Simon Yu

I am a 2nd year PhD student at Northeastern University, advised by Weiyan Shi. My research goal is to build self-improving agent systems via self-play and interaction with real-world feedback. I closely work with Chris Manning from Stanford and Natasha Jaques from UW. I am currently interning at MSR Redmond with Baolin Peng and Jianfeng Gao, working on AI for AI and self-improvement. Before that, I interned at Orby AI, mentored by Peng Qi. I am a main organizer of the NeurIPS 2025 Workshop on Multi-Turn Interactions in LLMs. I was a M.Res. and BSc student at the University of Edinburgh, also part of the EdinburghNLP, and I am part of the Cohere For AI community working with Marzieh Fadaee.

I work toward this goal from three angles:

1. Meta-Agents: Shepherd turns an agent’s execution into a reversible, Git-like trace, so meta-agents can inspect, fork, replay, and revert other agents’ runs to supervise, optimize, and train them. Code at shepherd-agents/shepherd.
2. Self-Play: SPIRAL shows that self-play on zero-sum games improves reasoning, with zero human-curated data.
3. Environment Scaling & Continual Learning: scaling what agents learn from and what they keep, including TextArena for multi-agent environments and evaluation, GEM for unified, scalable environment generation, and PolySkill for continually aggregating experience across new domains.

One of the most influential lessons to me is from The Bitter Lesson by Richard Sutton and The Era of Experience by David Silver and Richard Sutton. The idea is not just limited to AI but can be applied to any choice in life. Always choose the path that benefits in the long run, instead of the path that might be easier in the short run.