{"id":"W4385573345","doi":"10.18653/v1/2022.nllp-1.30","title":"Computing and Exploiting Document Structure to Improve Unsupervised Extractive Summarization of Legal Case Decisions","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":7,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Pittsburgh; National Science Foundation","keywords":"Automatic summarization; Computer science; Argumentative; Exploit; Representation (politics); Legal document; Artificial intelligence; Information retrieval; Ranking (information retrieval); Graph; Legal case; Domain (mathematical analysis); Data mining; Machine learning; Natural language processing; Theoretical computer science; Mathematics","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001364468,0.001631331,0.001316177,0.01027875,0.0009032187,0.002078221,0.001208801,0.001026588,0.001635618],"category_scores_gemma":[0.007887711,0.0004144931,0.001073191,0.005772842,0.0003656655,0.003037907,0.0009257226,0.001533451,0.002211644],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008266282,"about_ca_system_score_gemma":0.001812809,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.009066301,"about_ca_topic_score_gemma":0.02376994,"domain_scores_codex":[0.9988663,0.000243942,0.0001390414,0.0003467006,0.0002949006,0.0001090782],"domain_scores_gemma":[0.9959522,0.001787235,0.0004643763,0.0005052388,0.001166419,0.000124549],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000355195,0.0003991668,0.007084771,0.0008498112,0.0003305081,0.000302008,0.0007696355,0.02534123,0.02549691,0.005240721,0.04914315,0.8846869],"study_design_scores_gemma":[0.0002245808,0.0006403534,0.01663423,0.000256563,0.0007931878,0.0005897277,0.001293331,0.8537208,0.03484369,0.03720768,0.05365552,0.0001403173],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1715607,0.008936523,0.7752083,0.001486227,0.0006809985,0.001006476,0.01605458,0.0173822,0.007684096],"genre_scores_gemma":[0.372961,0.002625762,0.5389184,0.0003098594,0.0009738937,0.0006498934,0.07438777,0.0008124714,0.008360996],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01027875,"threshold_uncertainty_score":0.01802707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0169936252046372,"score_gpt":0.2676518000611616,"score_spread":0.2506581748565244,"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."}}