{"id":"W4411113084","doi":"10.18653/v1/2025.findings-naacl.168","title":"Language Modeling with Editable External Knowledge","year":2025,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; National Science Foundation","keywords":"Computer science; Human–computer interaction","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001781863,0.0009293765,0.0006457606,0.001213021,0.0004791631,0.002774957,0.00252718,0.001271078,0.005324288],"category_scores_gemma":[0.008658877,0.0006454459,0.002010844,0.0009723183,0.0008625502,0.003287386,0.001706081,0.001981124,0.002177146],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001016573,"about_ca_system_score_gemma":0.001086013,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01076768,"about_ca_topic_score_gemma":0.01698334,"domain_scores_codex":[0.9980963,0.0007162485,0.0001255947,0.0005000487,0.000453523,0.0001082121],"domain_scores_gemma":[0.9954432,0.002334827,0.0003460872,0.001248435,0.0005155288,0.0001119429],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001469641,0.0001489289,0.002117231,0.0001798016,0.0001642967,0.000542576,0.0004415943,0.815808,0.002932635,0.05998463,0.00825992,0.1092734],"study_design_scores_gemma":[0.00001184298,0.00001479901,0.000128804,0.00001011484,0.00001755936,0.00005368468,0.00002044506,0.9657487,0.001029956,0.02685773,0.006096699,0.000009695075],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01195714,0.0002104794,0.9751771,0.0004480256,0.00008282054,0.0001106978,0.001519682,0.005167793,0.005326175],"genre_scores_gemma":[0.447965,0.0006042228,0.5242463,0.000469051,0.0001854711,0.0004973169,0.006322179,0.00145325,0.01825722],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01076768,"threshold_uncertainty_score":0.02140999,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008067877120099843,"score_gpt":0.279647331919547,"score_spread":0.2715794547994472,"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."}}