{"id":"W4412889754","doi":"10.18653/v1/2025.acl-long.1420","title":"DeepReview: Improving LLM-based Paper Review with Human-like Deep Thinking Process","year":2025,"lang":"en","type":"article","venue":"","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Natural Science Foundation of China; Westlake University; Institute for Catastrophic Loss Reduction","keywords":"Computer science; Process (computing); Artificial intelligence; Programming language","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":["metaresearch"],"consensus_categories":["metaresearch"],"category_scores_codex":[0.1970003,0.001707746,0.003021856,0.009393313,0.003422156,0.01430192,0.005689968,0.005370161,0.04013354],"category_scores_gemma":[0.6153604,0.001893701,0.003053279,0.005596266,0.003714288,0.0149167,0.01128718,0.004667441,0.02132256],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002864698,"about_ca_system_score_gemma":0.02180741,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008940294,"about_ca_topic_score_gemma":0.002656294,"domain_scores_codex":[0.7595385,0.1663351,0.01738883,0.01227827,0.04211924,0.002340022],"domain_scores_gemma":[0.2533861,0.462288,0.04376547,0.116559,0.1129465,0.01105485],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0009812415,0.0003096418,0.0051325,0.005934943,0.001032211,0.0002417001,0.002150344,0.002995309,0.006813959,0.03126882,0.2650339,0.6781055],"study_design_scores_gemma":[0.00192662,0.0012725,0.005527138,0.003819791,0.001833504,0.0013984,0.001585975,0.111955,0.02360387,0.2263122,0.6199705,0.0007945196],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01380173,0.00676438,0.8188359,0.07865366,0.01802949,0.004882367,0.002998945,0.03132013,0.0247134],"genre_scores_gemma":[0.1464345,0.002217613,0.7834271,0.0243495,0.01125125,0.003736649,0.003583453,0.004513122,0.02048682],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8029997,"threshold_uncertainty_score":0.9902416,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02684378825721569,"score_gpt":0.3791996728400586,"score_spread":0.3523558845828429,"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."}}