{"id":"W4412691112","doi":"10.22260/isarc2025/0088","title":"RAG-Enhanced Safety Information Retrieval for Construction: Integration of Large Language Models with Domain-Specific Information","year":2025,"lang":"en","type":"article","venue":"Proceedings of the ... ISARC","topic":"Occupational Health and Safety Research","field":"Health Professions","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"University of Alberta","keywords":"Computer science; Domain (mathematical analysis); Information retrieval; Information integration; Natural language processing; Information model; Artificial intelligence; Data mining; Software engineering; Mathematics","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001251956,0.0001168952,0.0002445267,0.000239894,0.0005706655,0.0000146344,0.0002425236,0.0001595575,0.00004132602],"category_scores_gemma":[0.0002911487,0.00007908217,0.00006150335,0.0006996801,0.0001101908,0.001566144,0.00009157774,0.0003530544,0.000008571469],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002025667,"about_ca_system_score_gemma":0.0004473344,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002818446,"about_ca_topic_score_gemma":0.000009134334,"domain_scores_codex":[0.9982168,0.00003049528,0.0008836912,0.0001005411,0.0004562951,0.0003121734],"domain_scores_gemma":[0.9967685,0.0002955023,0.0006586305,0.0001332426,0.002090955,0.00005311373],"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":[0.009788407,0.00005474197,0.002377549,0.003800299,0.00004300884,1.120733e-8,0.01901472,0.00005959579,0.00651459,0.9296311,0.004169038,0.02454692],"study_design_scores_gemma":[0.03251492,0.001895091,0.09722422,0.00837117,0.0001737242,0.000008718759,0.2859755,0.01405245,0.1796526,0.1476584,0.2313545,0.001118695],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.605284,0.0001006198,0.2509224,0.005271408,0.0008858571,0.007295566,0.0005843083,0.00009888885,0.129557],"genre_scores_gemma":[0.9864309,0.00007280497,0.01246895,0.0003690589,0.00007414822,0.00016777,0.0001189686,0.000007322519,0.0002901032],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.7819728,"threshold_uncertainty_score":0.4389156,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02371043577322171,"score_gpt":0.368119619173015,"score_spread":0.3444091833997933,"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."}}