores_codex":[0.9968266,0.001422457,0.0002187017,0.0005622492,0.0007763481,0.0001936784],"domain_scores_gemma":[0.9935021,0.004254694,0.0004706715,0.0008949306,0.0006241382,0.0002534551],"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":[0.0001813927,0.0001016091,0.00150674,0.0002694862,0.0001852446,0.0002782455,0.0004595977,0.1529681,0.002838111,0.6932817,0.009388517,0.1385413],"study_design_scores_gemma":[0.00003426183,0.00002872049,0.0002114043,0.00002984408,0.00004362778,0.0001332746,0.00002702769,0.5999753,0.0004526685,0.3915264,0.00749618,0.00004134311],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00181249,0.0003555019,0.9948315,0.0008353207,0.00007483315,0.00003885863,0.0003021453,0.0005263443,0.001222971],"genre_scores_gemma":[0.2039946,0.001926024,0.7809874,0.00126356,0.0005781467,0.0007144646,0.001692958,0.0004518624,0.008390965],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004534267,"threshold_uncertainty_score":0.01964998,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W10957333","doi":"","title":"Data-Driven Response Generation in Social Media","year":2011,"lang":"en","type":"article","venue":"NPARC","topic":"Topic Modeling","field":"Computer Science","cited_by":580,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false},"ca_institutions":"National Research Council Canada","funders":"","keywords":"Phrase; Computer science; Machine translation; Natural language processing; Artificial intelligence; Social media; Speech recognition","authors":[{"name":"Alan Ritter","is_ca":false},{"name":"Colin Cherry","is_ca":true},{"name":"William B. Dolan","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1716762906428433,"gpt":0.2990671782790738,"spread":0.1273908876362305,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003495304,0.0009677841,0.0007475673,0.0008967205,0.0004544355,0.0009840966,0.0014568,0.001142541,0.004125992],"category_scores_gemma":[0.01579889,0.0003712614,0.0007520653,0.0009377684,0.0005710644,0.001186431,0.001338652,0.001205892,0.00345325],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005362378,"about_ca_system_score_gemma":0.0007050914,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001162037,"about_ca_topic_score_gemma":0.001522,"domain_scores_codex":[0.9956616,0.003038253,0.000142285,0.0004907762,0.0005154285,0.0001516234],"domain_scores_gemma":[0.9852346,0.01116302,0.0005277305,0.001161073,0.001674265,0.0002394733],"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":[0.002076373,0.001079308,0.007029182,0.001253629,0.0002594544,0.001183132,0.002088505,0.1609831,0.1091319,0.0154452,0.02616386,0.6733064],"study_design_scores_gemma":[0.00009622826,0.0002626068,0.0008574524,0.00002159264,0.00002614485,0.0001776934,0.0003232078,0.9413615,0.03727432,0.01210301,0.007444626,0.0000515788],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06410319,0.0001789618,0.9171577,0.0008398948,0.0002189033,0.0006938631,0.001981745,0.01191168,0.002914032],"genre_scores_gemma":[0.5138577,0.0001190871,0.4771859,0.0003484307,0.0001658991,0.001183972,0.003607155,0.0006597027,0.002872198],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004125992,"threshold_uncertainty_score":0.01848519,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2126034021","doi":"10.1017/s1351324901002765","title":"Discovery of inference rules for question-answering","year":2001,"lang":"en","type":"article","venue":"Natural Language Engineering","topic":"Topic Modeling","field":"Computer Science","cited_by":532,"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":"Question answering; Inference; Computer science; Dependency (UML); Set (abstract data type); Parsing; Artificial intelligence; Natural language processing; Rule of inference; Programming language","authors":[{"name":"Dekang Lin","is_ca":true},{"name":"Patrick Pantel","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.00715054306742324,"gpt":0.2517287714730841,"spread":0.2445782284056608,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01285425,0.001676511,0.001928733,0.007963133,0.002171141,0.003862006,0.004943471,0.00287621,0.004924751],"category_scores_gemma":[0.08408172,0.00161744,0.00342029,0.003867732,0.002223716,0.009593174,0.003825296,0.004992203,0.003836379],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001631844,"about_ca_system_score_gemma":0.00260738,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003865298,"about_ca_topic_score_gemma":0.00538563,"domain_scores_codex":[0.9877671,0.004144249,0.001487827,0.003281333,0.002952466,0.0003670568],"domain_scores_gemma":[0.9239963,0.06099043,0.002626938,0.006629165,0.005140124,0.0006169625],"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":[0.0002953379,0.0006702197,0.01407066,0.001270908,0.0004968832,0.001019094,0.0034076,0.02648856,0.01118162,0.1192391,0.0280059,0.7938541],"study_design_scores_gemma":[0.00008116165,0.00005574021,0.002439339,0.0002422722,0.0002205779,0.0006252131,0.0005403052,0.5626093,0.01483298,0.3893543,0.02891343,0.00008537435],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006798803,0.0003653209,0.9827832,0.000800819,0.00005199747,0.0003577637,0.0009942934,0.006214378,0.001633365],"genre_scores_gemma":[0.05577256,0.0003113596,0.9375631,0.0003386975,0.0001213919,0.0003748164,0.004171626,0.000391837,0.0009544628],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01285425,"threshold_uncertainty_score":0.06798059,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W4411403346","doi":"10.1145/3744746","title":"A Comprehensive Overview of Large Language Models","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Intelligent Systems and Technology","topic":"Topic Modeling","field":"Computer Science","cited_by":488,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Ontario Tech University","funders":"","keywords":"Computer science; Context (archaeology); Benchmarking; Frontier; Data science; Engineering ethics; Management science; Political science; Engineering; Management","authors":[{"name":"Humza Naveed","is_ca":false},{"name":"Asad Ullah Khan","is_ca":false},{"name":"Shi Qiu","is_ca":false},{"name":"Muhammad Saqib","is_ca":false},{"name":"Saeed Anwar","is_ca":false},{"name":"Muhammad Usman","is_ca":true},{"name":"Naveed Akhtar","is_ca":false},{"name":"Nick Barnes","is_ca":false},{"name":"Ajmal Mian","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.04191685863724989,"gpt":0.307768274351576,"spread":0.2658514157143261,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003503855,0.001737053,0.001459119,0.004465573,0.0007505256,0.004408585,0.002607852,0.001740491,0.0101801],"category_scores_gemma":[0.01213802,0.0009723078,0.002192413,0.005422747,0.0008275392,0.006518172,0.002088067,0.002954376,0.007107844],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001757167,"about_ca_system_score_gemma":0.003431213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005581778,"about_ca_topic_score_gemma":0.006109043,"domain_scores_codex":[0.9973854,0.001045759,0.0002951229,0.0004219268,0.0007491369,0.0001026403],"domain_scores_gemma":[0.9939644,0.004345627,0.0002787711,0.0006233522,0.0006724796,0.0001154213],"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":[0.0001097258,0.0001183819,0.002125902,0.004219814,0.0003959105,0.0003827566,0.0004809221,0.0510775,0.002023827,0.1914044,0.09091377,0.656747],"study_design_scores_gemma":[0.00002325386,0.00008463606,0.001411013,0.001423393,0.0001828638,0.0006307563,0.0001858462,0.1938123,0.001348266,0.3229011,0.4778834,0.000113182],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.002703269,0.1749848,0.7742202,0.008656428,0.001108882,0.0003236845,0.00836263,0.005980137,0.02365993],"genre_scores_gemma":[0.1168833,0.2717656,0.5455529,0.004233782,0.00604511,0.001974741,0.03205152,0.002558035,0.01893493],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.0101801,"threshold_uncertainty_score":0.03405577,"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":"W2045929671","doi":"10.1145/1376815.1376819","title":"Semantic text similarity using corpus-based word similarity and string similarity","year":2008,"lang":"en","type":"article","venue":"ACM Transactions on Knowledge Discovery from Data","topic":"Topic Modeling","field":"Computer Science","cited_by":485,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Ottawa","funders":"","keywords":"Computer science; Similarity (geometry); Semantic similarity; String metric; Artificial intelligence; Word (group theory); Natural language processing; Longest common subsequence problem; Focus (optics); String searching algorithm; String (physics); Representation (politics); Information retrieval; Matching (statistics); Pattern matching; Mathematics; Algorithm","authors":[{"name":"Aminul Islam","is_ca":true},{"name":"Diana Inkpen","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1204337695552436,"gpt":0.3074205502158795,"spread":0.186986780660636,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002177997,0.00092376,0.001464652,0.01774631,0.0009860167,0.002976211,0.001630763,0.001285142,0.003514882],"category_scores_gemma":[0.01814917,0.000312062,0.001163446,0.01682147,0.001240919,0.005136708,0.002024637,0.00100161,0.001902967],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009804009,"about_ca_system_score_gemma":0.001356322,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001658718,"about_ca_topic_score_gemma":0.001637855,"domain_scores_codex":[0.9943928,0.001124886,0.0007004199,0.001007105,0.002637846,0.0001368597],"domain_scores_gemma":[0.9921194,0.003030686,0.0009637382,0.00113907,0.00249476,0.0002522486],"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":[0.0005017755,0.0004054433,0.01289896,0.001379963,0.0005654665,0.0004173881,0.001131468,0.02129064,0.02876124,0.05392328,0.01116528,0.8675591],"study_design_scores_gemma":[0.0001940287,0.0009286882,0.02707705,0.0004368933,0.0005551387,0.002814416,0.001698498,0.6720217,0.04891734,0.1823989,0.06249095,0.0004662709],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05191324,0.002126402,0.930313,0.0002680698,0.0004651728,0.0006221626,0.00230973,0.002587464,0.009394729],"genre_scores_gemma":[0.2841628,0.001106184,0.7038279,0.000128194,0.0005406998,0.001079478,0.00547775,0.0004178632,0.003259144],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01774631,"threshold_uncertainty_score":0.01175845,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2100002341","doi":"","title":"Replicated Softmax: an Undirected Topic Model","year":2009,"lang":"en","type":"article","venue":"Neural Information Processing Systems","topic":"Topic Modeling","field":"Computer Science","cited_by":442,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Latent Dirichlet allocation; Softmax function; Computer science; Graphical model; Inference; Artificial intelligence; Topic model; Latent variable; Undirected graph; Dirichlet distribution; Theoretical computer science; Data mining; Machine learning; Algorithm; Graph; Deep learning; Mathematics","authors":[{"name":"Geoffrey E. Hinton","is_ca":true},{"name":"Ruslan Salakhutdinov","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03266415990600878,"gpt":0.2715287157289674,"spread":0.2388645558229586,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00253677,0.001011345,0.001215775,0.00111654,0.0004340289,0.001841758,0.003925291,0.001801184,0.004651243],"category_scores_gemma":[0.008755062,0.0006968254,0.001414285,0.001778101,0.001081588,0.003112356,0.002037663,0.002789772,0.002248765],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009531957,"about_ca_system_score_gemma":0.0009476447,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002743036,"about_ca_topic_score_gemma":0.003869815,"domain_scores_codex":[0.9985086,0.0006480836,0.00007517381,0.0003823524,0.0002740597,0.000111612],"domain_scores_gemma":[0.9977815,0.001374932,0.0001762602,0.0003973173,0.0001944201,0.0000756003],"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":[0.0007663705,0.0003113491,0.002839805,0.0003650409,0.0003339312,0.0003258665,0.0003670751,0.5802077,0.01394393,0.1008362,0.01360971,0.286093],"study_design_scores_gemma":[0.00001601577,0.00002263188,0.0002337582,0.00001131884,0.0000230329,0.00004081248,0.000009474785,0.9637141,0.001556163,0.03282509,0.00153249,0.00001513398],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01010815,0.0001901748,0.9862272,0.0003714082,0.00005935847,0.00005593687,0.0004783367,0.00101642,0.00149297],"genre_scores_gemma":[0.5163999,0.0006980925,0.4654991,0.0007429844,0.0003108467,0.000798258,0.002707825,0.0005161922,0.01232681],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004651243,"threshold_uncertainty_score":0.01555997,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1654173042","doi":"","title":"Unsupervised Modeling of Twitter Conversations","year":2010,"lang":"en","type":"article","venue":"NPARC","topic":"Topic Modeling","field":"Computer Science","cited_by":434,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false},"ca_institutions":"National Research Council Canada","funders":"Microsoft Research","keywords":"Computer science; Conversation; Task (project management); Cluster analysis; Visualization; Domain (mathematical analysis); Artificial intelligence; Social media; Natural language processing; Microblogging; Topic model; Data science; World Wide Web; Linguistics","authors":[{"name":"Alan Ritter","is_ca":false},{"name":"Colin Cherry","is_ca":true},{"name":"Bill Dolan","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03234468296183593,"gpt":0.247503627607715,"spread":0.215158944645879,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001008454,0.0005647665,0.0005435738,0.001160927,0.0005663057,0.001183551,0.001118301,0.0007092286,0.001428792],"category_scores_gemma":[0.006810176,0.0004108484,0.0007966434,0.0008705125,0.0006282655,0.001916764,0.001034436,0.001079712,0.0006879616],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007663103,"about_ca_system_score_gemma":0.0007617376,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005841764,"about_ca_topic_score_gemma":0.007299322,"domain_scores_codex":[0.9989691,0.0004945821,0.00004307236,0.0002786739,0.0001232621,0.00009131733],"domain_scores_gemma":[0.9976798,0.001403945,0.0002683688,0.0002877445,0.00027171,0.00008854204],"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":[0.0005433768,0.0002388259,0.02698207,0.0003584452,0.0002629847,0.0004497948,0.003433864,0.6984304,0.01961686,0.09021916,0.009850348,0.1496139],"study_design_scores_gemma":[0.000004942787,0.000009051157,0.00121353,0.000005403518,0.000006131157,0.00002595493,0.00006912276,0.9860108,0.0006517785,0.01067461,0.001320543,0.00000821147],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.11056,0.0002992035,0.8813415,0.000597939,0.00005559218,0.0001543512,0.001693289,0.001068754,0.004229306],"genre_scores_gemma":[0.8848689,0.0003092695,0.1065673,0.00009910824,0.0001573223,0.0004229722,0.002932727,0.0002598317,0.004382703],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005841764,"threshold_uncertainty_score":0.01161551,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2399456070","doi":"10.21437/interspeech.2013-596","title":"Investigation of recurrent-neural-network architectures and learning methods for spoken language understanding","year":2013,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":432,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Artificial intelligence; Spoken language; Artificial neural network; Natural language processing; Recurrent neural network","authors":[{"name":"Grégoire Mesnil","is_ca":true},{"name":"Xiaodong He","is_ca":true},{"name":"Li Deng","is_ca":false},{"name":"Yoshua Bengio","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0788804892206361,"gpt":0.329259228172978,"spread":0.2503787389523419,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003871588,0.0009838898,0.000629087,0.0006642244,0.0003391285,0.001034648,0.001641056,0.0009728762,0.00223805],"category_scores_gemma":[0.007587748,0.000485071,0.0005795592,0.0006231792,0.0003905948,0.003516411,0.0006893431,0.001848116,0.0007099769],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009496188,"about_ca_system_score_gemma":0.0009506891,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007121918,"about_ca_topic_score_gemma":0.01051028,"domain_scores_codex":[0.9993325,0.0002581689,0.00003857666,0.0001906834,0.0001190682,0.00006099109],"domain_scores_gemma":[0.9969962,0.001843985,0.0001605623,0.0002889866,0.0006336346,0.0000767626],"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":[0.0002202967,0.0002148119,0.002218265,0.0002376101,0.0001800082,0.000097116,0.0002677,0.6202718,0.01088962,0.02119873,0.002546108,0.3416579],"study_design_scores_gemma":[0.000004346771,0.00003575477,0.0001338044,0.000009121521,0.00001202527,0.00001038029,0.00001411422,0.9947379,0.001725879,0.002885603,0.0004257537,0.000005312399],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09572335,0.002678476,0.8915288,0.0008003135,0.000115919,0.00009542512,0.0002930383,0.00370084,0.005063816],"genre_scores_gemma":[0.6862236,0.001507006,0.3050776,0.0002122757,0.00008360602,0.0001352816,0.0009444337,0.0002810367,0.005535188],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007121918,"threshold_uncertainty_score":0.02047515,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3100107515","doi":"10.18653/v1/2020.findings-emnlp.63","title":"Document