{"meta":{"query_hash":"29178c4a3bef","filters":{"venue":"ACM Transactions on Recommender Systems"},"cohort_total":8,"direct_labels_cover":0,"predictions_cover":8,"exported":8,"export_cap":100000,"truncated":false,"label_status":"direct model label, unvalidated","prediction_status":"machine_predicted_unvalidated (Codex and Gemma teacher distillation)","score_status":"score_only:v0-immature-baseline","snapshot":{"source":"OpenAlex, pinned release, all 482 partitions","release":"2026-06-24","frame_built":"2026-07-12"},"permalink":"https://metacan.xera.ac/q/29178c4a3bef","api":"https://metacan.xera.ac/api/v1/cohort?venue=ACM+Transactions+on+Recommender+Systems"},"results":[{"id":"W4366598314","doi":"10.1145/3591109","title":"A Framework and Toolkit for Testing the Correctness of Recommendation Algorithms","year":2023,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Test suite; Correctness; Unit testing; Python (programming language); Algorithm; Recommender system; Suite; White-box testing; Implementation; Surprise; Regression testing; Code coverage; Integration testing; Test case; Software engineering; Software; Programming language; Machine learning; Software system; Software construction","score_opus":0.12685383430496341,"score_gpt":0.3261905453913853,"score_spread":0.1993367110864219,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4366598314","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0004971026,0.0000616433,0.98746556,0.00738224,0.0025934617,0.0010552289,0.000038947754,0.00056330406,0.00034252985],"genre_scores_gemma":[0.880877,0.00013249821,0.11650639,0.00039505356,0.00016721597,0.0014964332,0.000016203952,0.0000512458,0.00035795098],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99819285,0.0002319572,0.00057942525,0.0004602483,0.00019771414,0.00033782236],"domain_scores_gemma":[0.99518543,0.0033640768,0.00026541093,0.00093909755,0.00017203247,0.00007397591],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0013947599,0.0002207525,0.00034301323,0.00026784957,0.00046420263,0.00022869969,0.00085217366,0.00014866864,0.0000056983235],"category_scores_gemma":[0.00016077393,0.0001670127,0.00010832835,0.001043169,0.000037862475,0.00032883813,0.00004442964,0.00026040443,0.000007694377],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000015011786,0.00011073167,0.00020841467,0.0002559251,0.00015281954,0.0000010606703,0.0017172468,0.00034605584,0.000110299705,0.0072917305,0.009190926,0.98059976],"study_design_scores_gemma":[0.002094925,0.0017686385,0.0018971068,0.0016020923,0.00013903872,0.00027255208,0.006687361,0.7357277,0.0048576607,0.04858931,0.19478038,0.0015831878],"about_ca_topic_score_codex":0.00021573866,"about_ca_topic_score_gemma":0.000010625232,"teacher_disagreement_score":0.9790166,"about_ca_system_score_codex":0.00005276499,"about_ca_system_score_gemma":0.00004071408,"threshold_uncertainty_score":0.681058},"labels":[],"label_agreement":null},{"id":"W4386136194","doi":"10.1145/3616379","title":"Multi-Interest Multi-Round Conversational Recommendation System with Fuzzy Feedback Based User Simulator","year":2023,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Simon Fraser University","funders":"National Natural Science Foundation of China","keywords":"Computer science; Preference; Recommender system; Human–computer interaction; Fuzzy logic; Information retrieval; Artificial intelligence","score_opus":0.08600942693861284,"score_gpt":0.2924780118068677,"score_spread":0.20646858486825487,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4386136194","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0011589163,0.000020751544,0.98662245,0.0041606748,0.0032718198,0.0014548197,0.00010802636,0.0028478657,0.00035469147],"genre_scores_gemma":[0.94772416,0.000014301203,0.049534112,0.0004579197,0.00010917143,0.00072022143,0.00011811624,0.000096887445,0.001225081],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9960372,0.0006077971,0.0009797616,0.0011441846,0.0005082118,0.00072285684],"domain_scores_gemma":[0.9964235,0.0007302842,0.00042961212,0.0017900554,0.0003198038,0.0003067837],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0010986928,0.0005975267,0.00063806307,0.0008439642,0.0006493181,0.00070036034,0.0014078848,0.00029851784,0.00006274712],"category_scores_gemma":[0.000023534167,0.0005211759,0.00024330571,0.0013314955,0.000054915257,0.0010514278,0.00005178977,0.0005035742,0.00049880886],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0018552542,0.015112443,0.021847166,0.0117744515,0.010504894,0.00059721916,0.013144797,0.27520794,0.005048112,0.053570174,0.28755358,0.30378395],"study_design_scores_gemma":[0.0051491,0.00052687764,0.0015235785,0.0007805319,0.00006163912,0.00009143843,0.0023670562,0.93303037,0.0013512216,0.000029914756,0.053892788,0.0011954916],"about_ca_topic_score_codex":0.0005658465,"about_ca_topic_score_gemma":0.0002664694,"teacher_disagreement_score":0.9465653,"about_ca_system_score_codex":0.0006158156,"about_ca_system_score_gemma":0.0001653612,"threshold_uncertainty_score":0.999724},"labels":[],"label_agreement":null},{"id":"W4391436413","doi":"10.1145/3643138","title":"Measuring