{"meta":{"page":1,"per_page":50,"max_per_page":100,"total":5,"total_is_capped":false,"direct_labels_cover":1,"predictions_cover":5,"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"},"query_hash":"d9ff4133a46a","filters":{"venue":"Conference on Innovative Data Systems Research"}},"results":[{"id":"W3146147323","doi":"","title":"Lakehouse: A New Generation of Open Platforms that Unify Data Warehousing and Advanced Analytics.","year":2021,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":123,"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; Analytics; Data warehouse; Data science; Database","retraction":null,"screen_n_in":null,"score":{"opus":0.946872203848267,"gpt":0.6271471361658597,"spread":0.3197250676824073,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","scholarly_communication","open_science"],"consensus_categories":["open_science"],"category_scores_codex":[0.01920787,0.0001846625,0.0005634697,0.0005990985,0.0003330643,0.003047313,0.007935701,0.00009173081,0.000237139],"category_scores_gemma":[0.008996663,0.0001400913,0.00001142573,0.004048625,0.0003438755,0.004107798,0.01728054,0.0004329756,0.00007979927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000617585,"about_ca_system_score_gemma":0.001479583,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00130949,"about_ca_topic_score_gemma":0.001838082,"domain_scores_codex":[0.9932848,0.0007882873,0.001013928,0.001601335,0.002890029,0.0004215709],"domain_scores_gemma":[0.9875896,0.001291829,0.0004681597,0.007103241,0.003382313,0.000164864],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001245676,0.0001745498,0.001017623,0.0001318543,0.0001444601,0.00005834777,0.0007685958,0.00007274096,0.003483747,0.4615848,0.3228096,0.2096291],"study_design_scores_gemma":[0.003688337,0.0005489678,0.003738967,0.001484644,0.00003481059,0.00003717185,0.06252182,0.1739665,0.006345469,0.0286581,0.7179775,0.0009976816],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.249746,0.002770397,0.5759263,0.02307307,0.003440607,0.009840389,0.06722741,0.0001945618,0.06778129],"genre_scores_gemma":[0.9801367,0.0003758504,0.002712554,0.0001317854,0.0001367286,0.00001900054,0.01026055,0.00002079907,0.006206062],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7303907,"threshold_uncertainty_score":0.999351,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2585438896","doi":"","title":"The data civilizer system","year":2017,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":88,"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","retraction":null,"screen_n_in":null,"score":{"opus":0.6414419922836844,"gpt":0.5044311967031525,"spread":0.1370107955805319,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["sts","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":["open_science"],"category_scores_codex":[0.01486543,0.0002244579,0.0003606796,0.0002904342,0.004139095,0.009152149,0.04588917,0.00009965778,0.000006735707],"category_scores_gemma":[0.003496828,0.0001421178,0.00001844446,0.001222321,0.0007843953,0.002828222,0.02105016,0.0008378897,0.0008002319],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009898134,"about_ca_system_score_gemma":0.0006784677,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001244886,"about_ca_topic_score_gemma":0.0001892868,"domain_scores_codex":[0.9941769,0.001056758,0.0005586725,0.001625257,0.001777166,0.0008053111],"domain_scores_gemma":[0.9653193,0.0009541845,0.0004183412,0.03077381,0.002391115,0.000143205],"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":[0.00001211329,0.00003257725,0.000392178,0.00006721194,0.0001124209,0.00004364416,0.00008714563,8.424748e-7,0.0001534632,0.7966109,0.1535977,0.0488898],"study_design_scores_gemma":[0.0004722072,0.0001255221,0.00189211,0.0008289244,0.000007493811,0.00003289042,0.003070116,0.7378812,0.00008442844,0.0004526723,0.2547271,0.0004253858],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005776752,0.001722985,0.570177,0.03664406,0.00785681,0.003628636,0.01920058,0.001038225,0.3539549],"genre_scores_gemma":[0.9952725,0.00009849805,0.0005788262,0.0000165443,0.0003225112,0.00006067682,0.001108808,0.00001574625,0.002525951],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9894957,"threshold_uncertainty_score":0.9999778,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2583699183","doi":"","title":"Data Ingestion for the Connected World.","year":2017,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Data Quality and Management","field":"Decision Sciences","cited_by":68,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Ingestion; Medicine","retraction":null,"screen_n_in":null,"score":{"opus":0.9484017230905639,"gpt":0.6713623530691826,"spread":0.2770393700213813,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","sts","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":["metaresearch","open_science"],"category_scores_codex":[0.06057546,0.0001804614,0.0003630426,0.0006285084,0.002669919,0.007006608,0.02731622,0.00007358223,0.0002100638],"category_scores_gemma":[0.07575348,0.0001063969,0.00001733666,0.001698861,0.001072163,0.002467089,0.01232761,0.000597105,0.0007861222],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000614393,"about_ca_system_score_gemma":0.0004424358,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001090417,"about_ca_topic_score_gemma":0.002129748,"domain_scores_codex":[0.9925039,0.001398064,0.0008736888,0.001551561,0.003095837,0.000576882],"domain_scores_gemma":[0.9432418,0.01550063,0.001044693,0.03154089,