Ranking with a Pretrained Sequence-to-Sequence Model","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":414,"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; Ranking (information retrieval); Sequence (biology); Encoder; Relevance (law); Artificial intelligence; Transformer; Task (project management); Information retrieval; Machine learning; Natural language processing; Data mining; Engineering","authors":[{"name":"Rodrigo Nogueira","is_ca":false},{"name":"Zhiying Jiang","is_ca":false},{"name":"Ronak Pradeep","is_ca":false},{"name":"Jimmy Lin","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07025758913945321,"gpt":0.2760678251045512,"spread":0.205810235965098,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001633267,0.001034252,0.001177197,0.001172069,0.0004366365,0.00137022,0.001907405,0.001648138,0.006395116],"category_scores_gemma":[0.004292347,0.0005460125,0.0009966518,0.00131691,0.0006395476,0.003314969,0.0006740945,0.002312725,0.004700397],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001316016,"about_ca_system_score_gemma":0.001346013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005607946,"about_ca_topic_score_gemma":0.01160927,"domain_scores_codex":[0.999314,0.0002198881,0.0000429348,0.0001990787,0.00014369,0.00008055528],"domain_scores_gemma":[0.9982114,0.0009538711,0.0001320284,0.000281724,0.0003517485,0.00006917467],"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":[0.0003507402,0.0003121787,0.001335291,0.0002294301,0.0001054394,0.000151051,0.00009665953,0.6466268,0.00843034,0.01890357,0.01125096,0.3122075],"study_design_scores_gemma":[0.00001330891,0.00007242337,0.0001593582,0.000006927297,0.00001522267,0.00004709929,0.000007775307,0.9898121,0.001243443,0.007769107,0.000842837,0.0000104576],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0389448,0.0006051779,0.9490086,0.0007372547,0.0001474855,0.0002364425,0.0007849678,0.004130311,0.005404962],"genre_scores_gemma":[0.7078541,0.0006593933,0.2504365,0.00078615,0.0003527096,0.0006784671,0.002845371,0.000540605,0.03584675],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006395116,"threshold_uncertainty_score":0.02139384,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2252143850","doi":"","title":"Training recurrent neural networks","year":2013,"lang":"en","type":"dissertation","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":391,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Recurrent neural network; Initialization; Computer science; Sequence (biology); Artificial intelligence; Probabilistic logic; Gradient descent; Artificial neural network; Machine learning","authors":[{"name":"Geoffrey E. Hinton","is_ca":true},{"name":"Ilya Sutskever","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05334820093018022,"gpt":0.2900405536168376,"spread":0.2366923526866574,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001246452,0.0009404979,0.0008054597,0.0004271472,0.0003213097,0.0007097222,0.001250517,0.001037356,0.002523617],"category_scores_gemma":[0.005072165,0.0007340946,0.0007079023,0.0004503424,0.0004912346,0.001688757,0.0008513911,0.001475781,0.001044539],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007209933,"about_ca_system_score_gemma":0.0007994781,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003757732,"about_ca_topic_score_gemma":0.005838322,"domain_scores_codex":[0.9994215,0.0001587683,0.00003974407,0.0001830287,0.0001368809,0.0000600058],"domain_scores_gemma":[0.9988716,0.0006211288,0.00008946331,0.0001722756,0.0002122792,0.00003324347],"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":[0.00004351595,0.00004126527,0.0007837329,0.0001280584,0.00007154793,0.00004396626,0.00006913942,0.854232,0.005285557,0.01410209,0.00204717,0.1231519],"study_design_scores_gemma":[0.000003198337,0.00001319239,0.00006070561,0.000007467064,0.000005526495,0.000006534808,0.000003792675,0.9946853,0.00121044,0.003398882,0.0006016887,0.000003187521],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.03129365,0.0005644902,0.9614362,0.0002752891,0.0001022688,0.00005505772,0.0001360819,0.001451693,0.004685196],"genre_scores_gemma":[0.6097412,0.0008423302,0.379577,0.0002867925,0.0001095615,0.0002660913,0.0008437416,0.0004292493,0.00790417],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003757732,"threshold_uncertainty_score":0.008442402,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3043222912","doi":"10.3389/frai.2020.00042","title":"Using Topic Modeling Methods for Short-Text Data: A Comparative Analysis","year":2020,"lang":"en","type":"article","venue":"Frontiers in Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":368,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Ottawa","funders":"","keywords":"Latent Dirichlet allocation; Topic model; Computer science; Latent semantic analysis; Popularity; Information retrieval; Social media; Matrix decomposition; Context (archaeology); Artificial intelligence; Precision and recall; Natural language processing; Data science; Machine learning; Data mining; World Wide Web","authors":[{"name":"Rania Albalawi","is_ca":true},{"name":"Tet Yeap","is_ca":true},{"name":"Morad Benyoucef","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.4952072856871249,"gpt":0.4730338297708425,"spread":0.02217345591628234,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.02162341,0.001628618,0.001670609,0.01362742,0.0008241962,0.003625943,0.001420676,0.001817168,0.001288501],"category_scores_gemma":[0.05717194,0.0004620613,0.002518149,0.01202565,0.0007248482,0.007651937,0.001170807,0.001568822,0.001011166],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001083974,"about_ca_system_score_gemma":0.001122753,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002795435,"about_ca_topic_score_gemma":0.00255467,"domain_scores_codex":[0.9852224,0.008075548,0.001204108,0.001945828,0.003314315,0.0002377672],"domain_scores_gemma":[0.8810503,0.1052997,0.002955456,0.003512815,0.006563126,0.0006186923],"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":[0.001088889,0.0004043071,0.04807114,0.005398154,0.002497919,0.0003130276,0.002553266,0.03573433,0.003815238,0.01169664,0.0121784,0.8762488],"study_design_scores_gemma":[0.0001890472,0.001084155,0.09188557,0.001883877,0.001972463,0.001714524,0.004385911,0.7781143,0.008227656,0.03391255,0.07618051,0.000449441],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1432417,0.1404304,0.6987153,0.0029354,0.001111174,0.000753464,0.004320539,0.002074994,0.006416977],"genre_scores_gemma":[0.5255121,0.05606867,0.397253,0.0004059318,0.002110985,0.001026572,0.01501528,0.0006730448,0.001934416],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02162341,"threshold_uncertainty_score":0.1143569,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2912817604","doi":"10.18653/v1/n19-4013","title":"End-to-End Open-Domain Question Answering with","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":368,"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":"Question answering; Computer science; Open domain; Benchmark (surveying); Information retrieval; Reading (process); End-to-end principle; Natural language processing; Artificial intelligence; Domain (mathematical analysis); Reading comprehension; Language model; Open source; Programming language; Linguistics; Software","authors":[{"name":"Wei Yang","is_ca":false},{"name":"Yuqing Xie","is_ca":false},{"name":"Aileen Lin","is_ca":false},{"name":"Xingyu Li","is_ca":true},{"name":"Luchen Tan","is_ca":false},{"name":"Kun Xiong","is_ca":false},{"name":"Ming Li","is_ca":false},{"name":"Jimmy