{"id":"W7130374227","doi":"10.48448/xxg5-gk54","title":"Improving LLM’s Attachment to External Knowledge In Dialogue Generation Tasks Through Entity Anonymization","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Leverage (statistics); Task (project management); Knowledge graph; Domain knowledge; Knowledge-based systems; Simple (philosophy)","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":[0.01036884,0.001967631,0.0009250212,0.002052014,0.001680095,0.003100791,0.002446327,0.003249605,0.006832388],"category_scores_gemma":[0.03728606,0.0005736133,0.001123782,0.001275054,0.001186599,0.006611183,0.007016658,0.003190969,0.006961573],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001309555,"about_ca_system_score_gemma":0.001798836,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004632637,"about_ca_topic_score_gemma":0.006566078,"domain_scores_codex":[0.9845834,0.01024201,0.0006393277,0.002616629,0.001455408,0.0004632266],"domain_scores_gemma":[0.9804401,0.01080196,0.0008179064,0.005894573,0.001465331,0.0005801977],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001974751,0.001519976,0.01735216,0.001799221,0.0004088144,0.0009546104,0.005552873,0.06302061,0.01817249,0.01671287,0.141569,0.7309627],"study_design_scores_gemma":[0.0002812827,0.0005493551,0.007183032,0.0004004213,0.0001986171,0.0009489059,0.00264523,0.8123068,0.02834244,0.03721171,0.1096821,0.0002501443],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2353203,0.003731176,0.621349,0.004976099,0.001567525,0.001372963,0.01741327,0.08360516,0.03066441],"genre_scores_gemma":[0.6223959,0.0005365163,0.3221524,0.00161372,0.0002507931,0.0007138035,0.0384585,0.001714049,0.01216432],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01036884,"threshold_uncertainty_score":0.05483633,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05890460343334033,"score_gpt":0.3644919037656627,"score_spread":0.3055873003323224,"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."}}