{"id":"W4415595662","doi":"10.1016/j.ipm.2025.104449","title":"Leveraging historical information to boost retrieval-augmented generation in conversations","year":2025,"lang":"en","type":"article","venue":"Information Processing & Management","topic":"Topic Modeling","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Natural Sciences and Engineering Research Council of Canada; Amazon","keywords":"Boosting (machine learning); Context (archaeology); Text generation; Information system; Information needs","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0004703152,0.0001131484,0.000104867,0.0009498718,0.0001881476,0.0005853084,0.0003742708,0.00004483763,0.000005638912],"category_scores_gemma":[0.00005967124,0.0001254909,0.0000229827,0.001166382,0.000005836138,0.006181285,0.0002183357,0.00009820595,0.00007804245],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001146405,"about_ca_system_score_gemma":0.00008666274,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004209857,"about_ca_topic_score_gemma":0.000004543676,"domain_scores_codex":[0.9986263,0.00002286311,0.0006491445,0.000144283,0.0003559204,0.0002014971],"domain_scores_gemma":[0.9993271,0.00001043932,0.0001583822,0.0002845646,0.0001746095,0.00004495194],"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":[0.00001631876,0.00003255333,0.0002181842,0.00030438,0.00001555554,0.000001205357,0.009949646,0.0653628,0.00003563702,0.05186254,0.007595285,0.8646059],"study_design_scores_gemma":[0.0005988258,0.000010351,0.001640469,0.0001048794,0.000007125889,7.682746e-7,0.0003867289,0.8718935,0.000258555,0.0008388526,0.1240977,0.000162218],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004151338,0.00002263038,0.9767076,0.004804445,0.0004668346,0.0004468785,4.664103e-7,0.0001721292,0.01322762],"genre_scores_gemma":[0.9252659,0.000009946015,0.06759975,0.006311914,0.0000273636,0.00007565617,0.00005457755,0.00000319848,0.0006516599],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9211146,"threshold_uncertainty_score":0.5644143,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02004362624849737,"score_gpt":0.2488657579690393,"score_spread":0.2288221317205419,"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."}}