{"id":"W4402671823","doi":"10.18653/v1/2024.acl-long.208","title":"DataDreamer: A Tool for Synthetic Data Generation and Reproducible LLM Workflows","year":2024,"lang":"en","type":"article","venue":"","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute","funders":"Intelligence Advanced Research Projects Activity; Office of the Director of National Intelligence","keywords":"Workflow; Computer science; Synthetic data; Artificial intelligence; Database","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":"codex-gemma-dda1882f352a","candidate_categories":["scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.0115215,0.00008705826,0.0001216807,0.0002078022,0.0001797269,0.003025146,0.001109772,0.00002176442,0.0004096068],"category_scores_gemma":[0.004852706,0.00005813914,0.00002694199,0.0006252698,0.00005082005,0.0008818494,0.001314931,0.00004293289,0.0003093915],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000009518497,"about_ca_system_score_gemma":0.00003997922,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002891386,"about_ca_topic_score_gemma":0.0000434834,"domain_scores_codex":[0.9967215,0.00005453597,0.0003864311,0.002082093,0.0005798093,0.0001756023],"domain_scores_gemma":[0.9944751,0.0007955598,0.00003952721,0.004560421,0.00007992851,0.000049511],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000002253198,0.00001254967,0.00002758964,0.000005659729,0.000006453597,0.000001457544,0.00003569504,0.00004104496,0.000115859,0.003415823,0.5653246,0.431011],"study_design_scores_gemma":[0.00004036433,0.00001361237,0.00005239324,0.00001146512,0.00001218018,0.000004248277,0.00006382135,0.4875559,0.00008513944,0.003113254,0.508984,0.00006372168],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03529072,0.001557063,0.9420592,0.007467365,0.005030012,0.0007675477,0.0005538952,0.0003116236,0.006962618],"genre_scores_gemma":[0.7608845,0.00008129166,0.1281273,0.0007555896,0.001450449,0.00005380199,0.001365758,0.00003114079,0.1072501],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8139318,"threshold_uncertainty_score":0.9980098,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.4313535412028034,"score_gpt":0.457528050023825,"score_spread":0.02617450882102157,"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."}}