{"id":"W7077919379","doi":"10.48448/gt1h-8n75","title":"Swan and ArabicMTEB: Dialect-Aware, Arabic-Centric, Cross-Lingual, and Cross-Cultural Embedding Models and Benchmarks","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Arabic; Embedding; Benchmark (surveying); Suite; Field (mathematics); Language model; Natural language","routes":{"ca_aff":true,"ca_fund":false,"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":[0.003676997,0.003852857,0.001012653,0.00280815,0.0009838528,0.002252077,0.003013731,0.001861498,0.005290887],"category_scores_gemma":[0.01137079,0.0005366282,0.001660598,0.002417661,0.0007433865,0.003985526,0.003059818,0.002607557,0.004478056],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001571407,"about_ca_system_score_gemma":0.001627651,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01770595,"about_ca_topic_score_gemma":0.02655782,"domain_scores_codex":[0.9977786,0.0009055418,0.0001974997,0.0005550336,0.0004027998,0.000160635],"domain_scores_gemma":[0.9964759,0.001451269,0.0001622281,0.0009277795,0.0007771363,0.0002056694],"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":[0.001618673,0.001525532,0.01284284,0.002404012,0.0009757279,0.0005850826,0.0004542427,0.3272273,0.00724141,0.005304111,0.1601807,0.4796404],"study_design_scores_gemma":[0.0001946996,0.0005124815,0.003476593,0.0002138465,0.0001783582,0.0003497849,0.0005065131,0.9382048,0.01025908,0.0104323,0.03556438,0.0001072737],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5354797,0.02432072,0.2609464,0.00545419,0.004341451,0.002108025,0.06655047,0.05547083,0.04532824],"genre_scores_gemma":[0.5303785,0.00372085,0.2674356,0.00173983,0.0003504118,0.001686773,0.1753212,0.002640928,0.01672603],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.01770595,"threshold_uncertainty_score":0.03520578,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01905219705620825,"score_gpt":0.320158114463073,"score_spread":0.3011059174068648,"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."}}