{"id":"W7138891804","doi":"10.1109/adacis65663.2025.11436933","title":"HOL4PSG: AI-Driven Proof Sequence Generation for the HOL4 Theorem Prover","year":2025,"lang":"","type":"article","venue":"","topic":"Logic, programming, and type systems","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Concordia University","funders":"","keywords":"Mathematical proof; Automated theorem proving; Proof assistant; Proof complexity; Gas meter prover; Computer-assisted proof; Formal proof; Proof theory","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002824956,0.001292288,0.0005041509,0.001424539,0.0005349172,0.001248889,0.002586851,0.001075763,0.02112819],"category_scores_gemma":[0.01229414,0.0008017779,0.001156025,0.0006537598,0.001042492,0.001950874,0.002726552,0.002214502,0.007927446],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008319113,"about_ca_system_score_gemma":0.002379428,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001937597,"about_ca_topic_score_gemma":0.004127924,"domain_scores_codex":[0.9983998,0.0005389852,0.0001088111,0.0002994378,0.0005353426,0.0001176041],"domain_scores_gemma":[0.9945166,0.003553964,0.0002235853,0.001013221,0.0005380433,0.0001546012],"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":[0.0008519714,0.0004832582,0.003961885,0.002383868,0.0001884731,0.0007406387,0.0008416803,0.05487785,0.04072843,0.05720535,0.1007436,0.736993],"study_design_scores_gemma":[0.0004147083,0.0002705911,0.0007932827,0.0002132519,0.0000727068,0.0008282858,0.0001073811,0.7741306,0.05441997,0.06675629,0.101909,0.00008396313],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003929067,0.0001939397,0.8892648,0.0002195532,0.0001031583,0.0003374305,0.0009266137,0.1016724,0.003353063],"genre_scores_gemma":[0.07624695,0.0002648664,0.9056938,0.0004209897,0.0000449116,0.0004694144,0.003505808,0.00958259,0.003770598],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02112819,"threshold_uncertainty_score":0.0706808,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04919463481431056,"score_gpt":0.2964831204573316,"score_spread":0.247288485643021,"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."}}