{"id":"W4412887723","doi":"10.18653/v1/2025.findings-acl.1160","title":"Understanding the Influence of Synthetic Data for Text Embedders","year":2025,"lang":"en","type":"article","venue":"","topic":"Advanced Text Analysis Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research; Samsung; National Science Foundation","keywords":"Computer science; Information retrieval; Data science","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"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":[0.00902088,0.001466875,0.000656111,0.001189038,0.000687103,0.002244955,0.001514556,0.001729466,0.002554092],"category_scores_gemma":[0.05826413,0.0005736639,0.0008211061,0.001083349,0.001671384,0.004780002,0.002674273,0.002079768,0.002063062],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008212372,"about_ca_system_score_gemma":0.0006874786,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002149385,"about_ca_topic_score_gemma":0.003453942,"domain_scores_codex":[0.9943777,0.003559005,0.0003482987,0.0009196603,0.0005936343,0.0002017052],"domain_scores_gemma":[0.9578854,0.02949482,0.001501939,0.00758642,0.002839421,0.0006920001],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002314029,0.001282795,0.04881242,0.0027095,0.0005417904,0.0008794341,0.002457128,0.4800454,0.03929755,0.02116455,0.06226191,0.3382335],"study_design_scores_gemma":[0.0001994419,0.0009486036,0.009086793,0.0003096441,0.00009790375,0.0005701699,0.001216751,0.8876811,0.04640447,0.02872539,0.02465364,0.0001061313],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6684281,0.003278161,0.2930714,0.00354662,0.0009133064,0.000686263,0.01288781,0.006593175,0.01059518],"genre_scores_gemma":[0.8269252,0.000720992,0.1383714,0.0007808433,0.0001921289,0.0007046011,0.02821897,0.0009396916,0.00314614],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00902088,"threshold_uncertainty_score":0.04770756,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08669165571493846,"score_gpt":0.3481905455380297,"score_spread":0.2614988898230913,"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."}}