{"id":"W4414578107","doi":"10.48550/arxiv.2503.08398","title":"OpenRAG: Optimizing RAG End-to-End via In-Context Retrieval Learning","year":2025,"lang":"en","type":"preprint","venue":"ArXiv.org","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Relevance (law); Adaptation (eye); Bridge (graph theory); Range (aeronautics); Language model; Labrador Retriever","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow","research_integrity"],"consensus_categories":[],"category_scores_codex":[0.001556067,0.0005077193,0.0006883892,0.0007114901,0.0003131202,0.0005117299,0.002284918,0.0003871772,0.0002425278],"category_scores_gemma":[0.0006374166,0.0005715949,0.0002372958,0.001072052,0.00006626457,0.0004644383,0.003848609,0.002514649,0.0005617162],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003019454,"about_ca_system_score_gemma":0.0004849114,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002590723,"about_ca_topic_score_gemma":0.00006846646,"domain_scores_codex":[0.9958953,0.000487537,0.0008062567,0.001487513,0.0005741206,0.0007492903],"domain_scores_gemma":[0.9975731,0.0004307616,0.0003798242,0.001149735,0.0001925868,0.0002740257],"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":[0.0003057285,0.0003198959,0.1527451,0.0004818124,0.0003547752,0.0007126351,0.03661999,0.3630932,0.004528596,0.009122565,0.000943569,0.4307721],"study_design_scores_gemma":[0.004757008,0.0004812136,0.1990729,0.003261443,0.0001145385,0.00006974177,0.003382251,0.5824614,0.005370814,0.001569622,0.1949598,0.004499265],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2277974,0.0007189218,0.7362396,0.00253156,0.002813137,0.0008669821,0.000004291705,0.0005937981,0.02843432],"genre_scores_gemma":[0.9536135,0.00009316796,0.03588251,0.001681534,0.0001756266,0.00004253281,0.00002616204,0.00003689458,0.008448086],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7258161,"threshold_uncertainty_score":0.9997866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04396895051009762,"score_gpt":0.2890148634946892,"score_spread":0.2450459129845916,"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."}}