0.008521804,0.0001501352],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001051304,0.00003857412,0.0001399909,0.00002164601,0.00004079803,0.000002532953,0.00006615131,0.000001733041,0.00003463729,0.4347189,0.4871992,0.07763071],"study_design_scores_gemma":[0.0004705325,0.00009480969,0.004021222,0.0001702697,0.000006161026,0.000001234926,0.003682896,0.06365814,0.00003278553,0.01180344,0.9158978,0.000160641],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.005522109,0.0006263262,0.4533053,0.1717604,0.01007029,0.01648564,0.1798762,0.0002678325,0.1620859],"genre_scores_gemma":[0.9747884,0.00006685773,0.0002893289,0.0001664004,0.0004852634,0.0002847146,0.00710516,0.00001804651,0.01679582],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9692663,"threshold_uncertainty_score":0.9999919,"prediction_status":"machine_predicted_unvalidated"},"labels":[{"model":"gemma","categories":[],"domain":null,"study_design":"bench_or_experimental","genre":"empirical","about_ca_system":false,"about_ca_topic":false,"confidence":"low"},{"model":"gpt","categories":[],"domain":null,"study_design":"not_applicable","genre":"software","about_ca_system":false,"about_ca_topic":false,"confidence":"high"}],"label_agreement":"split"},{"id":"W2397606137","doi":"","title":"A Case for Thread Migration.","year":2013,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Mobile Agent-Based Network Management","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Thread (computing); Programming language","retraction":null,"screen_n_in":null,"score":{"opus":0.4392051732051863,"gpt":0.4339388316145014,"spread":0.005266341590684898,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.003486003,0.0002082585,0.0002588724,0.0005307274,0.0004162701,0.001549727,0.003613408,0.00008011074,0.00008891032],"category_scores_gemma":[0.0003992601,0.0001740592,0.00002089772,0.002469871,0.0001638894,0.001361854,0.001819389,0.0003770509,0.0008177004],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001866013,"about_ca_system_score_gemma":0.0004054644,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001517469,"about_ca_topic_score_gemma":0.000120416,"domain_scores_codex":[0.9962369,0.0005352336,0.000505607,0.001098609,0.0008939917,0.000729704],"domain_scores_gemma":[0.9919248,0.0007579608,0.0001575168,0.003488931,0.003534951,0.00013589],"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":[0.000007732569,0.00006610838,0.00004552213,0.00009449619,0.00003926949,0.000125253,0.0001672249,0.00003374083,0.0001711475,0.6381583,0.3251089,0.03598227],"study_design_scores_gemma":[0.0007070016,0.0005666235,0.00009193464,0.0002220537,0.000003027563,0.0001443139,0.001174591,0.8974717,0.0002554068,0.007750667,0.09121227,0.0004003731],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.007533293,0.00009632522,0.9750239,0.004087341,0.0006084349,0.005690206,0.0003550117,0.0001629454,0.006442506],"genre_scores_gemma":[0.985298,0.00002818086,0.007691807,0.0001712175,0.00024487,0.003743168,0.0005239033,0.00002242389,0.002276385],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9777647,"threshold_uncertainty_score":0.9999603,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null},{"id":"W2963114423","doi":"","title":"Reprowd: Crowdsourced Data Processing Made Reproducible.","year":2016,"lang":"en","type":"article","venue":"Conference on Innovative Data Systems Research","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false},"ca_institutions":"Simon Fraser University","funders":"","keywords":"Crowdsourcing; Computer science; Data science; Multidisciplinary approach; Open research; World Wide Web","retraction":null,"screen_n_in":null,"score":{"opus":0.8778779383011985,"gpt":0.5952426953214541,"spread":0.2826352429797444,"validation_status":"score_only:v0-immature-baseline"},"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaresearch","scholarly_communication","open_science","insufficient_payload"],"consensus_categories":["metaresearch","open_science"],"category_scores_codex":[0.09826914,0.0002998916,0.0005386921,0.001702592,0.0008842592,0.004136809,0.02002726,0.0001184383,0.0004858032],"category_scores_gemma":[0.06414337,0.000171462,0.000022056,0.009011733,0.0009407128,0.002580502,0.01753449,0.0005947058,0.003681949],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001398682,"about_ca_system_score_gemma":0.0009407757,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002554621,"about_ca_topic_score_gemma":0.00002427071,"domain_scores_codex":[0.9818556,0.002127251,0.001684147,0.006587493,0.006640322,0.001105175],"domain_scores_gemma":[0.9232846,0.005346508,0.001127678,0.05764461,0.01217353,0.0004230416],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.00007120815,0.00009218704,0.0008723363,0.00003683477,0.00002360794,0.00002461493,0.0001944935,0.00000329727,0.003624493,0.01956864,0.5757372,0.3997511],"study_design_scores_gemma":[0.001009893,0.0002263665,0.003800836,0.001580253,0.000007150498,0.00003107273,0.00665107,0.05412332,0.0008903958,0.008143147,0.922855,0.0006814513],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1094728,0.001306059,0.5122795,0.07458124,0.01091826,0.008692658,0.0277795,0.001340132,0.2536298],"genre_scores_gemma":[0.9558611,0.00002261785,0.0007744231,0.00006555922,0.0005513675,0.00005378053,0.001119878,0.00003125565,0.04152004],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8463882,"threshold_uncertainty_score":0.9970938,"prediction_status":"machine_predicted_unvalidated"},"labels":[],"label_agreement":null}]}