Lin","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02837405652156753,"gpt":0.2801999427192826,"spread":0.2518258861977151,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004099092,0.002476443,0.001933936,0.002387334,0.001346115,0.003964356,0.003276713,0.003357423,0.0267848],"category_scores_gemma":[0.01264427,0.0009680649,0.001991447,0.001846122,0.0007920894,0.007636747,0.009914985,0.003138681,0.03433625],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007565882,"about_ca_system_score_gemma":0.00105096,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002672359,"about_ca_topic_score_gemma":0.006292839,"domain_scores_codex":[0.9950918,0.001703465,0.0003339094,0.001730843,0.0008010856,0.0003389],"domain_scores_gemma":[0.9934837,0.003339895,0.0001484884,0.001863637,0.0008465429,0.0003177177],"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":[0.002543512,0.001018007,0.002410186,0.001192325,0.0005067444,0.0009380208,0.001902396,0.007788813,0.02027892,0.01885165,0.3589087,0.5836607],"study_design_scores_gemma":[0.0005034279,0.0003750497,0.001926737,0.0002125182,0.0002614712,0.0006683469,0.001901749,0.5731444,0.04403492,0.1470115,0.2297667,0.0001931923],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01717217,0.001689264,0.7188481,0.001406013,0.000807245,0.0007644944,0.01472607,0.2309439,0.01364269],"genre_scores_gemma":[0.1843385,0.0005287297,0.727059,0.001332212,0.0004135293,0.001071957,0.05978991,0.005131938,0.02033425],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0267848,"threshold_uncertainty_score":0.08960408,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2132052677","doi":"10.14569/ijacsa.2015.060121","title":"A Survey of Topic Modeling in Text Mining","year":2015,"lang":"en","type":"article","venue":"International Journal of Advanced Computer Science and Applications","topic":"Topic Modeling","field":"Computer Science","cited_by":368,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Concordia University","funders":"","keywords":"Latent Dirichlet allocation; Topic model; Computer science; Probabilistic latent semantic analysis; Natural language processing; Artificial intelligence; Latent semantic analysis; Field (mathematics); Probabilistic logic; Information retrieval","authors":[{"name":"Rubayyi Alghamdi","is_ca":true},{"name":"Khalid Alfalqi","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0669373198898449,"gpt":0.3294273301311625,"spread":0.2624900102413176,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007165624,0.002088693,0.003745923,0.009564827,0.001374747,0.004509262,0.003041179,0.002203124,0.002771244],"category_scores_gemma":[0.01681403,0.001105343,0.003300958,0.01821085,0.0009602183,0.008731122,0.001771802,0.002651416,0.002589453],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001565168,"about_ca_system_score_gemma":0.002375254,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004097731,"about_ca_topic_score_gemma":0.003425356,"domain_scores_codex":[0.9935291,0.002576702,0.0008018898,0.001331183,0.00157328,0.0001877044],"domain_scores_gemma":[0.9888678,0.008538554,0.0004837128,0.0007989721,0.001146201,0.000164848],"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":[0.0001315607,0.0002591586,0.008975076,0.004818047,0.0006170789,0.0002799109,0.0007470395,0.0223622,0.001581075,0.0481176,0.02942097,0.8826904],"study_design_scores_gemma":[0.00009492281,0.0002607831,0.01068683,0.00261626,0.0006839056,0.00203104,0.001393651,0.4482298,0.004080579,0.2384188,0.2912305,0.0002729779],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"review","genre_scores_codex":[0.006272367,0.182315,0.7961861,0.004210368,0.0008489597,0.0004448869,0.002055285,0.001336062,0.006331056],"genre_scores_gemma":[0.1228998,0.3139892,0.53741,0.001826519,0.007445916,0.001764888,0.008256959,0.0006603642,0.005746305],"genre_candidate":"review","genre_consensus":null,"teacher_disagreement_score":0.009564827,"threshold_uncertainty_score":0.03789592,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2963386218","doi":"","title":"A Structured Self-Attentive Sentence Embedding.","year":2017,"lang":"en","type":"article","venue":"International Conference on Learning Representations","topic":"Topic Modeling","field":"Computer Science","cited_by":365,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Embedding; Sentence; Computer science; Natural language processing; Artificial intelligence; Logical consequence; Regularization (linguistics)","authors":[{"name":"Zhouhan Lin","is_ca":true},{"name":"Minwei Feng","is_ca":false},{"name":"Cícero Nogueira dos Santos","is_ca":false},{"name":"Mo Yu","is_ca":false},{"name":"Bing Xiang","is_ca":false},{"name":"Bowen Zhou","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.05458417324847135,"gpt":0.3661503131841141,"spread":0.3115661399356427,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003845097,0.0006732736,0.0002888856,0.0006852043,0.0001572627,0.0005433972,0.0007281756,0.0006773546,0.003148202],"category_scores_gemma":[0.002068233,0.0002517008,0.0006085125,0.0004781448,0.0002613376,0.001856121,0.0006495633,0.0009040961,0.001371495],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003150561,"about_ca_system_score_gemma":0.0003484404,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001136549,"about_ca_topic_score_gemma":0.002111012,"domain_scores_codex":[0.9997467,0.00008131277,0.00001530225,0.00009648712,0.00004354144,0.00001665401],"domain_scores_gemma":[0.9995179,0.0001960093,0.00006180477,0.00008805542,0.0001064485,0.00002973703],"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":[0.0003573731,0.0003777756,0.003369444,0.0006048552,0.0002235885,0.0003437334,0.0007884331,0.06386077,0.07697031,0.03943884,0.02919059,0.7844743],"study_design_scores_gemma":[0.00001504928,0.0001746299,0.001582196,0.00005433674,0.00006841644,0.0002339192,0.00007608756,0.9481597,0.01209261,0.02355309,0.01396075,0.00002914075],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0269405,0.0008884918,0.9651569,0.0005855957,0.0003866068,0.0001306158,0.0009334377,0.001771541,0.003206329],"genre_scores_gemma":[0.5320228,0.001098548,0.4454457,0.0006175471,0.0004020461,0.0002968044,0.003976705,0.0003463911,0.01579355],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003148202,"threshold_uncertainty_score":0.01053178,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3173783447","doi":"10.18653/v1/2021.acl-long.72","title":"DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":362,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University Health Network; Vector Institute; University of Toronto","funders":"National Institutes of Health; Compute Canada; Canadian Institute for Advanced Research","keywords":"Computer science; Natural language processing; Artificial intelligence; Linguistics; Computational linguistics; Volume (thermodynamics); Association (psychology); Joint (building); Deep learning; Cognitive science; Psychology; Philosophy; Epistemology; Engineering","authors":[{"name":"John Giorgi","is_ca":true},{"name":"Osvald Nitski","is_ca":true},{"name":"Bo Wang","is_ca":true},{"name":"Gary D. Bader","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03108129907966992,"gpt":0.2870895336430146,"spread":0.2560082345633448,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001263657,0.00132465,0.0009180878,0.001678834,0.0007188595,0.001844311,0.002768557,0.001331372,0.008955889],"category_scores_gemma":[0.004383939,0.0008025209,0.001160055,0.001590054,0.0006686,0.004901622,0.00247002,0.003273334,0.006484618],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001008362,"about_ca_system_score_gemma":0.0009345738,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005396095,"about_ca_topic_score_gemma":0.01242438,"domain_scores_codex":[0.999358,0.0002180189,0.00003455754,0.000222862,0.0001113503,0.00005517026],"domain_scores_gemma":[0.9985093,0.0008536704,0.00006700624,0.0003077256,0.0001947715,0.00006750428],"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":[0.0005697056,0.0003328612,0.001206654,0.0005123017,0.0002208141,0.000184618,0.0002644453,0.03556187,0.01261357,0.02635219,0.1382617,0.7839192],"study_design_scores_gemma":[0.0001211037,0.0001115062,0.0005258629,0.00007573316,0.00005584304,0.0001076329,0.00009220035,0.8845556,0.01114882,0.07388984,0.02927219,0.00004367623],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01498035,0.00281955,0.9251804,0.001094118,0.0005837438,0.0001568,0.006015741,0.04332814,0.005841149],"genre_scores_gemma":[0.2472118,0.001427538,0.6886694,0.001225134,0.0004918542,0.0006046725,0.03292605,0.004430823,0.02301267],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008955889,"threshold_uncertainty_score":0.02996045,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3176169354","doi":"10.18653/v1/2021.acl-long.551","title":"ARBERT & MARBERT: Deep Bidirectional Transformers for Arabic","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":352,"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":"Social Sciences and Humanities Research Council of Canada; Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Arabic; Transformer; Computer science; Natural language processing; Linguistics; Artificial intelligence; Speech recognition; Electrical engineering; Engineering; Philosophy; Voltage","authors":[{"name":"Muhammad Abdul-Mageed","is_ca":true},{"name":"AbdelRahim Elmadany","is_ca":true},{"name":"El Moatez Billah Nagoudi","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02856709349934125,"gpt":0.2625137650090602,"spread":0.233946671509719,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004288811,0.001277789,0.0005017632,0.001025157,0.0008145044,0.002055159,0.00127468,0.0008686268,0.04826551],"category_scores_gemma":[0.0015774,0.0007814677,0.000946492,0.000791222,0.0005474907,0.006130095,0.002037643,0.00175423,0.01833704],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007569564,"about_ca_system_score_gemma":0.0007547188,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005249097,"about_ca_topic_score_gemma":0.01340077,"domain_scores_codex":[0.9997951,0.00004894977,0.00001536807,0.00005947737,0.00005079709,0.00003029001],"domain_scores_gemma":[0.9996636,0.0001275425,0.00001589097,0.00008676078,0.00008072062,0.00002564035],"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":[0.0006389768,0.0001025436,0.0007182117,0.0004742463,0.00008044076,0.000461668,0.0005039183,0.01285183,0.01372084,0.09503891,0.2871666,0.5882418],"study_design_scores_gemma":[0.0002031984,0.0001132504,0.0005828761,0.0001980522,0.0001029239,0.0004963652,0.0004440203,0.3194001,0.05625177,0.2882971,0.3338153,0.00009497983],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01623176,0.002260812,0.8129079,0.001796991,0.000931973,0.00015036,0.007943547,0.1135448,0.04423179],"genre_scores_gemma":[0.3332583,0.001724971,0.5817866,0.000833377,0.0002358029,0.0002392809,0.01438654,0.01209677,0.05543839],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.04826551,"threshold_uncertainty_score":0.1614642,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1596986901","doi":"","title":"Joint learning of words and meaning representations for open-text semantic parsing","year":2012,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":333,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; Computer Research Institute of Montréal","funders":"","keywords":"Computer science; Natural language processing; WordNet; Artificial intelligence; Parsing; Task (project management); Meaning (existential); Context (archaeology); Representation (politics); Natural language; Process (computing); Programming language","authors":[{"name":"Antoine Bordes","is_ca":false},{"name":"Xavier Glorot","is_ca":true},{"name":"Jason Weston","is_ca":false},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.108195437040194,"gpt":0.337927349450232,"spread":0.229731912410038,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002173387,0.001052022,0.0008713416,0.001605643,0.0007688122,0.001702633,0.001917317,0.001648406,0.002636871],"category_scores_gemma":[0.006784994,0.0006849588,0.001261983,0.001926081,0.001133097,0.006895683,0.002863141,0.003002489,0.002066795],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008147288,"about_ca_system_score_gemma":0.001067974,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001889571,"about_ca_topic_score_gemma":0.003224568,"domain_scores_codex":[0.9987385,0.0005744887,0.00007018543,0.0003884111,0.0001448402,0.00008355999],"domain_scores_gemma":[0.9975241,0.001371765,0.000189382,0.0005873692,0.0002211015,0.0001062381],"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":[0.0003463695,0.0004576266,0.005027874,0.0003723939,0.000222023,0.0003970336,0.001303087,0.1235978,0.01861509,0.07737226,0.02237031,0.7499181],"study_design_scores_gemma":[0.00002049111,0.00004264633,0.0007089802,0.00002539612,0.00003582191,0.0001170316,0.0001553111,0.8727391,0.007275514,0.1139151,0.004932628,0.00003194221],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01593908,0.0002566811,0.9775109,0.0004276821,0.00004895855,0.00006118214,0.0003326886,0.004356187,0.001066689],"genre_scores_gemma":[0.3869755,0.0003903414,0.605617,0.0002508987,0.0001283386,0.0001881253,0.003200021,0.0004991972,0.002750504],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002636871,"threshold_uncertainty_score":0.01149404,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2542835211","doi":"10.48550/arxiv.1610.09038","title":"Professor Forcing: A New Algorithm for Training Recurrent Networks","year":2016,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":330,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Forcing (mathematics); Computer science; Treebank; Sampling (signal processing); MNIST database; Algorithm; Handwriting; Sample (material); Artificial intelligence; Sequence (biology); Machine learning; Speech recognition; Artificial neural network; Mathematics; Parsing","authors":[{"name":"Alex Lamb","is_ca":true},{"name":"Anirudh Goyal","is_ca":true},{"name":"Ying Zhang","is_ca":true},{"name":"Saizheng Zhang","is_ca":true},{"name":"Aaron Courville","is_ca":true},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.1343603456892092,"gpt":0.2081793152473785,"spread":0.07381896955816938,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002548011,0.001122257,0.0008595421,0.0009528279,0.000603682,0.0008398616,0.002524542,0.001703619,0.004368776],"category_scores_gemma":[0.01030359,0.0008306931,0.0009816127,0.0006644577,0.0009004282,0.001993308,0.001715747,0.002585957,0.001395922],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007792272,"about_ca_system_score_gemma":0.001077074,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003082427,"about_ca_topic_score_gemma":0.006692782,"domain_scores_codex":[0.9990917,0.0003201575,0.00006232283,0.000225765,0.0002163354,0.00008378628],"domain_scores_gemma":[0.9973463,0.001535648,0.0002214404,0.0004721756,0.0003367315,0.00008763255],"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":[0.0002270067,0.0001047225,0.001729453,0.0001004282,0.0001266499,0.0001594392,0.0002525342,0.606744,0.01030259,0.02838058,0.006562183,0.3453104],"study_design_scores_gemma":[0.00001672903,0.0000316588,0.00005794697,0.00000642583,0.000007404342,0.00002466037,0.000007224135,0.9910966,0.001604004,0.006121716,0.001019589,0.000006036152],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.006626271,0.00008350059,0.9901815,0.0001290693,0.00003406203,0.00004456113,0.00005473184,0.001892573,0.0009536844],"genre_scores_gemma":[0.2448703,0.0001514528,0.7476967,0.0002795838,0.0001111423,0.0004633223,0.0005427433,0.0008366616,0.005048158],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.004368776,"threshold_uncertainty_score":0.01461506,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1828724394","doi":"10.48550/arxiv.1410.2455","title":"BilBOWA: Fast Bilingual Distributed Representations without Word Alignments","year":2014,"lang":"en","type":"article","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":314,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal; Canadian Institute for Advanced Research","funders":"","keywords":"Computer science; Natural language processing; Artificial intelligence; Task (project management); Word (group theory); Bilingual dictionary; Sentence; Feature (linguistics); Set (abstract data type); Machine translation; Translation (biology); Speech recognition; Linguistics","authors":[{"name":"Stephan Gouws","is_ca":false},{"name":"Yoshua Bengio","is_ca":true},{"name":"Greg