{"id":"W4385570153","doi":"10.18653/v1/2023.acl-industry.3","title":"Hunt for Buried Treasures: Extracting Unclaimed Embodiments from Patent Specifications","year":2023,"lang":"en","type":"article","venue":"","topic":"Machine Learning in Materials Science","field":"Materials Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Institute for Catastrophic Loss Reduction","keywords":"Task (project management); Computer science; Competitor analysis; Automation; Intellectual property; Inference; Patent visualisation; Transformer; Artificial intelligence; Natural language processing; Software engineering; Engineering drawing; Data science; Engineering; Business; Systems engineering","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":[0.001641069,0.001528472,0.0004982803,0.003720293,0.001024474,0.002042159,0.001558188,0.00214498,0.006687509],"category_scores_gemma":[0.009790031,0.0005797896,0.00145783,0.0020272,0.000800286,0.005271658,0.002095009,0.001618151,0.004953901],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001702586,"about_ca_system_score_gemma":0.002690766,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01381394,"about_ca_topic_score_gemma":0.03181194,"domain_scores_codex":[0.9982503,0.0003446792,0.0002742494,0.0005067461,0.0004812088,0.0001426762],"domain_scores_gemma":[0.9930448,0.004009157,0.0007025984,0.001223155,0.0008590255,0.0001613122],"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":[0.001112599,0.0006189509,0.02736532,0.005085926,0.0002508617,0.004883692,0.002561053,0.01857418,0.06477515,0.05730089,0.259476,0.5579954],"study_design_scores_gemma":[0.0003872955,0.0004095109,0.02335421,0.0006074058,0.0002939571,0.004808687,0.003890604,0.2758473,0.1568758,0.0582941,0.4748696,0.00036161],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2901987,0.004683544,0.3452494,0.003641419,0.00052131,0.0008648471,0.2763772,0.05058243,0.0278812],"genre_scores_gemma":[0.290039,0.001143411,0.2910405,0.0006119612,0.00008107703,0.0005049077,0.407099,0.001462644,0.008017522],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01381394,"threshold_uncertainty_score":0.02746707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1958081442768627,"score_gpt":0.3208010541663181,"score_spread":0.1249929098894554,"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."}}