{"rank":null,"reference":false,"run_id":"8a920104cc0d","task":"clinicalqa@0","model":{"id":"gscfwid/qwen3-8b-pericomp","slug":"gscfwid/qwen3-8b-pericomp","name":"PeriComp-8B (Qwen3-8B + LoRA)","type":"llm","developer":"First Affiliated Hospital of Sun Yat-sen University"},"tier":"agent","metric":{"name":"accuracy","value":0.722488038277512,"ci_low":0.6947129186602872,"ci_high":0.7492822966507177},"n":1045,"parse_rate":0.9942583732057416,"cost_usd":0.0,"eval_date":"2026-10-29","url":"https://medbench.kr/r/8a920104cc0d","status":"done","listed":true,"current_spec":true,"metrics":{"accuracy":{"value":0.722488038277512,"ci_low":0.6947129186602872,"ci_high":0.7492822966507177,"n":1045,"n_unparsed":6},"parse_rate":0.9942583732057416,"error_rate":0.0,"parser":"p0.2","by_category":{"Allergy and Immunology":{"value":0.6944444444444444,"n":36},"Cardiology":{"value":0.797752808988764,"n":89},"Dermatology":{"value":0.6666666666666666,"n":12},"Emergency Medicine":{"value":0.71875,"n":32},"Endocrinology":{"value":0.7647058823529411,"n":68},"Endocrinology, metabolic, and nutritional disorders":{"value":0.0,"n":1},"Forensic Medicine":{"value":1.0,"n":1},"Gastroenterology":{"value":0.8035714285714286,"n":112},"General Surgery":{"value":0.5689655172413793,"n":58},"Geriatric Medicine":{"value":1.0,"n":1},"Gynecology":{"value":0.6571428571428571,"n":70},"Hematology and Oncology":{"value":0.8541666666666666,"n":48},"Infectious Diseases":{"value":0.6538461538461539,"n":52},"Musculoskeletal (e.g., costochondritis)":{"value":1.0,"n":1},"Nephrology":{"value":0.6396396396396397,"n":111},"Neurology / Neurosurgery":{"value":0.6530612244897959,"n":49},"Obstetrics":{"value":0.7222222222222222,"n":18},"Ophthalmology":{"value":1.0,"n":1},"Orthopedic Surgery":{"value":0.875,"n":24},"Otolaryngology":{"value":1.0,"n":14},"Otolaryngology (ENT)":{"value":0.8571428571428571,"n":14},"Pediatrics":{"value":0.7321428571428571,"n":56},"Pharmacology":{"value":1.0,"n":1},"Psychiatry":{"value":0.6666666666666666,"n":45},"Pulmonology (Respiratory Medicine)":{"value":0.7076923076923077,"n":65},"Radiology":{"value":1.0,"n":1},"Rheumatology":{"value":0.6938775510204082,"n":49},"Urology":{"value":0.7692307692307693,"n":13},"Vascular Surgery":{"value":1.0,"n":2},"nan":{"value":0.0,"n":1}}},"config":{"max_tokens":4096,"prompt":"clinicalqa-p0","temperature":0},"run_key":"c2bcf0438b8c7a6968f1046384a0e9fdeafc3e25fc286ff3b999b84cccd237e8","harness_version":"p0.1","task_spec_hash":"77e37cd0ca6bded7fba03792ec1922342b3f6a7dc3ce602881ce196e6aa7377c","model_id":"gscfwid/qwen3-8b-pericomp","tokens_in":0,"tokens_out":177517,"dataset":{"slug":"clinicalqa","title":"ClinicalQA: Korean Medical Knowledge QA Benchmark","license":"CC BY 4.0 (upstream Apache-2.0)","license_url":"https://creativecommons.org/licenses/by/4.0/","citation":"SNUH-HARI. ClinicalQA: Korean Medical Knowledge QA Benchmark. Seoul National University Hospital Healthcare AI Research Institute, 2025. doi:10.23421/2026.0800"},"predictions_url":"https://medbench.kr/artifacts/8a920104cc0d/predictions.parquet","bibtex":"@misc{medbench_8a920104cc0d,\n  title = {PeriComp-8B (Qwen3-8B + LoRA) on Korean clinical knowledge (KMLE-level MCQ) (clinicalqa@0)},\n  author = {{medbench.kr}},\n  year = {2026},\n  howpublished = {\\url{https://medbench.kr/r/8a920104cc0d}},\n  note = {Evaluated 2026-10-29, harness p0.1, run key c2bcf0438b8c}\n}"}