S. Corrado","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.0557204607827267,"gpt":0.2038580054545029,"spread":0.1481375446717763,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001423148,0.001579095,0.001329589,0.001414354,0.0006224874,0.001489254,0.002293299,0.001105908,0.006139134],"category_scores_gemma":[0.00438692,0.0006792836,0.00115888,0.001902235,0.0005303894,0.003994748,0.003399884,0.00231312,0.007450167],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006179924,"about_ca_system_score_gemma":0.001758224,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003571508,"about_ca_topic_score_gemma":0.00664028,"domain_scores_codex":[0.9991549,0.0002739545,0.00004692126,0.0002794465,0.0001474419,0.000097322],"domain_scores_gemma":[0.9991197,0.0002364443,0.00007184264,0.0003450045,0.0001507128,0.00007641689],"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":[0.000584359,0.0004473462,0.002567116,0.0002916734,0.0002427283,0.0001606861,0.0002838374,0.08080888,0.0144467,0.02349649,0.04621321,0.8304571],"study_design_scores_gemma":[0.00008670871,0.0001339271,0.0005172955,0.00003189188,0.00003804093,0.0001293184,0.00008422535,0.9384073,0.006106061,0.04155922,0.01286743,0.00003861929],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01702086,0.0005784828,0.9694914,0.0002666806,0.0001847987,0.0001025871,0.001469337,0.009184909,0.001700857],"genre_scores_gemma":[0.2696638,0.0007109785,0.6981762,0.0005450118,0.0002919687,0.0008935911,0.01769568,0.001955802,0.01006704],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006139134,"threshold_uncertainty_score":0.02053744,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3035038672","doi":"10.18653/v1/2020.acl-main.204","title":"DeeBERT: Dynamic Early Exiting for Accelerating BERT Inference","year":2020,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":301,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Waterloo; Vector Institute","funders":"Vector Institute; Natural Sciences and Engineering Research Council of Canada; Government of Canada; Canadian Institute for Advanced Research","keywords":"Inference; Computer science; Transformer; Language model; Redundancy (engineering); Artificial intelligence; Machine learning; Operating system; Engineering","authors":[{"name":"Ji Xin","is_ca":true},{"name":"Raphael Tang","is_ca":true},{"name":"Jaejun Lee","is_ca":true},{"name":"Yaoliang Yu","is_ca":true},{"name":"Jimmy Lin","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.07716608598601515,"gpt":0.2962549808226925,"spread":0.2190888948366773,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001873976,0.001999327,0.00124931,0.001392576,0.0008672692,0.001821755,0.003288496,0.001838304,0.01076242],"category_scores_gemma":[0.01125251,0.001242795,0.001288172,0.0009307718,0.0008216222,0.005520203,0.003142605,0.004623414,0.005447936],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001262262,"about_ca_system_score_gemma":0.002217198,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01194943,"about_ca_topic_score_gemma":0.02968272,"domain_scores_codex":[0.9990557,0.0002157725,0.00005726882,0.0002950476,0.0002426563,0.0001336565],"domain_scores_gemma":[0.9975017,0.001370215,0.0001181243,0.0005075997,0.0003532481,0.0001491134],"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":[0.001119746,0.0002759195,0.004740184,0.0004085226,0.0002362869,0.0003643305,0.0005149883,0.1959019,0.02042255,0.03581836,0.04730601,0.6928912],"study_design_scores_gemma":[0.00003995478,0.00003569089,0.0001953178,0.00001552397,0.00002196136,0.00005568617,0.0000319895,0.9761317,0.005982973,0.01294112,0.0045294,0.00001864353],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.009571702,0.0004315911,0.9644204,0.0003155717,0.0001688183,0.0001003873,0.000520601,0.02273075,0.001740136],"genre_scores_gemma":[0.3148634,0.0005258096,0.6637363,0.0007053777,0.0002250567,0.0003175901,0.004135612,0.004512577,0.01097833],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01194943,"threshold_uncertainty_score":0.03600389,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W3021792233","doi":"10.1145/3440755","title":"Evolution of Semantic Similarity—A Survey","year":2021,"lang":"en","type":"article","venue":"ACM Computing Surveys","topic":"Topic Modeling","field":"Computer Science","cited_by":281,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Lakehead University","funders":"","keywords":"Semantic similarity; Similarity (geometry); Semantic computing; Field (mathematics); Natural language; Open research; Semantics (computer science); Strengths and weaknesses; Natural language understanding","authors":[],"retraction":null,"screen_n_in":null,"score":{"opus":0.04743587048223354,"gpt":0.2804255950414988,"spread":0.2329897245592652,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004251526,0.0009379275,0.002044269,0.009730139,0.0008492635,0.005013178,0.002066736,0.001677228,0.003443049],"category_scores_gemma":[0.01840182,0.0008003798,0.00114108,0.01266281,0.001722562,0.01221627,0.002485662,0.001785815,0.001766172],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002027616,"about_ca_system_score_gemma":0.001764487,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002374003,"about_ca_topic_score_gemma":0.001347349,"domain_scores_codex":[0.9954259,0.000997416,0.0004939847,0.001147851,0.001765591,0.0001692958],"domain_scores_gemma":[0.989225,0.006573655,0.0006124779,0.001191091,0.002130132,0.0002676721],"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":[0.00008821595,0.0001266123,0.007892082,0.002075478,0.0001497737,0.00008897273,0.0005095289,0.005097257,0.0009662251,0.09305164,0.0101629,0.8797913],"study_design_scores_gemma":[0.00003334989,0.0003803133,0.01927441,0.002392288,0.0002888713,0.002635614,0.002151858,0.09958244,0.004810144,0.4270085,0.4412337,0.0002086417],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.03389379,0.6764765,0.2451459,0.006067893,0.001069599,0.0001921098,0.001122306,0.0009298247,0.03510216],"genre_scores_gemma":[0.2969955,0.5122386,0.1733301,0.001209006,0.003599612,0.0003092073,0.003185789,0.0005612049,0.008570908],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.009730139,"threshold_uncertainty_score":0.02248448,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2798416089","doi":"10.18653/v1/p18-1224","title":"Neural Natural Language Inference Models Enhanced with External Knowledge","year":2018,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":276,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Ottawa; Queen's University","funders":"","keywords":"Computer science; Inference; Leverage (statistics); Artificial intelligence; Artificial neural network; Natural language; Machine learning; Task (project management); Language model; Natural language processing","authors":[{"name":"Chen Qian","is_ca":false},{"name":"Xiaodan Zhu","is_ca":true},{"name":"Zhen-Hua Ling","is_ca":false},{"name":"Diana Inkpen","is_ca":true},{"name":"Si Wei","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02114323554206217,"gpt":0.2839945996636008,"spread":0.2628513641215386,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003058482,0.00123667,0.001093076,0.001244867,0.0006000409,0.001654775,0.002749116,0.00140637,0.003623985],"category_scores_gemma":[0.01168359,0.0005331033,0.001449886,0.001383403,0.0007375263,0.00563747,0.001595257,0.004252959,0.001896816],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001316927,"about_ca_system_score_gemma":0.001175407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01136504,"about_ca_topic_score_gemma":0.02094911,"domain_scores_codex":[0.9988484,0.0003805815,0.00006970411,0.0004661746,0.0001464876,0.00008865963],"domain_scores_gemma":[0.9943321,0.003983988,0.0002861684,0.0007111231,0.000567384,0.0001192272],"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":[0.0002985098,0.0004113036,0.004725055,0.0004009572,0.000321806,0.0002705841,0.0003647004,0.6348302,0.003143611,0.03346653,0.02020707,0.3015596],"study_design_scores_gemma":[0.000007623727,0.000009348512,0.0002101642,0.00001103884,0.00002040065,0.00001770875,0.000009387268,0.9872328,0.0004559403,0.01082104,0.001198419,0.000006094028],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04581743,0.002790848,0.9334731,0.00203104,0.0002171201,0.0001410145,0.003249643,0.005528044,0.006751845],"genre_scores_gemma":[0.6542633,0.001734517,0.3122284,0.0009993066,0.0005093563,0.0005682085,0.01787026,0.0005600259,0.01126673],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01136504,"threshold_uncertainty_score":0.02259773,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W1879966306","doi":"","title":"Long Short-Term Memory Over Recursive