Commonality in Recommendation of Cultural Content to Strengthen Cultural Citizenship","year":2024,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":true,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"McGill University","funders":"University College London; Canadian Institute for Advanced Research","keywords":"Citizenship; Content (measure theory); Cultural diversity; Sociology; Political science; Anthropology; Law","score_opus":0.2429439671267938,"score_gpt":0.3216424187380756,"score_spread":0.07869845161128178,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4391436413","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"empirical","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.03356733,0.00042356088,0.93923616,0.015889544,0.0045217825,0.0014694203,0.00011254568,0.000950415,0.0038292613],"genre_scores_gemma":[0.9926954,0.000058088684,0.0063726204,0.0002074374,0.000068462214,0.000316857,0.00002051527,0.000025975152,0.00023464217],"study_design_codex":"design_other","study_design_gemma":"not_applicable","domain_scores_codex":[0.99705184,0.0005276948,0.0010142756,0.0006599608,0.00032866438,0.0004175753],"domain_scores_gemma":[0.998178,0.00044609423,0.00013921685,0.0008921828,0.00017601196,0.00016848573],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0011984278,0.00033229758,0.00051964074,0.00050921144,0.00014263378,0.00045038125,0.0010300457,0.00015199555,0.00003201865],"category_scores_gemma":[0.00003954746,0.0002748459,0.00021834178,0.00086021976,0.000026254756,0.00066613377,0.000050314877,0.0004544679,0.000035504607],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00014419394,0.001005245,0.002165866,0.0018578595,0.0011618895,0.000039836843,0.024791427,0.0032439805,0.0029954293,0.04121766,0.04154966,0.87982696],"study_design_scores_gemma":[0.010648042,0.0054394933,0.016978415,0.022181252,0.00042744714,0.0014660589,0.06370506,0.2897325,0.08789403,0.01821143,0.47270915,0.010607146],"about_ca_topic_score_codex":0.0015048577,"about_ca_topic_score_gemma":0.00024603136,"teacher_disagreement_score":0.9591281,"about_ca_system_score_codex":0.0003920838,"about_ca_system_score_gemma":0.000048882823,"threshold_uncertainty_score":0.9999704},"labels":[],"label_agreement":null},{"id":"W4405194834","doi":"10.1145/3707693","title":"Modelling Concept Drift in Dynamic Data Streams for Recommender Systems","year":2024,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Data Stream Mining Techniques","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"University of British Columbia","funders":"","keywords":"Recommender system; STREAMS; Computer science; Concept drift; Data stream mining; Dynamic data; Data mining; Data science; Information retrieval; Database; Computer network","score_opus":0.10291693760084104,"score_gpt":0.33782853265750123,"score_spread":0.2349115950566602,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4405194834","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.00010955513,0.0015391873,0.98614824,0.0027804,0.0049473243,0.0013864107,0.0012611799,0.00148518,0.00034251012],"genre_scores_gemma":[0.88185316,0.00037666823,0.11494656,0.00016365468,0.0001296972,0.0010336287,0.00059933134,0.00011727367,0.0007800239],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9962245,0.00029672397,0.0009277218,0.0015053654,0.0003747741,0.00067090656],"domain_scores_gemma":[0.9942353,0.0012133537,0.00013825156,0.004180354,0.00007918914,0.0001535789],"candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0012391916,0.0004506963,0.0005494932,0.00065261964,0.00022309858,0.0011017971,0.0039600856,0.0002653505,0.000012002879],"category_scores_gemma":[0.000027989934,0.0004405583,0.00015934753,0.000778743,0.000042259788,0.0017166085,0.00013683112,0.00058804924,0.000037820882],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000053307765,0.0007984843,0.00002067227,0.0013594056,0.0009005929,0.000058340116,0.0036369606,0.36752746,0.000045281522,0.015738618,0.10141645,0.5084444],"study_design_scores_gemma":[0.00037352738,0.00014475425,5.596057e-7,0.00079471915,0.000039582374,0.00004902617,0.0006721936,0.92403895,0.00005506324,0.0007760401,0.07261916,0.00043642192],"about_ca_topic_score_codex":0.0014034216,"about_ca_topic_score_gemma":0.00013013443,"teacher_disagreement_score":0.8817436,"about_ca_system_score_codex":0.00038894298,"about_ca_system_score_gemma":0.00015832318,"threshold_uncertainty_score":0.99993515},"labels":[],"label_agreement":null},{"id":"W4407654895","doi":"10.1145/3718488","title":"Balancing