{"id":"W4308614207","doi":"10.48550/arxiv.2211.03229","title":"Computing and Exploiting Document Structure to Improve Unsupervised Extractive Summarization of Legal Case Decisions","year":2022,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Artificial Intelligence in Law","field":"Social Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Automatic summarization; Argumentative; Computer science; Exploit; Representation (politics); Legal document; Artificial intelligence; Ranking (information retrieval); Graph; Domain (mathematical analysis); Information retrieval; Legal case; Data mining; Machine learning; Theoretical computer science; Mathematics; Law; Political science","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005245422,0.0001997989,0.0002826836,0.0002206314,0.0009771767,0.0001122029,0.0004468717,0.0001943367,0.000367804],"category_scores_gemma":[0.0005311463,0.0002537661,0.0001000339,0.0005634813,0.0002761722,0.0003332615,0.001269228,0.0005164824,0.000003113586],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004498168,"about_ca_system_score_gemma":0.0003144479,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0133235,"about_ca_topic_score_gemma":0.005363382,"domain_scores_codex":[0.9981145,0.0003645628,0.0002998589,0.0007224079,0.0001872949,0.0003113781],"domain_scores_gemma":[0.9982092,0.000650886,0.0003172101,0.0003495068,0.0002666733,0.0002065585],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"qualitative","study_design_scores_codex":[0.0001509234,0.0001023311,0.005202257,0.00005456616,0.0001331978,0.001629086,0.03523897,0.429769,0.001118899,0.510488,0.00005119036,0.0160616],"study_design_scores_gemma":[0.001059681,0.0005986968,0.001837741,0.0007647371,0.0008704114,0.00006600485,0.5505255,0.1624579,0.007399723,0.2618149,0.009297183,0.003307493],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9556995,0.00001850406,0.0404744,0.00009043444,0.0006033655,0.0006329413,0.00009047139,0.00005904861,0.002331353],"genre_scores_gemma":[0.9984439,0.00005108292,0.0009678603,0.00004097183,0.0001210874,9.690066e-7,0.00001709484,0.00001762398,0.0003394209],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5152866,"threshold_uncertainty_score":0.9999915,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1010511661901489,"score_gpt":0.2823362853334743,"score_spread":0.1812851191433254,"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."}}