{"id":"W4414153271","doi":"10.1016/j.eswa.2025.129655","title":"The power of text similarity in identifying AI-LLM paraphrased documents: The case of BBC news articles and ChatGPT","year":2025,"lang":"en","type":"article","venue":"Expert Systems with Applications","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Calgary","funders":"","keywords":"Similarity (geometry); Task (project management); Benchmark (surveying); Generative grammar; Power (physics); Revenue","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":[0.007110476,0.0005407254,0.0007092745,0.01140198,0.003344744,0.00597668,0.001342964,0.003148895,0.005507607],"category_scores_gemma":[0.06894235,0.0004275022,0.0004738765,0.008124547,0.002157419,0.009066674,0.003023935,0.001723876,0.002694533],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001216189,"about_ca_system_score_gemma":0.001126698,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008955238,"about_ca_topic_score_gemma":0.009777203,"domain_scores_codex":[0.9941332,0.003259786,0.0004464957,0.0006202885,0.001164251,0.0003759806],"domain_scores_gemma":[0.9056955,0.07807218,0.004010832,0.004528781,0.006852578,0.0008401277],"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":[0.005071802,0.0009337754,0.09735958,0.003144718,0.0004623549,0.01640916,0.04600232,0.0161531,0.06638074,0.0523103,0.02426023,0.671512],"study_design_scores_gemma":[0.0003554186,0.00106395,0.1254584,0.0009810438,0.0007646747,0.02996591,0.04423346,0.5220339,0.06491574,0.1256752,0.08407681,0.0004754693],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8036401,0.005222904,0.1081948,0.004864914,0.0002720123,0.0005000415,0.001981922,0.001728456,0.07359492],"genre_scores_gemma":[0.9546155,0.0005642981,0.03962215,0.0002213767,0.0002025694,0.00006708888,0.001087542,0.0002647492,0.003354795],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01140198,"threshold_uncertainty_score":0.03760427,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01692661947228411,"score_gpt":0.2993149508804233,"score_spread":0.2823883314081391,"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."}}