Popular repositories Loading

1. mmrag-toolkit mmrag-toolkit Public

   A minimal toolkit for Multimodal RAG — retrieve images and text, ground answers with vision-language models

   Python

2. vlm-faithfulness vlm-faithfulness Public

   Probing and evaluating factual faithfulness in Vision-Language Models — hallucination detection, POPE-style eval, and contrastive decoding

   Python

3. visual-scratchpad visual-scratchpad Public

   Visual chain-of-thought reasoning via a step-by-step visual working memory — decompose, operate, then answer

   Python

4. sgxs2014 sgxs2014 Public

   My GitHub profile README