{"id":"W4286905174","doi":"10.18653/v1/2022.findings-emnlp.19","title":"Salient Phrase Aware Dense Retrieval: Can a Dense Retriever Imitate a Sparse One?","year":2022,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":36,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Phrase; Computer science; Salient; Artificial intelligence; Natural language processing; Labrador Retriever; Information retrieval; Medicine","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001612898,0.0006244724,0.00149037,0.0009632793,0.0009228851,0.002606259,0.001919517,0.001907094,0.007887177],"category_scores_gemma":[0.009181601,0.0006085531,0.0004683477,0.001012214,0.001212281,0.01079373,0.002473552,0.001332514,0.00742517],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003902503,"about_ca_system_score_gemma":0.0009270568,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003522962,"about_ca_topic_score_gemma":0.005553595,"domain_scores_codex":[0.9992493,0.0001720074,0.00004676718,0.0001992755,0.0002023152,0.0001303259],"domain_scores_gemma":[0.9968527,0.00112532,0.0001874833,0.001116662,0.0005014203,0.0002164123],"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":[0.003209937,0.0009691234,0.006219118,0.0008566311,0.0002243436,0.0008571787,0.001394034,0.01877316,0.1600301,0.03572308,0.06519365,0.7065497],"study_design_scores_gemma":[0.0004641838,0.001419231,0.006239248,0.0001147378,0.0003979281,0.002173704,0.001957069,0.7519554,0.0698045,0.1243604,0.04088464,0.0002290666],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3170975,0.006974966,0.5965815,0.00770233,0.001584941,0.0005220646,0.001348376,0.01740941,0.05077898],"genre_scores_gemma":[0.8373392,0.001463006,0.142519,0.001367033,0.0005759281,0.0001002148,0.001308435,0.0009129272,0.01441435],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007887177,"threshold_uncertainty_score":0.02638525,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02471533952713897,"score_gpt":0.2593173851862666,"score_spread":0.2346020456591276,"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."}}