Research Focus
At MiroMind, he works on large language models, multimodal AI systems, and agentic frameworks — building AI systems that push the boundaries of reasoning.
Active Topics
- Agentic AI & Reasoning Models
- How can we build AI agents that reason effectively and interact with complex environments?
- What architectures enable robust multi-step reasoning in LLMs?
Projects
- MiroFlow — Agent framework developed at MiroMind (2.5k+ GitHub stars). github.com/MiroMindAI/MiroFlow
- MiroThinker — Agentic model developed at MiroMind (1.5k GitHub stars). github.com/MiroMindAI/MiroThinker
- MiroMind-M1 — Foundation reasoning model developed at MiroMind. github.com/MiroMindAI/MiroMind-M1
Publications
- MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling — Technical Report 2025
- MiroFlow: Towards High-Performance, Robust, and Open-Source Reproducible Agent Framework for General Research — Technical Report 2025
- MiroMind-M1: An Open-Source Advancement in Mathematical Reasoning via Context-Aware Multi-Stage Policy Optimization — Technical Report 2025
- 100 Days After DeepSeek-R1: A Survey on Replication Studies and More Directions for Reasoning Language Models — Technical Review 2025