{"id":"W4403232958","doi":"10.1007/978-3-031-74186-9_17","title":"Enhancing RAG-Retrieval to Improve LLMs Robustness and Resilience to Hallucinations","year":2024,"lang":"en","type":"book-chapter","venue":"Lecture notes in computer science","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"Université du Québec à Chicoutimi","funders":"","keywords":"Computer science; Robustness (evolution); Resilience (materials science); Computer security; Artificial intelligence","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.007911555,0.0004649042,0.0005477844,0.002773748,0.0004881073,0.003325457,0.003591567,0.0001671289,0.00007855456],"category_scores_gemma":[0.003435028,0.0003716752,0.0001028861,0.003074951,0.0005209788,0.0004377733,0.004973651,0.0005590971,0.0004767665],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002741096,"about_ca_system_score_gemma":0.0003450865,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000354416,"about_ca_topic_score_gemma":0.0004293403,"domain_scores_codex":[0.9915017,0.00004139285,0.0009662135,0.003634043,0.003130861,0.0007257544],"domain_scores_gemma":[0.9946398,0.00188344,0.0002009752,0.002243651,0.0005674352,0.0004647128],"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":[0.00001981893,0.0000202064,0.00001951352,0.00003496892,0.000008800413,0.00008321246,0.001590366,0.2639486,0.001173047,0.008031944,0.0008072176,0.7242622],"study_design_scores_gemma":[0.0002933754,0.0005044991,0.0005540218,0.001326543,0.00003752093,0.00005947603,0.00001790168,0.7555236,0.004305399,0.198378,0.0373892,0.001610438],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.002989164,0.000221451,0.9822562,0.003615869,0.006497021,0.0006889551,0.0000354422,0.000109341,0.003586554],"genre_scores_gemma":[0.5641589,0.00001682555,0.3798765,0.004466643,0.001568552,0.00002344977,0.00001173163,0.00009663003,0.04978082],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7226518,"threshold_uncertainty_score":0.9998735,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03832758125998025,"score_gpt":0.3308532131839272,"score_spread":0.292525631923947,"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."}}