{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0009782433,0.001069332,0.001067424,0.0005415871,0.0005828331,0.0009490353,0.002472684,0.001401183,0.01294792],"category_scores_gemma":[0.002953649,0.000534351,0.0006628607,0.0002462587,0.0006262992,0.001369963,0.001733966,0.002014425,0.002947295],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006522664,"about_ca_system_score_gemma":0.001493411,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005824122,"about_ca_topic_score_gemma":0.01093504,"domain_scores_codex":[0.9995099,0.000113513,0.00002885487,0.0001163692,0.00016628,0.0000650154],"domain_scores_gemma":[0.9993355,0.0002409982,0.00004027603,0.0001512263,0.0001579672,0.00007403659],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0008586968,0.000780578,0.001771826,0.0004115986,0.000227608,0.000469561,0.0001507506,0.4214709,0.01424226,0.02871922,0.04090238,0.4899948],"study_design_scores_gemma":[0.00009166304,0.00009018386,0.0001022209,0.00001039941,0.00002622326,0.00004780876,0.00001305566,0.9824449,0.004062441,0.009352492,0.00374178,0.00001673934],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.02702159,0.0002731653,0.9362931,0.0005061759,0.000356893,0.0002829499,0.0004228528,0.02640564,0.008437721],"genre_scores_gemma":[0.4433547,0.000155213,0.5325643,0.0004390141,0.00007345252,0.0002783686,0.0008937398,0.001091619,0.02114955],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01294792,"threshold_uncertainty_score":0.04331511,"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."}}