Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning
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
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- PossiblePossibly related (embedding) · 54%Atomic-man007/Awesome_Multimodel_LLM →
- LinkedLinked via arxiv author · 85%Yiyang Fang →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Pei Fu →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Jinjie Li →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Jian Liang →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Wenke Huang →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Ruijie Luo →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”
- LinkedLinked via arxiv author · 85%Shaojie Zhang →
“Switch-Reasoner: Learn When to Think in Multitask Mixtures via Reinforcement Learning”