[LLaVa] Some improvements (#27895)

* More improvements

* Improve variable names

* Update READMEs, improve docs

docs/source/en/index.md

@@ -169,7 +169,7 @@ Flax), PyTorch, and/or TensorFlow.
 |                          [LiLT](model_doc/lilt)                          |       ✅        |         ❌         |      ❌      |
 |                         [LLaMA](model_doc/llama)                         |       ✅        |         ❌         |      ✅      |
 |                        [Llama2](model_doc/llama2)                        |       ✅        |         ❌         |      ✅      |
-|                         [Llava](model_doc/llava)                         |       ✅        |         ❌         |      ❌      |
+|                         [LLaVa](model_doc/llava)                         |       ✅        |         ❌         |      ❌      |
 |                    [Longformer](model_doc/longformer)                    |       ✅        |         ✅         |      ❌      |
 |                        [LongT5](model_doc/longt5)                        |       ✅        |         ❌         |      ✅      |
 |                          [LUKE](model_doc/luke)                          |       ✅        |         ❌         |      ❌      |

docs/source/en/model_doc/llava.md

@@ -14,19 +14,27 @@ rendered properly in your Markdown viewer.
 
 -->
 
-# Llava
+# LLaVa
 
 ## Overview
 
-Llava is an open-source chatbot trained by fine-tuning LlamA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture. In other words, it is an multi-modal version of LLMs fine-tuned for chat / instructions.
+LLaVa is an open-source chatbot trained by fine-tuning LlamA/Vicuna on GPT-generated multimodal instruction-following data. It is an auto-regressive language model, based on the transformer architecture. In other words, it is an multi-modal version of LLMs fine-tuned for chat / instructions.
 
-The Llava model was proposed in [Improved Baselines with Visual Instruction Tuning](https://arxiv.org/pdf/2310.03744) by Haotian Liu, Chunyuan Li, Yuheng Li and Yong Jae Lee.
+The LLaVa model was proposed in [Visual Instruction Tuning](https://arxiv.org/abs/2304.08485) and improved in [Improved Baselines with Visual Instruction Tuning](https://arxiv.org/pdf/2310.03744) by Haotian Liu, Chunyuan Li, Yuheng Li and Yong Jae Lee.
 
 The abstract from the paper is the following:
 
 *Large multimodal models (LMM) have recently shown encouraging progress with visual instruction tuning. In this note, we show that the fully-connected vision-language cross-modal connector in LLaVA is surprisingly powerful and data-efficient. With simple modifications to LLaVA, namely, using CLIP-ViT-L-336px with an MLP projection and adding academic-task-oriented VQA data with simple response formatting prompts, we establish stronger baselines that achieve state-of-the-art across 11 benchmarks. Our final 13B checkpoint uses merely 1.2M publicly available data, and finishes full training in ∼1 day on a single 8-A100 node. We hope this can make state-of-the-art LMM research more accessible. Code and model will be publicly available*
 
-Tips:
+<img src="https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/model_doc/llava_architecture.jpg"
+alt="drawing" width="600"/>
+
+<small> LLaVa architecture. Taken from the <a href="https://arxiv.org/abs/2304.08485">original paper.</a> </small>
+
+This model was contributed by [ArthurZ](https://huggingface.co/ArthurZ) and [ybelkada](https://huggingface.co/ybelkada).
+The original code can be found [here](https://github.com/haotian-liu/LLaVA/tree/main/llava).
+
+## Usage tips
 
 - We advise users to use `padding_side="left"` when computing batched generation as it leads to more accurate results. Simply make sure to call `processor.tokenizer.padding_side = "left"` before generating.
 
@@ -44,15 +52,20 @@ For multiple turns conversation:
 "USER: <image>\n<prompt1>ASSISTANT: <answer1>USER: <prompt2>ASSISTANT: <answer2>USER: <prompt3>ASSISTANT:"
 ```
 
-This model was contributed by [ArthurZ](https://huggingface.co/ArthurZ) and [ybelkada](https://huggingface.co/ybelkada).
-The original code can be found [here](https://github.com/haotian-liu/LLaVA/tree/main/llava).
-
-Check out also this [Google Colab demo](https://colab.research.google.com/drive/1qsl6cd2c8gGtEW1xV5io7S8NHh-Cp1TV?usp=sharing) on how to run Llava on a free-tier Google colab instance.
-
 ### Using Flash Attention 2
 
 Flash Attention 2 is an even faster, optimized version of the previous optimization, please refer to the [Flash Attention 2 section of performance docs](https://huggingface.co/docs/transformers/perf_infer_gpu_one).
 
+## Resources
+
+A list of official Hugging Face and community (indicated by 🌎) resources to help you get started with BEiT.
+
+<PipelineTag pipeline="image-to-text"/>
+
+- A [Google Colab demo](https://colab.research.google.com/drive/1qsl6cd2c8gGtEW1xV5io7S8NHh-Cp1TV?usp=sharing) on how to run Llava on a free-tier Google colab instance leveraging 4-bit inference.
+- A [similar notebook](https://github.com/NielsRogge/Transformers-Tutorials/blob/master/LLaVa/Inference_with_LLaVa_for_multimodal_generation.ipynb) showcasing batched inference. 🌎
+
+
 ## LlavaConfig
 
 [[autodoc]] LlavaConfig