Structures","year":2015,"lang":"en","type":"article","venue":"NPARC","topic":"Topic Modeling","field":"Computer Science","cited_by":273,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":true,"about_ca":false},"ca_institutions":"University of Ottawa; National Research Council Canada","funders":"","keywords":"Computer science; Leverage (statistics); Artificial intelligence; Parsing; Theoretical computer science; Natural language processing","authors":[{"name":"Xiaodan Zhu","is_ca":true},{"name":"Parinaz Sobihani","is_ca":true},{"name":"Hongyu Guo","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.03825664726599463,"gpt":0.2786430462546798,"spread":0.2403863989886852,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003425917,0.0007202094,0.0004390269,0.0003954242,0.0002779401,0.0006850936,0.001100569,0.0008038584,0.003685961],"category_scores_gemma":[0.001819494,0.0002927791,0.0004237522,0.0007259872,0.000537596,0.003171249,0.0006705835,0.001055758,0.0009293194],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007314786,"about_ca_system_score_gemma":0.000626676,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00461016,"about_ca_topic_score_gemma":0.006161918,"domain_scores_codex":[0.9998429,0.00003837896,0.00001201304,0.00006259535,0.00002184835,0.00002229811],"domain_scores_gemma":[0.9995536,0.0002427983,0.0000525937,0.00006645988,0.00006413904,0.00002038757],"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":[0.0003908758,0.00008687101,0.001696897,0.000607393,0.0001310248,0.0004139446,0.0004215442,0.3729364,0.04380392,0.1294599,0.009376834,0.4406744],"study_design_scores_gemma":[0.0000108648,0.00005750349,0.0003521832,0.00002848585,0.00003197629,0.0000580734,0.00002555855,0.9413443,0.00597361,0.04972975,0.0023758,0.00001195893],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.07111851,0.002478018,0.9135227,0.0006922918,0.000144425,0.00004954445,0.0006518159,0.003265755,0.008077065],"genre_scores_gemma":[0.8760471,0.001998465,0.1136267,0.0002087939,0.0001236056,0.0001082826,0.001026798,0.0001791229,0.006681083],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00461016,"threshold_uncertainty_score":0.01233071,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2399880602","doi":"10.1609/aaai.v31i1.10983","title":"A Hierarchical Latent Variable Encoder-Decoder Model for Generating Dialogues","year":2017,"lang":"en","type":"preprint","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":266,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false},"ca_institutions":"HEC Montréal; McGill University; Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Canada Research Chairs; Samsung; Compute Canada; Samsung Advanced Institute of Technology; Canadian Institute for Advanced Research","keywords":"Computer science; Latent variable; Generative grammar; Generative model; Latent variable model; Artificial intelligence; Context (archaeology); Variable (mathematics); Artificial neural network; Process (computing); Encoder; Machine learning; Task (project management); Mathematics; Engineering","authors":[{"name":"Iulian Vlad Serban","is_ca":true},{"name":"Alessandro Sordoni","is_ca":false},{"name":"Ryan Lowe","is_ca":true},{"name":"Laurent Charlin","is_ca":true},{"name":"Joëlle Pineau","is_ca":true},{"name":"Aaron Courville","is_ca":true},{"name":"Yoshua Bengio","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.2165077802991398,"gpt":0.3321390255953345,"spread":0.1156312452961947,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001826564,0.000785258,0.0006308528,0.0006664997,0.0004003156,0.0008479111,0.00156732,0.001351961,0.004114923],"category_scores_gemma":[0.005816707,0.0006030776,0.0008672085,0.000783503,0.0006976786,0.001462466,0.00104987,0.002071222,0.001402666],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009869706,"about_ca_system_score_gemma":0.001193766,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004889777,"about_ca_topic_score_gemma":0.007904901,"domain_scores_codex":[0.9991054,0.0004548704,0.00003956445,0.00021829,0.0001127911,0.00006908795],"domain_scores_gemma":[0.997772,0.001676795,0.0001154056,0.000146173,0.0002117753,0.00007781691],"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":[0.000594786,0.0001941489,0.002140855,0.0002640819,0.0001391064,0.0003158968,0.0008445431,0.7428105,0.01004121,0.08645752,0.005839127,0.1503582],"study_design_scores_gemma":[0.00001445125,0.00001772841,0.00007847111,0.000005349767,0.000009764588,0.00002335599,0.000008004518,0.9899182,0.0004981742,0.008949878,0.0004702242,0.00000641198],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01573862,0.0002706356,0.9802473,0.0004614954,0.00006503199,0.00007346088,0.0003497094,0.001170079,0.001623702],"genre_scores_gemma":[0.625735,0.0003710089,0.3616284,0.0003208843,0.0001409859,0.0005460798,0.001253875,0.0003475849,0.009656114],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004889777,"threshold_uncertainty_score":0.01376581,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2160825952","doi":"10.1145/1008992.1009024","title":"Dependence language model for information retrieval","year":2004,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":261,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Language model; Term (time); Smoothing; Linkage (software); Divergence-from-randomness model; Artificial intelligence; Probabilistic logic; Independence (probability theory); Graph; Sentence; Term Discrimination; Axiom; Natural language processing; Theoretical computer science; Visual Word; Mathematics; Image retrieval","authors":[{"name":"Jianfeng Gao","is_ca":false},{"name":"Jian‐Yun Nie","is_ca":true},{"name":"Guangyuan Wu","is_ca":false},{"name":"Guihong Cao","is_ca":false}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02058730128944972,"gpt":0.2583534641804145,"spread":0.2377661628909648,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002306017,0.0008487065,0.001182874,0.002180255,0.0006049019,0.0015316,0.002291122,0.001322519,0.004899609],"category_scores_gemma":[0.008259196,0.0005337208,0.001565557,0.002124184,0.000837755,0.005835143,0.001087348,0.001996858,0.003029622],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001795441,"about_ca_system_score_gemma":0.001141232,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006065931,"about_ca_topic_score_gemma":0.003293014,"domain_scores_codex":[0.9980962,0.0007543518,0.0001251166,0.0003672727,0.0005280552,0.0001290545],"domain_scores_gemma":[0.9965218,0.002226819,0.0002242516,0.0004670895,0.0004818903,0.00007820795],"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":[0.0002374779,0.000187508,0.001813381,0.0003515487,0.0002047915,0.0003861456,0.0004654235,0.2084091,0.004276613,0.6257082,0.01896271,0.1389971],"study_design_scores_gemma":[0.00001756666,0.00003336337,0.0002483443,0.00001631712,0.00003776564,0.0001572768,0.00001917855,0.81051,0.000634023,0.1784032,0.009895104,0.00002784289],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005759574,0.0009676288,0.9856995,0.0006978651,0.00008254485,0.00007657433,0.0006543316,0.001043658,0.005018403],"genre_scores_gemma":[0.5493494,0.003569176,0.4092633,0.001407789,0.0007845023,0.001133261,0.004346891,0.0008364758,0.02930925],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006065931,"threshold_uncertainty_score":0.01639086,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2469060249","doi":"10.18653/v1/n16-1108","title":"Pairwise Word Interaction Modeling with Deep Neural Networks for Semantic Similarity Measurement","year":2016,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":249,"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 Science Foundation","keywords":"Computer science; Artificial intelligence; SemEval; Pairwise comparison; Similarity (geometry); Natural language processing; Semantic similarity; Word (group theory); Sentence; Semantics (computer science); Focus (optics); Artificial neural network; Selection (genetic algorithm); Mathematics","authors":[{"name":"Hua He","is_ca":false},{"name":"Jimmy