{"id":"W4415974188","doi":"10.1016/j.procs.2025.09.549","title":"Content Safety and Response Quality in LLMs: A Data-Centric Refinement Approach","year":2025,"lang":"en","type":"article","venue":"Procedia Computer Science","topic":"Text Readability and Simplification","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Chicoutimi","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Readability; Performance improvement; Quality management; Quality (philosophy); Coherence (philosophical gambling strategy); Natural language generation","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.006106526,0.0001470383,0.0002224472,0.0003805514,0.0002378203,0.0003841975,0.002767989,0.00004568907,7.806344e-7],"category_scores_gemma":[0.0005462882,0.0001291532,0.00001922888,0.002383828,0.0003486284,0.001200166,0.00222009,0.0001590112,0.000004207888],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002049748,"about_ca_system_score_gemma":0.0007047034,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00007760456,"about_ca_topic_score_gemma":0.0000196197,"domain_scores_codex":[0.9971364,0.0002044544,0.00050893,0.001269239,0.0004858568,0.0003951549],"domain_scores_gemma":[0.997697,0.0003407176,0.0001092315,0.001553901,0.0001796168,0.000119519],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0003947565,0.0008021315,0.033593,0.0002334842,0.0000121423,0.000005597402,0.004329355,0.0003573021,0.001777688,0.2795681,0.0003251575,0.6786013],"study_design_scores_gemma":[0.0005407641,0.00004672926,0.514718,0.0000329941,0.00000234314,0.000008043658,0.00004996826,0.4806948,0.000175672,0.001210025,0.002343748,0.0001768844],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.09911843,0.0001918941,0.8962634,0.003310495,0.0002567082,0.0004099119,0.000005109956,0.000104146,0.0003399305],"genre_scores_gemma":[0.825217,0.00003261137,0.1738598,0.0007730111,0.00002191671,0.00002367351,0.000005501235,0.000002536395,0.00006391456],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7260986,"threshold_uncertainty_score":0.5266713,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08154692548813655,"score_gpt":0.3217937313509582,"score_spread":0.2402468058628217,"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."}}