| { |
| "meta_info": { |
| "case_id": "SEED-V-05.cnt", |
| "bench_subset": "NeuroBench-Core", |
| "difficult": 2.0, |
| "original_dataset": "SEED-V" |
| }, |
| "agent_input": { |
| "data_path": "data/core/SEED-V-05.cnt", |
| "instruction": "Please select standard EEG channels and rank alpha relative power among the delta, theta, alpha, beta, and gamma bands. Compute the global rank from the mean relative power across all standard EEG channels, then compute the occipital rank from the mean relative power across occipital channels only. Each rank must be an integer from 1 to 5, where 1 is the highest and 5 is the lowest. Use Welch PSD estimation. Define relative power as the integrated power in the target frequency band divided by the integrated total power from 0.5 to 45 Hz. Determine whether the global average metric masks regional differences, and briefly explain your conclusion." |
| }, |
| "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 two alpha relative power ranks from the agent report:\n1) global alpha relative power rank\n2) occipital alpha relative power rank\n\n### STRICT CONSTRAINTS (MANDATORY)\n1. Output ONLY a valid JSON object.\n2. DO NOT include Markdown code blocks.\n3. DO NOT include any conversational text, explanations, or units.\n4. Keys must be EXACTLY \"global_alpha_rank\" and \"occipital_alpha_rank\".\n5. Each value must be an integer from 1 to 5, or null if missing.\n\n### OUTPUT TEMPLATE\n{\"global_alpha_rank\": <int|null>, \"occipital_alpha_rank\": <int|null>}", |
| "metrics": [ |
| { |
| "metric_id": "global_alpha_rank_accuracy", |
| "type": "categorical_check", |
| "target_key": "global_alpha_rank", |
| "weight": 30, |
| "params": { |
| "gt_value": 3 |
| } |
| }, |
| { |
| "metric_id": "occipital_alpha_rank_accuracy", |
| "type": "categorical_check", |
| "target_key": "occipital_alpha_rank", |
| "weight": 30, |
| "params": { |
| "gt_value": 3 |
| } |
| }, |
| { |
| "metric_id": "global_occipital_masking_consistency", |
| "type": "semantic_consistency", |
| "target_key": "full_text", |
| "weight": 40, |
| "params": { |
| "judge_prompt": "You are judging semantic consistency for an EEG task comparing global and occipital alpha relative power ranks. The true conclusion is: the global average metric does not mask regional indicator differences. Use the stated true conclusion as the ground truth; do not infer a different conclusion. Mark the response as correct only if its explanation matches this true conclusion. Mark it as incorrect if it claims masking when the true conclusion says no masking, or claims no masking when the true conclusion says masking. Output only a valid JSON object with exactly one key named status and a boolean value True or False." |
| } |
| } |
| ] |
| } |
| } |