Lin","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.06594859114948542,"gpt":0.254434726655701,"spread":0.1884861355062156,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001084474,0.001311028,0.0009068003,0.001176386,0.0004417759,0.0009593716,0.001936393,0.001179044,0.002591803],"category_scores_gemma":[0.003792297,0.000336842,0.0007605235,0.001634022,0.000473515,0.003506211,0.001568923,0.002259257,0.0009082591],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001066135,"about_ca_system_score_gemma":0.0007343958,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004366143,"about_ca_topic_score_gemma":0.006876721,"domain_scores_codex":[0.9991273,0.0002564343,0.00005369119,0.0002769439,0.0002025909,0.00008301179],"domain_scores_gemma":[0.9991449,0.0003997593,0.0001257799,0.0001212125,0.0001629202,0.00004543471],"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":[0.0005363186,0.0004491663,0.004463384,0.0003572618,0.0004131364,0.0002182715,0.0003972073,0.4341693,0.03605181,0.0351723,0.009083742,0.4786881],"study_design_scores_gemma":[0.000004175802,0.00001846977,0.0002387105,0.000003466236,0.00001217347,0.00001194931,0.00001180684,0.9865912,0.001556227,0.01120029,0.0003459932,0.000005574125],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05118299,0.0006651219,0.9432856,0.0003003496,0.00007980983,0.00006523413,0.0003916463,0.001896408,0.002132826],"genre_scores_gemma":[0.8077451,0.0003656186,0.1859069,0.0002513886,0.0001241777,0.0002502176,0.00148864,0.0002180204,0.003649857],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004366143,"threshold_uncertainty_score":0.008681417,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2151752770","doi":"10.1023/b:inrt.0000011209.19643.e2","title":"Augmenting Naive Bayes Classifiers with Statistical Language Models","year":2004,"lang":"en","type":"article","venue":"Information Retrieval","topic":"Topic Modeling","field":"Computer Science","cited_by":247,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Alberta","funders":"","keywords":"Naive Bayes classifier; Artificial intelligence; Computer science; Machine learning; Bayes error rate; Bayes classifier; Bayes' theorem; Classifier (UML); Conditional independence; Bayesian programming; Natural language processing; Pattern recognition (psychology); Bayes factor; Bayesian probability; Support vector machine","authors":[{"name":"Fuchun Peng","is_ca":false},{"name":"Dale Schuurmans","is_ca":true},{"name":"Shaojun Wang","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.01324613556970664,"gpt":0.2371975164125819,"spread":0.2239513808428753,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.007708231,0.001956006,0.002820781,0.004448228,0.001149722,0.00313319,0.002370568,0.00246437,0.004933987],"category_scores_gemma":[0.03147248,0.001034707,0.002005483,0.003635254,0.000610854,0.007949124,0.001710059,0.003128775,0.007152606],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008729438,"about_ca_system_score_gemma":0.001638349,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007244185,"about_ca_topic_score_gemma":0.01027938,"domain_scores_codex":[0.9940152,0.00296964,0.0004259123,0.00075319,0.001595023,0.0002410549],"domain_scores_gemma":[0.9773309,0.01723344,0.0004739081,0.001404635,0.003345979,0.0002110581],"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":[0.0005874375,0.0006005458,0.003049169,0.0005393695,0.0004737809,0.0001465533,0.0002047348,0.06320536,0.005727659,0.006089303,0.02235456,0.8970215],"study_design_scores_gemma":[0.00008858077,0.0001329543,0.000735759,0.00008684809,0.0003267741,0.0001773122,0.00007593527,0.9551525,0.003909549,0.03268756,0.006566804,0.00005937217],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02238299,0.004859613,0.9585693,0.001283084,0.0007891072,0.0002622874,0.0008154029,0.006360177,0.00467802],"genre_scores_gemma":[0.3074251,0.003535208,0.6714925,0.001190279,0.001743997,0.0005941053,0.004714453,0.0008093884,0.008495064],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.007708231,"threshold_uncertainty_score":0.04076552,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2982424689","doi":"10.1016/j.yjbinx.2019.100057","title":"A survey of word embeddings for clinical text","year":2019,"lang":"en","type":"review","venue":"Journal of Biomedical Informatics","topic":"Topic Modeling","field":"Computer Science","cited_by":240,"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; St. Michael's Hospital","funders":"","keywords":"Word (group theory); Computer science; Blueprint; Natural language processing; Artificial intelligence; Information retrieval; Data science; Linguistics","authors":[{"name":"Faiza Khan Khattak","is_ca":true},{"name":"Serena Jeblee","is_ca":true},{"name":"Chloé Pou-Prom","is_ca":true},{"name":"Mohamed Abdalla","is_ca":true},{"name":"Christopher Meaney","is_ca":true},{"name":"Frank Rudzicz","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.2610558891467246,"gpt":0.4618205473794558,"spread":0.2007646582327313,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002183706,0.001623285,0.001235335,0.004612558,0.0002426929,0.001571496,0.001178296,0.001001727,0.004363752],"category_scores_gemma":[0.009502555,0.0004849583,0.0009553081,0.004683227,0.0006978462,0.004003488,0.001165664,0.001344221,0.00346131],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006663167,"about_ca_system_score_gemma":0.001439395,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001381379,"about_ca_topic_score_gemma":0.001300752,"domain_scores_codex":[0.9990318,0.0003434995,0.0001510125,0.0002024821,0.0002371403,0.00003409075],"domain_scores_gemma":[0.9961141,0.002882621,0.0002325636,0.000178641,0.0005324529,0.00005961197],"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":[0.00003651737,0.00004447413,0.000471043,0.00599844,0.00006443582,0.00002334078,0.00008023765,0.0009381684,0.0007137039,0.005695798,0.01524314,0.9706906],"study_design_scores_gemma":[0.00005193171,0.0004418559,0.007413792,0.01603786,0.0004073126,0.001767762,0.0005956394,0.01583408,0.006046632,0.04248301,0.908711,0.0002091378],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"review","genre_gemma":"review","genre_scores_codex":[0.00321069,0.9350227,0.05090174,0.002678816,0.0009685528,0.0001655183,0.001313388,0.0005181994,0.005220424],"genre_scores_gemma":[0.02339877,0.9185818,0.04852227,0.0009238912,0.001102048,0.0003584776,0.003166955,0.0002126492,0.003733077],"genre_candidate":"review","genre_consensus":"review","teacher_disagreement_score":0.004612558,"threshold_uncertainty_score":0.01459819,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2000545282","doi":"10.1145/383952.384024","title":"Exploiting redundancy in question answering","year":2001,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":237,"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":"Question answering; Computer science; Redundancy (engineering); Information retrieval; Questions and answers; Artificial intelligence","authors":[{"name":"Charles L. A. Clarke","is_ca":true},{"name":"Gordon V. Cormack","is_ca":true},{"name":"Thomas R. Lynam","is_ca":true}],"retraction":null,"screen_n_in":null,"score":{"opus":0.02908871098396682,"gpt":0.2689517235673928,"spread":0.239863012583426,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.009978613,0.001621947,0.002615532,0.008480045,0.001832579,0.002521109,0.002767067,0.00243824,0.003470486],"category_scores_gemma":[0.05300068,0.001020398,0.001876485,0.005314915,0.001615333,0.007497381,0.004567026,0.001738486,0.001933559],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001005584,"about_ca_system_score_gemma":0.00119321,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002561537,"about_ca_topic_score_gemma":0.002223436,"domain_scores_codex":[0.9839057,0.009558165,0.0009483606,0.002059374,0.00277599,0.0007522675],"domain_scores_gemma":[0.9560141,0.03465769,0.001710215,0.004651231,0.002615417,0.0003513822],"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":[0.001248046,0.0005393752,0.01316969,0.002280975,0.0004538396,0.001075769,0.00551295,0.03103916,0.02550307,0.04166134,0.02705671,0.850459],"study_design_scores_gemma":[0.0002176735,0.0005838433,0.007786311,0.0003164012,0.0005968403,0.002421229,0.001598642,0.6524826,0.02818538,0.2604492,0.04516568,0.0001961822],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09009948,0.005122905,0.8883029,0.002420482,0.0002111034,0.0005610199,0.001427866,0.006452028,0.005402248],"genre_scores_gemma":[0.5268651,0.001614456,0.4608095,0.0008762655,0.0008160981,0.000638994,0.004951752,0.0005373049,0.00289042],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.009978613,"threshold_uncertainty_score":0.05277252,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}