{"id":"W4416036138","doi":"10.18653/v1/2025.emnlp-main.1075","title":"Improving Context Fidelity via Native Retrieval-Augmented Reasoning","year":2025,"lang":"","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Alliance de recherche numérique du Canada","keywords":"Context (archaeology); Natural language; Fidelity; Natural (archaeology); Natural language understanding","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":[0.001960335,0.001603717,0.00240186,0.001738027,0.0009826476,0.002535687,0.002625749,0.001857062,0.007641313],"category_scores_gemma":[0.01226902,0.0006424639,0.001329433,0.001419134,0.0007152082,0.00691696,0.004574246,0.001944386,0.002861767],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000579229,"about_ca_system_score_gemma":0.001553629,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006666288,"about_ca_topic_score_gemma":0.0100605,"domain_scores_codex":[0.9975067,0.0005784072,0.000194148,0.0006971765,0.0007764045,0.0002472269],"domain_scores_gemma":[0.9955604,0.001596048,0.0001537978,0.001914415,0.0006530152,0.0001223631],"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":[0.001377232,0.0007866228,0.002824381,0.0007852581,0.0002658897,0.0004101006,0.0006343838,0.04552665,0.04123369,0.01273456,0.02344984,0.8699713],"study_design_scores_gemma":[0.0003295501,0.0003661329,0.001567082,0.00009145308,0.0004902235,0.0005557906,0.0004324904,0.9220186,0.02943992,0.03028955,0.01432017,0.00009907542],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.08223024,0.006525199,0.8819255,0.0006027843,0.0006383535,0.0002724249,0.0009463243,0.01555634,0.0113028],"genre_scores_gemma":[0.6532009,0.001471597,0.3353485,0.0005716381,0.0003183206,0.0001236847,0.00204763,0.000751908,0.006165895],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.007641313,"threshold_uncertainty_score":0.02556276,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01701550256221682,"score_gpt":0.2702245270101477,"score_spread":0.2532090244479308,"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."}}