Advancing Multimodal Reasoning: From Optimized Cold Start to Staged Reinforcement Learning

Understand

Inspired by the remarkable reasoning capabilities of Deepseek-R1 in complex textual tasks, many works attempt to incentivize similar capabilities in Multimodal Large Language Models (MLLMs) by directly applying reinforcement learning (RL).

• However, they still struggle to activate complex reasoning.
• In this paper, rather than examining multimodal RL in isolation, we delve into current training pipelines and identify three crucial phenomena: 1) Effective cold start initialization is critical for enhancing MLLM reasoning.
• Intriguingly, we find that initializing with carefully selected text data alone can lead to performance surpassing many recent multimodal reasoning models, even before multimodal RL.