{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003429842,0.001157977,0.001206268,0.005450229,0.0008824331,0.003176612,0.001153788,0.001533122,0.004211226],"category_scores_gemma":[0.01660377,0.0003819751,0.0009835777,0.003644906,0.0005530113,0.008168432,0.00298665,0.0008994211,0.002862401],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00102159,"about_ca_system_score_gemma":0.001525658,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002615832,"about_ca_topic_score_gemma":0.004488521,"domain_scores_codex":[0.993435,0.002029654,0.0007689456,0.0005989517,0.002737104,0.0004304055],"domain_scores_gemma":[0.9916333,0.003477356,0.0004477378,0.002253617,0.001852922,0.000334969],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002056782,0.001081343,0.01229859,0.001581408,0.0004287557,0.000285092,0.0003478552,0.01379496,0.01244664,0.02629204,0.08699522,0.8423914],"study_design_scores_gemma":[0.0005034741,0.002652926,0.01121742,0.0004357779,0.000691151,0.001127905,0.001109783,0.8292393,0.03990649,0.04649923,0.06646402,0.0001525991],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5777419,0.03576883,0.2680689,0.003704695,0.002455028,0.002008196,0.01594681,0.03520228,0.05910336],"genre_scores_gemma":[0.7879354,0.004609993,0.1618961,0.0003548454,0.0004585493,0.0005109644,0.03551025,0.0009653395,0.007758689],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005450229,"threshold_uncertainty_score":0.01813895,"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."}}