Datasets:
| { | |
| "meta_info": { | |
| "case_id": "case39_29", | |
| "bench_subset": "NeuroBench-Multi", | |
| "difficult": 1, | |
| "original_dataset": "Multi_QA" | |
| }, | |
| "agent_input": { | |
| "data_path": [], | |
| "instruction": "Answer this standalone multimodal neurophysiology multiple-choice question: Which physiological quantities are estimated by conventional functional near-infrared spectroscopy? Options: A: It directly records cardiac electrical depolarization. B: It directly measures absolute cerebral blood flow in every vessel. C: It estimates changes in oxyhemoglobin and deoxyhemoglobin concentration. D: It directly records neuronal action-potential voltage. Select exactly one option, A, B, C, or D. Return only one JSON object with the single field answer and no explanation or extra fields." | |
| }, | |
| "eval_config": { | |
| "parser_prompt": "### ROLE\nYou are a precise JSON Extraction Engine for neurophysiological data. Your sole task is to convert natural language reports into structured JSON data.\n\n### TASK\nExtract the answer choice from the agent's report.\n- answer: exactly one of A, B, C, or D.\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks.\n3. DO NOT include conversational text, explanations, or extra fields.\n4. The JSON key must be EXACTLY \"answer\".\n5. The answer value must be exactly one of \"A\", \"B\", \"C\", or \"D\".\n6. If the answer is missing or unclear, return {\"answer\": null}.\n\n### OUTPUT TEMPLATE\n{\"answer\": \"<A | B | C | D or null>\"}", | |
| "metrics": [ | |
| { | |
| "metric_id": "knowledge_answer_accuracy", | |
| "type": "categorical_check", | |
| "target_key": "answer", | |
| "weight": 100, | |
| "params": { | |
| "gt_value": "C" | |
| } | |
| } | |
| ] | |
| } | |
| } | |