{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003680667,0.00006736262,0.0001142327,0.00009655236,0.00009964743,0.00005329858,0.002419344,0.00002381451,0.000004093762],"category_scores_gemma":[0.0002437657,0.00004403191,0.0000384315,0.0005119488,0.0001009551,0.0005004503,0.0006846165,0.00004742552,0.00000125705],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003659465,"about_ca_system_score_gemma":0.00005170902,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00000989307,"about_ca_topic_score_gemma":0.0000228161,"domain_scores_codex":[0.9992709,0.00002248434,0.000177983,0.0002836997,0.0001161744,0.0001287899],"domain_scores_gemma":[0.9976441,0.0004864973,0.00006992794,0.00172599,0.00005963293,0.00001386849],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00000205897,0.00001565744,0.0001063583,0.0000194548,0.00003865455,2.24705e-7,0.0001003237,0.001499294,0.0009657632,0.9868296,0.003759811,0.006662784],"study_design_scores_gemma":[0.0001266466,0.00002553447,0.0002801397,0.00006924062,0.00003796446,8.158871e-7,0.0003688757,0.4683729,0.006068369,0.5222933,0.002228804,0.0001274357],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0005066977,0.00004026575,0.9928261,0.00180719,0.00002170966,0.0001787861,0.000002366727,0.0001231838,0.004493679],"genre_scores_gemma":[0.9019654,0.00001050837,0.09737827,0.0003326949,0.00000272265,0.00001140394,0.000001177951,0.000002539622,0.0002952728],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9014587,"threshold_uncertainty_score":0.4495783,"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."}}