Paper summary
Paper summary: LLaVA-CoT: Let Vision Language Models Reason Step-by-Step
LLaVA-CoT trains a vision-language model to reason in explicit stages (summary, visual interpretation, logical reasoning, conclusion) and adds a stage-wise retracing search at test time.
- Rather than prompting for "think step by step", the model is trained to move through four fixed stages on its own.
- A 100k-sample dataset with structured reasoning annotations makes this possible.