{"id":"W4396827151","doi":"10.1145/3613904.3641895","title":"The HaLLMark Effect: Supporting Provenance and Transparent Use of Large Language Models in Writing with Interactive Visualization","year":2024,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":46,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Agency (philosophy); Computer science; Visualization; Control (management); Artificial intelligence; Sociology; Social science","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":[0.00592105,0.001124913,0.0005607522,0.001808775,0.001038613,0.004648219,0.001661929,0.001586673,0.006669917],"category_scores_gemma":[0.0317663,0.0006173418,0.001036773,0.0009173047,0.002071636,0.007528046,0.00589358,0.00222519,0.001165956],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008723572,"about_ca_system_score_gemma":0.001554215,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002806212,"about_ca_topic_score_gemma":0.004455874,"domain_scores_codex":[0.9977024,0.001413424,0.0001474238,0.0002904641,0.0003380202,0.0001081731],"domain_scores_gemma":[0.9785096,0.01459358,0.001061292,0.004001191,0.001000207,0.0008341518],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.003593715,0.0008451228,0.03025098,0.002547931,0.0003499016,0.003048826,0.08003781,0.04524808,0.08728436,0.1577967,0.09102321,0.4979734],"study_design_scores_gemma":[0.0007571189,0.0007114696,0.008231998,0.0007287387,0.0003071805,0.001227733,0.005717388,0.4403775,0.08582647,0.1976733,0.2578811,0.0005599443],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08169367,0.0005782403,0.8344019,0.002213918,0.0002981444,0.0002907128,0.001822717,0.06896207,0.009738576],"genre_scores_gemma":[0.4692174,0.0003912196,0.5179427,0.0003673413,0.0001137166,0.0003493801,0.001316044,0.00452764,0.005774531],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006669917,"threshold_uncertainty_score":0.0313139,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01848380606514614,"score_gpt":0.3086321793749555,"score_spread":0.2901483733098093,"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."}}