Language Models Can See Better: Visual Contrastive Decoding for LLM Multimodal Reasoning
CoRR（2025）
Key words
contrastive-example decoding,output distributions,In-Context Learning
AI Read Science
Must-Reading Tree
Example
Generate MRT to find the research sequence of this paper
Chat Paper
Summary is being generated by the instructions you defined