{"id":"W4413267863","doi":"10.1145/3696630.3728530","title":"LicenseGPT: A Fine-tuned Foundation Model for Publicly Available Dataset License Compliance","year":2025,"lang":"en","type":"article","venue":"","topic":"Scientific Computing and Data Management","field":"Decision Sciences","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University; Huawei Technologies (Canada)","funders":"","keywords":"Foundation (evidence); License; Compliance (psychology); Computer science; Computer security; Law","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":[0.008140398,0.00199607,0.0008642787,0.005762366,0.001660432,0.00285734,0.004199395,0.002465322,0.009151245],"category_scores_gemma":[0.05149226,0.0006789102,0.002225295,0.004145962,0.0010666,0.004487111,0.003650748,0.003464197,0.00906607],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002508453,"about_ca_system_score_gemma":0.005223762,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.03176346,"about_ca_topic_score_gemma":0.05620645,"domain_scores_codex":[0.9927973,0.001642737,0.0007544588,0.002280571,0.002045706,0.0004792288],"domain_scores_gemma":[0.9812768,0.007702778,0.001227304,0.005636062,0.003602112,0.0005550368],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0007180375,0.0009322097,0.04381666,0.001602076,0.0004448869,0.0007074955,0.0006115289,0.04986381,0.005621118,0.009366404,0.6370425,0.2492733],"study_design_scores_gemma":[0.0004915356,0.0004044251,0.02789697,0.000542635,0.0001753667,0.0008224319,0.0007946637,0.6243117,0.01135358,0.01898256,0.3140034,0.0002206969],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"software","genre_scores_codex":[0.1208786,0.001939965,0.1981888,0.003533741,0.001144971,0.001792575,0.4374335,0.2135454,0.02154247],"genre_scores_gemma":[0.1260545,0.0002756791,0.1787344,0.0008749187,0.00010899,0.001644024,0.6840987,0.003992607,0.004216177],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.03176346,"threshold_uncertainty_score":0.06315714,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3884682337846641,"score_gpt":0.4575310168340584,"score_spread":0.06906278304939434,"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."}}