Code

MixEval-X Image / Video#434

Code

Merged

Luodian merged 10 commits intomainEvolvingLMMs-Lab/lmms-eval:mainfrom pufanyi/mix_eval_xEvolvingLMMs-Lab/lmms-eval:pufanyi/mix_eval_xCopy head branch name to clipboardDec 3, 2024

+1,036-332

Commits

Commits on Nov 30, 2024

• reformat mix eval

  pufanyi

  committed
  1fad5d9

  Copy full SHA for 1fad5d9

• video2text fix

  pufanyi

  committed
  715df25

  Copy full SHA for 715df25

Commits on Dec 1, 2024

• video

  pufanyi

  committed
  cc232f7

  Copy full SHA for cc232f7

• Add image-to-text evaluation tasks and templates

  pufanyi

  committed
  b89847a

  Copy full SHA for b89847a

• Refactor image-to-text evaluation tasks and update configurations

  pufanyi

  committed
  5e23413

  Copy full SHA for 5e23413

• Enhance LlamaVision model with video loading improvements and configuration updates

  pufanyi

  committed
  4fddfff

  Copy full SHA for 4fddfff

• fix internvl2

  pufanyi

  committed
  b4ccbe1

  Copy full SHA for b4ccbe1

• llava vid default numframe too small

  pufanyi

  committed
  aa5c5bf

  Copy full SHA for aa5c5bf

Commits on Dec 2, 2024

• add max_new_tokens

  pufanyi

  committed
  11540d1

  Copy full SHA for 11540d1

• remove "with a short phrase" for gpt4v

  pufanyi

  committed
  57f8c43

  Copy full SHA for 57f8c43