{"id":"W6930689841","doi":"10.5281/zenodo.15350354","title":"Novel Table Search (NTS) Datasets","year":2025,"lang":"en","type":"dataset","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Stochastic processes and statistical mechanics","field":"Mathematics","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Table (database); Table (database); Benchmark (surveying); Benchmark (surveying); Column (typography); Column (typography); Ground truth","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001768639,0.003176008,0.001435175,0.00541912,0.001505291,0.003038215,0.004454468,0.002448649,0.054858],"category_scores_gemma":[0.01167835,0.0006567326,0.001830141,0.008085477,0.0006827944,0.003155356,0.00346435,0.001802427,0.06310349],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002215498,"about_ca_system_score_gemma":0.003534613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01450689,"about_ca_topic_score_gemma":0.03259862,"domain_scores_codex":[0.9967878,0.0005441793,0.0004597009,0.0008944503,0.001000305,0.0003136353],"domain_scores_gemma":[0.9962611,0.0008510576,0.0002560573,0.001321745,0.001025998,0.0002840098],"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":[0.0001681944,0.0001004594,0.0007486081,0.0008923374,0.0000476473,0.00006491456,0.00003366464,0.001330209,0.0005528572,0.001854809,0.9830858,0.01112059],"study_design_scores_gemma":[0.000661591,0.0001439593,0.003789294,0.0004024042,0.00006681239,0.0004606851,0.0003217168,0.0149382,0.004525905,0.009264256,0.965319,0.0001060798],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"dataset","genre_gemma":"dataset","genre_scores_codex":[0.002798481,0.0007032845,0.002450042,0.0002805179,0.0001686714,0.0002082698,0.9768648,0.01071314,0.005812806],"genre_scores_gemma":[0.001457091,0.0001073919,0.004102385,0.00007857221,0.00001213693,0.0002051941,0.9925019,0.0003589427,0.001176375],"genre_candidate":"dataset","genre_consensus":"dataset","teacher_disagreement_score":0.054858,"threshold_uncertainty_score":0.1835183,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07339077179870761,"score_gpt":0.3243007152849959,"score_spread":0.2509099434862883,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}