dataset_path: MME-Benchmarks/Video-MME-v2
dataset_kwargs:
  token: True
  cache_dir: videomme_v2
  video: True
test_split: test
output_type: generate_until
doc_to_visual: !function utils.videomme_v2_doc_to_visual
doc_to_text: !function utils.videomme_v2_doc_to_text
doc_to_target: "answer"
generation_kwargs:
  max_new_tokens: 64
  temperature: 0
  top_p: 1.0
  num_beams: 1
  do_sample: false
process_results: !function utils.videomme_v2_process_results
metric_list:
  - metric: videomme_v2_overall_score
    aggregation: !function utils.videomme_v2_aggregate_overall_score
    higher_is_better: true
  - metric: videomme_v2_relevance_score
    aggregation: !function utils.videomme_v2_aggregate_relevance_score
    higher_is_better: true
  - metric: videomme_v2_logic_score
    aggregation: !function utils.videomme_v2_aggregate_logic_score
    higher_is_better: true
  - metric: videomme_v2_level_1_score
    aggregation: !function utils.videomme_v2_aggregate_level_1_score
    higher_is_better: true
  - metric: videomme_v2_level_2_score
    aggregation: !function utils.videomme_v2_aggregate_level_2_score
    higher_is_better: true
  - metric: videomme_v2_level_3_score
    aggregation: !function utils.videomme_v2_aggregate_level_3_score
    higher_is_better: true
  - metric: videomme_v2_overall_acc
    aggregation: !function utils.videomme_v2_aggregate_overall_acc
    higher_is_better: true
  - metric: videomme_v2_relevance_acc
    aggregation: !function utils.videomme_v2_aggregate_relevance_acc
    higher_is_better: true
  - metric: videomme_v2_logic_acc
    aggregation: !function utils.videomme_v2_aggregate_logic_acc
    higher_is_better: true
  - metric: videomme_v2_level_1_acc
    aggregation: !function utils.videomme_v2_aggregate_level_1_acc
    higher_is_better: true
  - metric: videomme_v2_level_2_acc
    aggregation: !function utils.videomme_v2_aggregate_level_2_acc
    higher_is_better: true
  - metric: videomme_v2_level_3_acc
    aggregation: !function utils.videomme_v2_aggregate_level_3_acc
    higher_is_better: true
lmms_eval_specific_kwargs:
  default:
    pre_prompt: ""
    post_prompt: "\nAnswer with the option's letter from the given choices directly."
  qwen3_vl:
    format: "qwen3_vl"
    pre_prompt: "Question: "
    post_prompt: "Answer with the option letter only."
metadata:
  - version: 0.0
