dataset_path: m-a-p/SimpleVQA
dataset_kwargs:
  token: false
output_type: generate_until
doc_to_visual: !function utils.simplevqa_doc_to_visual
doc_to_text: !function utils.simplevqa_doc_to_text
doc_to_target: "answer"
process_results: !function utils.simplevqa_process_results
generation_kwargs:
  max_new_tokens: 32
  temperature: 0
  top_p: 1.0
  num_beams: 1
  do_sample: false
metric_list:
  - metric: exact_match
    aggregation: mean
    higher_is_better: true
lmms_eval_specific_kwargs:
  default:
    pre_prompt: ""
    post_prompt: "\nAnswer the question using a short phrase."
  qwen3_vl:
    pre_prompt: "Question: "
    post_prompt: "\nAnswer with a short phrase only."
metadata:
  - version: 0.0
