{"id":"W4416036933","doi":"10.18653/v1/2025.emnlp-main.125","title":"REARANK: Reasoning Re-ranking Agent via Reinforcement Learning","year":2025,"lang":"","type":"article","venue":"","topic":"Reinforcement Learning in Robotics","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada; Canadian Institute for Advanced Research; Nvidia","keywords":"Reinforcement learning; Control (management); Action (physics); Feature (linguistics); Stability (learning theory)","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","sts","scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.002903386,0.0009929154,0.0009494431,0.0009357939,0.001965824,0.001904074,0.002804526,0.0004755907,0.001575398],"category_scores_gemma":[0.0008876291,0.001065798,0.0005102468,0.002749129,0.0002616969,0.001329378,0.002918924,0.002056936,0.001002499],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009078495,"about_ca_system_score_gemma":0.0006532407,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003703359,"about_ca_topic_score_gemma":0.00001234322,"domain_scores_codex":[0.9915388,0.0006086182,0.002170598,0.001816226,0.001794324,0.00207141],"domain_scores_gemma":[0.9953877,0.0004838587,0.0009159076,0.002307084,0.0005247178,0.0003807756],"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":[0.00004883666,0.00004221054,0.00179215,0.0002219252,0.0003162336,0.000058966,0.002101426,0.8846804,0.0003136713,0.05822869,0.002008647,0.05018676],"study_design_scores_gemma":[0.001262664,0.0004340089,0.0006558656,0.001611205,0.0001546902,0.00001401677,0.0003302957,0.9345927,0.001372389,0.000271043,0.05835504,0.0009461387],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0002961929,0.0006530699,0.7269705,0.001367379,0.00299023,0.0007449908,1.238536e-7,0.0005897867,0.2663877],"genre_scores_gemma":[0.7931597,0.0005812119,0.03138192,0.001847079,0.0002145718,0.00003844031,0.00001347033,0.00005864602,0.1727049],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7928635,"threshold_uncertainty_score":0.9997754,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01683852968899304,"score_gpt":0.2687740419002488,"score_spread":0.2519355122112558,"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."}}