metadata: 2026-07-07 revised taxonomy (task_type format vocabulary multiple_choice/yes_no/short_answer/free_form; score_type incl. none; re-verified protocol evidence; repaired prompt provenance)

metadata.json

@@ -9,11 +9,11 @@
       "modalities": [
         "single_image_start"
       ],
-      "task_type": "multiple_choice_vqa",
+      "task_type": "multiple_choice",
       "score_type": "rule",
       "score_protocol": {
-        "reference": "official We-Math/We-Math@README + evaluation/four_dimensional_metrics.py & evaluation/accuracy.py — 'we use string matching to directly extract answers, which eliminates the high cost of using additional models for further answer extraction' (rule-based option-letter extraction, exact compare).",
-        "note": "Official headline is the four-dimensional metric set (IK/IG/CM/RM) plus step-wise accuracy, computed by grouping one-step sub-problems with their multi-step composite problem (problem_id/step_key, vendored in VLMEvalKit vlmeval/dataset/utils/wemath.py:16-238); a per-sample accuracy scorer will not reproduce those group scores (grouping keys ARE shipped in extra). VLMEvalKit grades WeMath via mcq_vanilla_eval (utils/multiple_choice.py:475, rule can_infer + GPT extraction fallback), i.e. llm_extract — disagrees with the official rule-only protocol; official preferred."
+        "reference": "official We-Math/We-Math@README + evaluation/four_dimensional_metrics.py & evaluation/accuracy.py — 'we use string matching to directly extract answers, which eliminates the high cost of using additional models for further answer extraction' (rule-based option-letter extraction, exact compare). [Quote re-verified verbatim in the official README 2026-07-07; rule pipeline re-verified vendored at VLMEvalKit@vlmeval/dataset/utils/wemath.py:28-41 (split on 'Answer', strip, first char in A-H, compare).]",
+        "note": "Official headline is the four-dimensional metric set (IK/IG/CM/RM) plus step-wise accuracy, computed by grouping one-step sub-problems with their multi-step composite problem (problem_id/step_key, vendored in VLMEvalKit vlmeval/dataset/utils/wemath.py:16-238); a per-sample accuracy scorer will not reproduce those group scores (grouping keys ARE shipped in extra). VLMEvalKit grades WeMath via mcq_vanilla_eval (image_mcq.py WeMath.evaluate -> utils/multiple_choice.py mcq_vanilla_eval, rule can_infer + GPT extraction fallback), i.e. llm_extract — disagrees with the official rule-only protocol; official preferred."
       },
       "prompt_template": "<image>{{ question }}\nOptions:\n{% for k, v in options.items() %}{{ k }}. {{ v }}{% if not loop.last %}\n{% endif %}{% endfor %}\nPlease select the correct answer from the options above. \n",
       "mapping_from_source": {