Raushan Turganbay
Raushan Turganbay
# What does this PR do? As was pointed out in #28932 , StoppingCriteria needs to stop generation per batch element and return a boolean tensor of `batch_size`. This PR...
# What does this PR do? Adds a watermarking technique proposed in this [paper](https://arxiv.org/abs/2301.10226) to `transformers` logits processor. I added only the simple method (algorithm 2 from paper) and the...
# What does this PR do? This PR adds the ability to pass in device when initializing `LogitsProcessors` and is one more step towards `compile` compatibility. ## Who can review?...
# What does this PR do? Small part of the issue #28981 . This PR makes sure that Logits Processor and Stopping Criteria are compatible with `torch.compile` when `fullgraph=True`. The...
# What does this PR do? This PR enables compile for Phi models. Checked the correctness by running speed benchmark script (the results is below) and a test for dynamic...
# What does this PR do? An implementation of quantized cache with `quanto` library. Introduces a new `CacheConfig` to store cache related arguments and a new cache class `QuantoQuantizedCache`. The...
# What does this PR do? Following #30716 this PR makes the return type of Idefics2 models' cache follow the same way, by returning the type we had in the...
# What does this PR do? Fixes #29640 . Add a new model, Video Llava to the library. This is a draft PR, will add more here later. The model...
Hey! Video-LLaVa is now available in the Transformers library! Feel free to check it out [here](https://huggingface.co/LanguageBind/Video-LLaVA-7B-hf). Thanks to @LinB203 for helping to ship the model 🤗 To get the model,...
# What does this PR do? As it was discussed under [this PR](https://github.com/huggingface/transformers/pull/29583), position ids in some models are not calculated/inferred from attn mask in `forward`, which gives incorrect positions...