Embedding Spectrum for Recommendation","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Embedding; Spectrum (functional analysis); Artificial intelligence; Physics","score_opus":0.02706613366870131,"score_gpt":0.30838186907996673,"score_spread":0.2813157354112654,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4407654895","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.00006529279,0.000090187816,0.96835345,0.018029615,0.005733314,0.001162201,0.000022572527,0.0008668066,0.0056765852],"genre_scores_gemma":[0.91635823,0.00008603506,0.077712156,0.0015101355,0.00015981961,0.0013446205,0.000023738543,0.000043180913,0.0027620983],"study_design_codex":"design_other","study_design_gemma":"not_applicable","domain_scores_codex":[0.99768454,0.0002091873,0.00071091694,0.0007156517,0.00017351317,0.00050618604],"domain_scores_gemma":[0.99764276,0.0006795718,0.00020141981,0.0012765222,0.00010016876,0.00009957986],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0009528586,0.00031828077,0.00043504525,0.00059180387,0.0005821033,0.00048099432,0.0011531934,0.000183821,0.000027676557],"category_scores_gemma":[0.000035446108,0.00031068435,0.00023275122,0.0006482645,0.000014957868,0.0006123244,0.000035631547,0.00029068146,0.000021095955],"study_design_candidate":"not_applicable","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.00007000907,0.0006583611,0.00026873883,0.00074525404,0.000749342,0.0000036330146,0.0010257798,0.0019426001,0.00040449298,0.10231221,0.12396514,0.76785445],"study_design_scores_gemma":[0.0021055876,0.0004617247,0.000116988085,0.00074844644,0.000073831776,0.000053515545,0.00095790176,0.22913346,0.010548188,0.023421915,0.731346,0.0010324155],"about_ca_topic_score_codex":0.00018728281,"about_ca_topic_score_gemma":0.00004218187,"teacher_disagreement_score":0.9162929,"about_ca_system_score_codex":0.0003377813,"about_ca_system_score_gemma":0.0000844943,"threshold_uncertainty_score":0.99993455},"labels":[],"label_agreement":null},{"id":"W4409168760","doi":"10.1145/3728373","title":"Recall, Robustness, and Lexicographic Evaluation","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Topic Modeling","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Microsoft (Canada)","funders":"","keywords":"Lexicographical order; Robustness (evolution); Recall; Computer science; Artificial intelligence; Cognitive psychology; Mathematics; Psychology; Chemistry; Combinatorics","score_opus":0.05255388057789485,"score_gpt":0.2979719475362509,"score_spread":0.24541806695835605,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4409168760","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0028481542,0.00054354855,0.9857339,0.006949252,0.0020312583,0.00042933415,0.0000019069095,0.00019666842,0.0012659614],"genre_scores_gemma":[0.98180306,0.000116289004,0.017025106,0.00036537668,0.000027567932,0.00020779941,0.0000019828217,0.00000802161,0.00044479402],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.99856746,0.00023845551,0.0002930857,0.00044948998,0.00025402594,0.000197485],"domain_scores_gemma":[0.9986079,0.00018907915,0.000057365258,0.0009819386,0.00010612751,0.00005761541],"candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007612761,0.0001408424,0.00016920247,0.00033209627,0.00025464135,0.00021035054,0.00055313733,0.00010347725,0.000010985371],"category_scores_gemma":[0.000024376604,0.00013673333,0.000055846005,0.0005191104,0.000020604724,0.0003479113,0.000019470957,0.00020089927,0.0000052783466],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0000102113745,0.00015286343,0.00024654096,0.00011817156,0.0001578036,0.0000013639491,0.00036365603,0.07333764,0.000078356505,0.0129876705,0.0013390189,0.9112067],"study_design_scores_gemma":[0.00068932294,0.00004358389,0.00032396917,0.00014113678,0.000046126876,0.0000139650865,0.00021337875,0.98759556,0.00017424102,0.0036144082,0.0069367653,0.00020756446],"about_ca_topic_score_codex":0.00009074969,"about_ca_topic_score_gemma":0.000029143903,"teacher_disagreement_score":0.9789549,"about_ca_system_score_codex":0.000079191814,"about_ca_system_score_gemma":0.0000728432,"threshold_uncertainty_score":0.5575824},"labels":[],"label_agreement":null},{"id":"W4412841400","doi":"10.1145/3757327","title":"A