{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0007535756,0.0002634743,0.0002702836,0.0002693294,0.0004574497,0.0002578223,0.001689856,0.00007438353,0.0002487243],"category_scores_gemma":[0.0002329523,0.0002509293,0.0001203277,0.001525016,0.00006474047,0.0003782716,0.001832636,0.0005830295,0.00003138045],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004028561,"about_ca_system_score_gemma":0.0002887327,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002674492,"about_ca_topic_score_gemma":0.00005776455,"domain_scores_codex":[0.9970936,0.0002236829,0.000367459,0.0007799732,0.0009699932,0.0005653553],"domain_scores_gemma":[0.9981877,0.0001012032,0.0001827224,0.001117342,0.0001903674,0.0002206894],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001673447,0.002705199,0.002766601,0.0004091978,0.0004403661,0.0121619,0.02069305,0.0003582465,0.1008776,0.2685953,0.5017331,0.08758603],"study_design_scores_gemma":[0.004823371,0.00249525,0.0007583379,0.0002775821,0.0002045091,0.002004838,0.001494422,0.07192453,0.6086769,0.2246276,0.0772242,0.005488457],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2080136,0.00427957,0.7557431,0.0194002,0.001389597,0.001585279,0.0001008559,0.008131811,0.001355964],"genre_scores_gemma":[0.7785577,0.00001840667,0.2143856,0.004588081,0.000046701,0.00003869721,0.00001219803,0.00003381486,0.00231872],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5705441,"threshold_uncertainty_score":0.9999943,"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."}}