fix: Passing empty list to images/videos for some multi-modal models

[HollowMan6]

What does this PR do?

Alternative root fix for verl-project/verl#3281

For example, this PR fixes the following error when we pass an empty list of images to Qwen2.5-VL:

  File "torchdata/stateful_dataloader/worker.py", line 242, in _worker_loop
    data = fetcher.fetch(index)  # type: ignore[union-attr]
  File "torch/utils/data/_utils/fetch.py", line 52, in fetch
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "torch/utils/data/_utils/fetch.py", line 52, in <listcomp>
    data = [self.dataset[idx] for idx in possibly_batched_index]
  File "verl/utils/dataset/rl_dataset.py", line 248, in __getitem__
    model_inputs = self.processor(text=[raw_prompt], images=images, videos=videos, return_tensors="pt")
  File "transformers/models/qwen2_5_vl/processing_qwen2_5_vl.py", line 150, in __call__
    image_inputs = self.image_processor(images=images, **output_kwargs["images_kwargs"])
  File "transformers/image_processing_utils_fast.py", line 637, in __call__
    return self.preprocess(images, *args, **kwargs)
  File "transformers/models/qwen2_vl/image_processing_qwen2_vl_fast.py", line 151, in preprocess
    return super().preprocess(images, videos, **kwargs)
  File "transformers/image_processing_utils_fast.py", line 662, in preprocess
    return self._preprocess_image_like_inputs(
  File "transformers/models/qwen2_vl/image_processing_qwen2_vl_fast.py", line 173, in _preprocess_image_like_inputs
    batch_feature = self._preprocess(images, **kwargs)
  File "transformers/models/qwen2_vl/image_processing_qwen2_vl_fast.py", line 211, in _preprocess
    grouped_images, grouped_images_index = group_images_by_shape(images, disable_grouping=disable_grouping)
  File "transformers/image_transforms.py", line 917, in group_images_by_shape
    device = images[0][0].device if is_nested else images[0].device
IndexError: list index out of range

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline,
  Pull Request section?
• Was this discussed/approved via a Github issue or the forum? Please add a link
  to it if that's the case.
• Did you make sure to update the documentation with your changes? Here are the
  documentation guidelines, and
  here are tips on formatting docstrings.
• Did you write any new necessary tests?

Who can review?

Anyone in the community is free to review the PR once the tests have passed. Feel free to tag
members/contributors who may be interested in your PR.

@ArthurZucker, @amyeroberts, @qubvel

[zucchini-nlp]

Thanks for the PR @doubao2021-ai !

We had another one in progress here (#36682) which is more complete and fixes all models. I'd prefer that one to be merged

[HollowMan6]

Hi @zucchini-nlp, thanks for your reply! I think #36682 doesn't consider the situation when there's no image at all and batch size is 1, i.e., images = []. Feel free to take over this PR as I don't have the bandwidth to fix all models. Thanks in advance!

[zucchini-nlp]

@HollowMan6 if there are no images at all, it is recommended to simply pass images=None which is the default for no input

[HollowMan6]

Yeah, that's what I had proposed in verl-project/verl#3281, but maybe it can be good as well to enforce some checks at the transformers library side.

[github-actions]

[For maintainers] Suggested jobs to run (before merge)

run-slow: glm4v, llama4, qwen2_vl, qwen3_vl

[zucchini-nlp]

Hey @HollowMan6 , the feature was added in transformers already in one of the old PRs. I think you can close it now. Passing empty lists for images should work images = [[im1, im2], [], [im3]]

[HollowMan6]

Oh okay, good to know! Thanks!