Comparative Survey Of Algorithmic Feed Recommendation System Designs","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"McGill University","funders":"","keywords":"Computer science; Information retrieval; Data science","score_opus":0.10218141728614505,"score_gpt":0.32943248951364307,"score_spread":0.22725107222749802,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4412841400","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0004212749,0.00015139976,0.9875606,0.0011411614,0.0031878017,0.0012404801,0.000091677495,0.00064985163,0.005555781],"genre_scores_gemma":[0.97900236,0.000038039372,0.019813465,0.00012946167,0.00002452832,0.00044364028,0.00003521898,0.000021756356,0.000491533],"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","domain_scores_codex":[0.9957732,0.00152949,0.0012236756,0.0007371948,0.00030983213,0.00042658998],"domain_scores_gemma":[0.9961624,0.001139982,0.00047788417,0.0016473503,0.00045349274,0.00011886812],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0019276998,0.0004047931,0.0009398984,0.0007932249,0.00035492054,0.00024049467,0.0015689032,0.00023273224,0.000017416556],"category_scores_gemma":[0.000028321218,0.00038095447,0.00019393573,0.0015635913,0.000045418463,0.0005242018,0.00005143546,0.00037486528,0.000029326002],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.0005902902,0.0052382746,0.0044572433,0.0046513816,0.006523965,0.000023656145,0.009305003,0.004261125,0.0026045192,0.0945284,0.14439382,0.72342235],"study_design_scores_gemma":[0.012476227,0.003752335,0.028527733,0.0077583375,0.0005508575,0.0003745423,0.014599362,0.7268342,0.09482791,0.0034260375,0.101129346,0.005743132],"about_ca_topic_score_codex":0.0038843546,"about_ca_topic_score_gemma":0.00026307398,"teacher_disagreement_score":0.9785811,"about_ca_system_score_codex":0.00041436337,"about_ca_system_score_gemma":0.00018393483,"threshold_uncertainty_score":0.9998642},"labels":[],"label_agreement":null},{"id":"W4413167298","doi":"10.1145/3760763","title":"Revisiting LightGCN: Unexpected Inflexibility, Inconsistency, and A Remedy Towards Improved Recommendation","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Recommender Systems","topic":"Recommender Systems and Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"route_ca_aff":true,"route_ca_fund":false,"route_ca_venue":false,"route_about_ca":false,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Computer science; Psychology","score_opus":0.028871919565577144,"score_gpt":0.2931055492405019,"score_spread":0.26423362967492475,"validation_status":"score_only:v0-immature-baseline","prediction":{"id":"W4413167298","genre_codex":"methods","genre_gemma":"empirical","domain_codex":null,"domain_gemma":null,"model_version":"codex-gemma-dda1882f352a","genre_candidate":"methods","genre_consensus":null,"domain_candidate":null,"domain_consensus":null,"prediction_status":"machine_predicted_unvalidated","genre_scores_codex":[0.0011995852,0.0003791422,0.95888674,0.024087984,0.0023574643,0.0011810104,0.000017320583,0.001109722,0.010781021],"genre_scores_gemma":[0.966481,0.00026961055,0.030469505,0.0011254064,0.00009878037,0.0005154604,0.000015547485,0.000032681077,0.0009920419],"study_design_codex":"design_other","study_design_gemma":"not_applicable","domain_scores_codex":[0.99646413,0.0006513942,0.0011180771,0.0010359939,0.00023653141,0.0004938524],"domain_scores_gemma":[0.99698925,0.0004981393,0.0003193663,0.0017617451,0.00025490724,0.0001765778],"candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0014553635,0.00043518812,0.0006284509,0.0006656381,0.00060045667,0.0006564917,0.001042878,0.00029375972,0.000026381187],"category_scores_gemma":[0.00011852013,0.00040789813,0.0001783842,0.0010685949,0.000050173006,0.0008003274,0.00012868627,0.0005518218,0.000010376159],"study_design_candidate":"design_other","study_design_consensus":null,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_system_candidate":false,"about_ca_system_consensus":false,"study_design_scores_codex":[0.000034554083,0.00023379947,0.0003925878,0.00045488426,0.00032174992,0.0000049134474,0.000848035,0.000015133852,0.00079234195,0.008752807,0.0033359865,0.9848132],"study_design_scores_gemma":[0.008886474,0.0018451044,0.0068724765,0.005483892,0.0004428178,0.00073501363,0.0067690704,0.18575387,0.026761387,0.035283636,0.7159429,0.0052233534],"about_ca_topic_score_codex":0.000982259,"about_ca_topic_score_gemma":0.00009390263,"teacher_disagreement_score":0.9795899,"about_ca_system_score_codex":0.0003334691,"about_ca_system_score_gemma":0.00016266029,"threshold_uncertainty_score":0.9998373},"labels":[],"label_agreement":null}]}