dataset_path: etri-vilab/holisafe-bench
dataset_kwargs:
  token: True
test_split: test
output_type: generate_until
doc_to_visual: !function utils.holisafe_doc_to_visual
doc_to_text: !function utils.holisafe_doc_to_text
doc_to_target: "{{type}}"
process_results: !function utils.holisafe_process_results
generation_kwargs:
  max_new_tokens: 512
  temperature: 0.0
  top_p: 1.0
  num_beams: 1
  do_sample: false
lmms_eval_specific_kwargs:
  default:
    pre_prompt: ""
    post_prompt: ""
metric_list:
  - metric: holisafe_avg_asr
    aggregation: !function utils.holisafe_aggregate_avg_asr
    higher_is_better: false
  - metric: holisafe_rr
    aggregation: !function utils.holisafe_aggregate_rr
    higher_is_better: false
  - metric: holisafe_asr_ssu
    aggregation: !function utils.holisafe_aggregate_asr_ssu
    higher_is_better: false
  - metric: holisafe_asr_usu
    aggregation: !function utils.holisafe_aggregate_asr_usu
    higher_is_better: false
  - metric: holisafe_asr_uuu
    aggregation: !function utils.holisafe_aggregate_asr_uuu
    higher_is_better: false
  - metric: holisafe_asr_suu
    aggregation: !function utils.holisafe_aggregate_asr_suu
    higher_is_better: false
