4 papers
One Frozen Simulator Is Not Enough: Simulator Collapse in Multi-Agent RL
Simon Yu, Nicholas Tomlin, Marwa Abdulhai +7
Multi-agent reinforcement learning for human-AI interaction typically relies on a single large language model to simulate user behavior. We show that this approach systematically f…
Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces
Simon Yu, Derek Chong, Ananjan Nandi +4
As LLM agent systems take on more complex tasks, they increasingly rely on meta-agents: higher-order agents that create, operate on and manage other agents. Meta-agent operations s…
Composing Policy Gradients and Prompt Optimization for Language Model Programs
Noah Ziems, Dilara Soylu, Lakshya A Agrawal +10
Group Relative Policy Optimization (GRPO) has proven to be an effective tool for post-training language models (LMs). However, AI systems are increasingly expressed as modular prog…
GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
Lakshya A Agrawal, Shangyin Tan, Dilara Soylu +14
Large language models (LLMs) are increasingly adapted to downstream tasks via reinforcement learning (RL) methods like Group Relative Policy Optimization (GRPO), which often requir…