{"id":"W4416058056","doi":"10.48550/arxiv.2510.19788","title":"AutumnBench Public Benchmark","year":2025,"lang":"","type":"preprint","venue":"ArXiv.org","topic":"Social Robot Interaction and HRI","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fonds de recherche du Québec – Nature et technologies","keywords":"Benchmarking; Suite; Maximization; Action (physics); Protocol (science); Reinforcement learning","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity","insufficient_payload"],"consensus_categories":["research_integrity","insufficient_payload"],"category_scores_codex":[0.0008980766,0.001140142,0.001402081,0.000822996,0.0008154475,0.0003607131,0.001905743,0.002345812,0.1209437],"category_scores_gemma":[0.0008951828,0.001333256,0.001254886,0.001053339,0.0004126672,0.0002914277,0.001589472,0.004305379,0.02071103],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006605958,"about_ca_system_score_gemma":0.001173033,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00176391,"about_ca_topic_score_gemma":0.0003619735,"domain_scores_codex":[0.9926624,0.001043186,0.001776886,0.0023779,0.0005950594,0.001544572],"domain_scores_gemma":[0.9943179,0.0008364608,0.001007147,0.002525383,0.0007139754,0.0005990598],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0001994186,0.001523442,0.731214,0.0002862945,0.002043609,0.00009344379,0.006532967,0.00002033182,0.000132455,0.03396697,0.1979448,0.02604228],"study_design_scores_gemma":[0.0008905025,0.00009996025,0.4161106,0.0002790972,0.0002242531,0.00001506814,0.001771238,0.00004772371,0.000064115,0.001421201,0.5782015,0.0008747287],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.286862,0.001369517,0.0005967728,0.01745516,0.04914812,0.001085985,0.0001012704,0.0003594166,0.6430218],"genre_scores_gemma":[0.7769262,0.0007903455,0.0001111173,0.005498536,0.0024533,0.0003616906,0.0002899951,0.00007337583,0.2134955],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.4900641,"threshold_uncertainty_score":0.9989493,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1379247616944918,"score_gpt":0.4053124748547807,"score_spread":0.2673877131602889,"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."}}