Metis-HOME: Hybrid Optimized Mixture-of-Experts for Multimodal Reasoning
On the map: LLM reasoning → Multimodal reasoning
also in 1 more
Abstract (arXiv)
Inspired by recent advancements in LLM reasoning, the field of multimodal reasoning has seen remarkable progress, achieving significant performance gains on intricate tasks such as mathematical problem-solving. Despite this progress, current multimodal large reasoning models exhibit two key limitations. They tend to employ computationally expensive reasoning even for simple queries, leading to inefficiency. Furthermore, this focus on specialized reasoning often impairs their broader, more general understanding capabilities. In this paper, we propose Metis-HOME: a Hybrid Optimized Mixture-of-Experts framework designed to address this trade-off. Metis-HOME enables a ''Hybrid Thinking'' paradigm by structuring the original dense model into two distinct expert branches: a thinking branch tailored for complex, multi-step reasoning, and a non-thinking branch optimized for rapid, direct inference on tasks like general VQA and OCR. A lightweight, trainable router dynamically allocates queries to the most suitable expert. We instantiate Metis-HOME by adapting the Qwen2.5-VL-7B into an MoE architecture. Comprehensive evaluations reveal that our approach not only substantially enhances complex reasoning abilities but also improves the model's general capabilities, reversing the degradation trend observed in other reasoning-specialized models. Our work establishes a new paradigm for building powerful and versatile MLLMs, effectively resolving the prevalent reasoning-vs-generalization dilemma. Code and weights are available at https://github.com/MM-Thinking/Metis-HOME.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Multimodal reasoning, one of 17 directions in LLM reasoning. Builds on 34 of its 50 references that are in the map. First posted on arXiv 23 Oct 2025.
Builds on
34 of 50 references.
Closest ancestors · 14
Same direction, newest first
- 2025DeepSketcher: Internalizing Visual Manipulation for Multimodal Reasoning30 Sep
- 2025Breaking the SFT Plateau: Multimodal Structured Reinforcement Learning for Chart-to-Code Generation19 Aug
- 2025Chart-R1: Chain-of-Thought Supervision and Reinforcement for Advanced Chart Reasoner21 Jul
- 2025Metis-RISE: RL Incentivizes and SFT Enhances Multimodal Reasoning Model Learning16 Jun
- 2025SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models10 Apr
- 2025VL-Rethinker: Incentivizing Self-Reflection of Vision-Language Models with Reinforcement Learning10 Apr
- 2025VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model10 Apr
- 2025Kimi-VL Technical Report10 Apr
- 2025R1-Onevision: Advancing Generalized Multimodal Reasoning through Cross-Modal Formalization13 Mar
- 2025LMM-R1: Empowering 3B LMMs with Strong Reasoning Abilities Through Two-Stage Rule-Based RL10 Mar
- 2025MM-Eureka: Exploring the Frontiers of Multimodal Reasoning with Rule-based Reinforcement Learning10 Mar
- 2024LogicVista: Multimodal LLM Logical Reasoning Benchmark in Visual Contexts6 Jul
2 more
Other references · 20
- 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 8,851
- 2025Qwen3 Technical Reportcited by 8,195
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
- 2025Qwen2.5-VL Technical Reportcited by 5,418
- 2024GPT-4o System Cardcited by 4,467
- 2025DAPO: An Open-Source LLM Reinforcement Learning System at Scalecited by 2,678
- 2025InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Modelscited by 1,769
- 2024OpenAI o1 System Cardcited by 1,220
- 2016A Diagram Is Worth A Dozen Imagescited by 793
- 2025VAPO: Efficient and Reliable Reinforcement Learning for Advanced Reasoning Taskscited by 281
- 2023How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Compositioncited by 222
- 2025Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learningcited by 159
8 more
- 2025AdaptThink: Reasoning Models Can Learn When to Thinkcited by 141
- 2025On the Generalization of SFT: A Reinforcement Learning Perspective with Reward Rectificationcited by 121
- 2025AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learningcited by 86
- 2024LLaVA-MoLE: Sparse Mixture of LoRA Experts for Mitigating Data Conflicts in Instruction Finetuning MLLMscited by 84
- 2025Think Only When You Need with Large Hybrid-Reasoning Modelscited by 64
- 2025Learning When to Think: Shaping Adaptive Reasoning in R1-Style Models via Multi-Stage RLcited by 59
- 2025Igniting VLMs toward the Embodied Spacecited by 53
- 2025OmniActor: A Generalist GUI and Embodied Agent for 2D&3D Worldscited by 4
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2025Towards Efficient Multimodal Unified Reasoning Model via Model Merging
- 2025AStar: Boosting Multimodal Reasoning with Automated Structured Thinking
- 2025Why Reasoning Matters? A Survey of Advancements in Multimodal Reasoning (v1)
- 2025Dynamic Experts Search: Enhancing Reasoning in Mixture-of-Experts LLMs at Test Time
- 2025Can Atomic Step Decomposition Enhance the Self-structured Reasoning of Multimodal Large Models?
- 2025MathSE: Improving Multimodal Mathematical Reasoning via Self-Evolving Iterative Reflection and Reward-Guided Fine-Tuning
- 2024MAmmoTH-VL: Eliciting Multimodal Reasoning with Instruction Tuning at Scale
- 2025Perception, Reason, Think, and Plan: A Survey on Large Multimodal Reasoning Models