{"id":"W4385570706","doi":"10.18653/v1/2023.acl-industry.50","title":"Evaluating Embedding APIs for Information Retrieval","year":2023,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Embedding; Volume (thermodynamics); Computer science; Library science; Information retrieval; Artificial intelligence; Physics","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0008465969,0.00005814701,0.00006064967,0.0001632816,0.0001180121,0.0002455174,0.0004311189,0.00004159911,0.000004849729],"category_scores_gemma":[0.0006680447,0.00004853659,0.00003104206,0.0006727738,0.000007025213,0.001563881,0.0001792458,0.00005737881,0.00006417977],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003053355,"about_ca_system_score_gemma":0.00003730101,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003951632,"about_ca_topic_score_gemma":2.751908e-7,"domain_scores_codex":[0.999316,0.00001302676,0.0001543032,0.0001150614,0.0002271303,0.0001744901],"domain_scores_gemma":[0.9994326,0.0001349465,0.00006334641,0.0001940977,0.0001491858,0.00002578599],"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":[0.00001489057,0.000005387291,0.00002080893,0.000116841,0.000008276603,0.000001408206,0.001920938,0.0001796442,0.01600756,0.2568991,0.01184219,0.712983],"study_design_scores_gemma":[0.0001289981,0.00005648528,0.00001219413,0.00001970188,0.000001699408,0.000002906721,0.00003488686,0.8832302,0.05930538,0.05611226,0.000995525,0.00009973955],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.002666896,0.0000402412,0.9932837,0.0006733369,0.0001549842,0.0002040519,0.00000107568,0.002399265,0.0005764773],"genre_scores_gemma":[0.09190515,0.000001573479,0.9073433,0.0003471495,0.00003652223,0.00001692001,0.0000102119,0.000003654854,0.0003354712],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.8830506,"threshold_uncertainty_score":0.2367529,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05085128616261112,"score_gpt":0.3954791912835629,"score_spread":0.3446279051209518,"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."}}