{"id":"W4399353928","doi":"10.1145/3626772.3657670","title":"Retrieval-Augmented Conversational Recommendation with Prompt-based Semi-Structured Natural Language State Tracking","year":2024,"lang":"en","type":"article","venue":"","topic":"Speech and dialogue systems","field":"Computer Science","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto; University of Waterloo","funders":"","keywords":"Computer science; Natural language; Tracking (education); Natural (archaeology); Natural language processing; Artificial intelligence; State (computer science); Information retrieval; Programming language","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":[0.0002366602,0.0001280669,0.0001072112,0.0001373823,0.00006399454,0.0003818029,0.0002133199,0.00003658042,0.0001172184],"category_scores_gemma":[0.00002734551,0.00009122925,0.0000387062,0.0004792112,0.00002229983,0.0005388464,0.00002883001,0.0001370612,0.00005406034],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009944619,"about_ca_system_score_gemma":0.0001855097,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00006113875,"about_ca_topic_score_gemma":0.00005100535,"domain_scores_codex":[0.9989138,0.00006085304,0.0001803088,0.000352519,0.0002885102,0.0002040075],"domain_scores_gemma":[0.9995013,0.0001044366,0.00005116152,0.0001956474,0.00008194401,0.00006553857],"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":[0.001727212,0.0003341877,0.006282948,0.001499359,0.001053563,0.001162436,0.02602482,0.00202791,0.295148,0.02590212,0.07109222,0.5677452],"study_design_scores_gemma":[0.002059945,0.000265717,0.005889802,0.0002221765,0.00002028338,0.00008119533,0.0002706077,0.8068804,0.1755736,0.0007284806,0.007394899,0.0006129657],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1596456,0.0003861421,0.8262436,0.005112634,0.003614132,0.0008020806,0.0000427433,0.001622832,0.002530212],"genre_scores_gemma":[0.9896109,6.962721e-7,0.009085484,0.0003477951,0.0001014943,0.000006281018,0.0001848296,0.00001054884,0.0006519734],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8299652,"threshold_uncertainty_score":0.3720221,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.00979204426577812,"score_gpt":0.2436184272932854,"score_spread":0.2338263830275073,"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."}}