{"id":"W4409588626","doi":"10.36227/techrxiv.174494685.55828887/v1","title":"Investigations on using Evidence-Based GraphRag Pipeline using LLM Tailored for Answering USMLE Medical Exam Questions","year":2025,"lang":"en","type":"preprint","venue":"","topic":"Online Learning and Analytics","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Lakehead University","keywords":"Pipeline (software); Medical education; Computer science; Medicine; Medical physics; Programming language","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":[0.008668308,0.0009800461,0.0006308299,0.003031297,0.0007683548,0.003409184,0.002188125,0.002441399,0.01034736],"category_scores_gemma":[0.03655001,0.0004376382,0.001182262,0.001573514,0.001055378,0.006194334,0.003251553,0.001992917,0.003599219],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001260635,"about_ca_system_score_gemma":0.002673666,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005527983,"about_ca_topic_score_gemma":0.00860143,"domain_scores_codex":[0.9951591,0.002607581,0.0003646253,0.0007579609,0.000953839,0.0001568929],"domain_scores_gemma":[0.9770454,0.01529706,0.0008465349,0.004191731,0.002205629,0.0004136078],"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":[0.001187805,0.001095537,0.01431405,0.002782455,0.0003432049,0.001219019,0.003544651,0.03314427,0.04259645,0.03281454,0.03271953,0.8342385],"study_design_scores_gemma":[0.0005617465,0.001154448,0.005881798,0.0007804022,0.0004552908,0.001647202,0.001632262,0.7058387,0.08783064,0.06992932,0.124021,0.0002672258],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07776728,0.001773992,0.8140193,0.005403385,0.0002835275,0.001947769,0.003591083,0.08365906,0.01155461],"genre_scores_gemma":[0.2504108,0.000426575,0.7389363,0.001194801,0.00006007474,0.0002595693,0.004814344,0.001002803,0.002894766],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01034736,"threshold_uncertainty_score":0.04584295,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1435985957218573,"score_gpt":0.3898692550282539,"score_spread":0.2462706593063967,"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."}}