ZachSun

/

Qwen2.5-gfn-3B

like 0

This PR significantly enhances the model card for the LaCoT model by adding the following information, all extracted directly from the paper and its associated GitHub repository:

• A clear introduction and link to the paper: Latent Chain-of-Thought for Visual Reasoning.
• The full abstract of the paper.
• A link to the official GitHub repository: https://github.com/heliossun/LaCoT.
• Relevant metadata tags: pipeline_tag: image-text-to-text and library_name: transformers. The library_name is supported by evidence from config.json, tokenizer_config.json, and installation instructions in the GitHub README.
• Links to various model checkpoints mentioned in the GitHub README.
• Detailed data preparation and installation instructions from the GitHub README.
• Guidance on training and evaluation, linking to relevant scripts in the GitHub repository.

Please note that a license was not added as no explicit license information was found in the provided context, adhering strictly to the documentation guidelines. Similarly, no specific inference code snippet was added to the "Sample Usage" section, as none was found directly in the GitHub README, as per the explicit instruction to avoid making up code.