{"id":"W6947999492","doi":"10.48448/chfw-wa16","title":"ChatRetriever: Adapting Large Language Models for Generalized and Robust Conversational Dense Retrieval","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Biomedical Text Mining and Ontologies","field":"Biochemistry, Genetics and Molecular Biology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Robustness (evolution); Language model; Generalization; Rewriting; Session (web analytics); Interpretation (philosophy); Training set","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":[],"consensus_categories":[],"category_scores_codex":[0.0005522319,0.000191489,0.0001907883,0.0001598324,0.00009412919,0.00007046491,0.0002439614,0.000328929,0.00006245201],"category_scores_gemma":[0.0002601838,0.0001619265,0.00005594625,0.0001828672,0.0005705167,0.000004109027,0.0002122787,0.000120796,0.00001229399],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002190408,"about_ca_system_score_gemma":0.0002644089,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003532061,"about_ca_topic_score_gemma":0.00006654973,"domain_scores_codex":[0.9985957,0.00001912729,0.0001638628,0.0006118516,0.000267215,0.000342212],"domain_scores_gemma":[0.9994813,0.00002223352,0.00009287646,0.00022743,0.00006499334,0.0001111525],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0003556394,0.0001700653,0.00009154123,0.0007912817,0.0004196349,0.0000502757,0.001001148,0.0002427795,0.07113152,0.0152187,0.88638,0.02414746],"study_design_scores_gemma":[0.002603489,0.0005709515,0.0000141035,0.0002834986,0.0001975725,0.00006681246,0.001427317,0.1212941,0.007528043,0.001874074,0.863077,0.00106304],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"other","genre_scores_codex":[0.03994914,0.136914,0.677241,0.005563327,0.006884319,0.00336241,0.005105079,0.00108629,0.1238944],"genre_scores_gemma":[0.1135837,0.001543429,0.1764615,0.002009823,0.003527485,0.00004385353,0.001376849,0.0005796323,0.7008737],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5769793,"threshold_uncertainty_score":0.660317,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03032577417322737,"score_gpt":0.3014703699196813,"score_spread":0.2711445957464539,"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."}}