paper-with-me

홈 › Papers

Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning

2026-07-09 · Yiyang Fang, Pei Fu, Jinjie Li, Jian Liang, Wenke Huang, Ruijie Luo, Shaojie Zhang, Jian Luan, Yi R. Fung, Mang Ye arxiv

Multimodal Large Language Models (MLLMs) often follow a fixed Think-then-Answer paradigm, which is inefficient in heterogeneous multitask settings because simple inputs may not require explicit reasoning while difficult ones can benefit substantially from it. Learning when to think is also unstable during post-training, where imbalanced rollouts can drive the model toward always-thinking or always-direct behavior. We propose Switch-Reasoner, a GRPO-based framework that learns to adaptively select reasoning modes for MLLMs. It treats thinking as a virtual tool invocation and allows the model to either answer directly or invoke explicit reasoning before answering. To stabilize this decision, we introduce a dual-level regulation mechanism that balances the overall use of Thinking Mode and Direct Mode while providing sample-level supervision based on the relative benefit of the two choices. Experiments on 11 multimodal tasks show that Switch-Reasoner reduces unnecessary reasoning while maintaining strong performance, achieving a better accuracy-efficiency trade-off.

📄 PDF Abstract BibTeX arXiv:2607.08572

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

ThinkSwitcher: When to Think Hard, When to Think Fast

2025-05-20 · Guosheng Liang, Longguang Zhong, ZiYi Yang, Xiaojun Quan

Large reasoning models (LRMs) excel at solving complex tasks by leveraging long chain-of-thought (CoT) reasoning. However, this often leads to overthinking on simple tasks, resulting in unnecessary computational overhead…

ProxyThinker: Test-Time Guidance through Small Visual Reasoners

2025-05-30 · Zilin Xiao, Jaywon Koo, Siru Ouyang, Jefferson Hernandez 외

Recent advancements in reinforcement learning with verifiable rewards have pushed the boundaries of the visual reasoning capabilities in large vision-language models (LVLMs). However, training LVLMs with reinforcement fi…

Visual Reasoning

Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning

2026-06-02 · Yu Xia, Zhouhang Xie, Xin Xu, Byungkyu Kang 외 arxiv

Large language models improve final-answer accuracy through extended chain-of-thought reasoning, but often spend tokens inefficiently and offer little inference-time control. Existing efficient reasoning methods control …

Reinforcement Learning

Step Back to Leap Forward: Self-Backtracking for Boosting Reasoning of Language Models

2025-02-06 · Xiao-Wen Yang, Xuan-Yi Zhu, Wen-Da Wei, Ding-Chu Zhang 외

The integration of slow-thinking mechanisms into large language models (LLMs) offers a promising way toward achieving Level 2 AGI Reasoners, as exemplified by systems like OpenAI's o1. However, several significant challe…

Think Then Embed: Generative Context Improves Multimodal Embedding

2025-10-06 · Xuanming Cui, Jianpeng Cheng, Hong-you Chen, Satya Narayan Shukla 외 arxiv

There is a growing interest in Universal Multimodal Embeddings (UME), where models are required to generate task-specific representations. While recent studies show that Multimodal Large Language Models (MLLMs) perform w…