{"id":"W3214455632","doi":"10.18653/v1/2021.emnlp-main.77","title":"Contextualized Query Embeddings for Conversational Search","year":2021,"lang":"en","type":"article","venue":"Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing","topic":"Topic Modeling","field":"Computer Science","cited_by":41,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canada First Research Excellence Fund","keywords":"Computer science; Leverage (statistics); Inference; Security token; Relevance (law); Pipeline (software); Information retrieval; Query expansion; Query language; Artificial intelligence","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":[0.001528358,0.0002092998,0.000406463,0.0001410444,0.0001425354,0.0002663512,0.001095299,0.0001382064,0.00003392003],"category_scores_gemma":[0.002933444,0.0001554843,0.0001574349,0.0008547008,0.0001143692,0.0004999717,0.0004836491,0.000560144,0.00000127803],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000109687,"about_ca_system_score_gemma":0.0004046171,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001117802,"about_ca_topic_score_gemma":0.000002822563,"domain_scores_codex":[0.9978356,0.0001133703,0.0004976453,0.0006551637,0.0005029588,0.000395313],"domain_scores_gemma":[0.9978395,0.0006282433,0.0002411876,0.0002224463,0.0009941956,0.00007447814],"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":[0.0001031106,0.000139312,0.001474496,0.000526541,0.00003350245,0.000006205812,0.01286653,0.00003979215,0.1856029,0.09630091,0.0001195604,0.7027871],"study_design_scores_gemma":[0.001147327,0.00004950101,0.0006241996,0.0007332677,0.00001815083,0.00002719991,0.004115626,0.6975945,0.2719545,0.02313994,0.0002231293,0.0003726969],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1644545,0.004331561,0.8044723,0.01841967,0.0009631083,0.0009185073,0.00001029069,0.000147779,0.006282305],"genre_scores_gemma":[0.5235511,0.000006272351,0.4754031,0.0005803296,0.00005810444,0.00002474042,0.00000114491,0.000009250772,0.0003659665],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7024144,"threshold_uncertainty_score":0.6340467,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.092333595524727,"score_gpt":0.4395395092037929,"score_spread":0.3472059136790658,"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."}}