4 papers
Asynchrony Is Nearly Free for Evolution Strategies on Long-Horizon LLM Agentic Tasks
William Hoy, Jingxuan Fan, Nurcin Celik +1
Synchronous policy updates can limit the efficiency of post-training for long-horizon LLM agents. When trajectories span multiple interaction turns, completion times can vary subst…
Continuous Context Management
William Hoy, Jingxuan Fan, Nurcin Celik +1
Long-horizon large language model (LLM) agents commonly retain their complete interaction history until compaction is triggered at a predefined threshold. We study Continuous Conte…
Matching Accuracy, Different Geometry: Evolution Strategies vs GRPO in LLM Post-Training
William Hoy, Binxu Wang, Xu Pan
Evolution Strategies (ES) have emerged as a scalable gradient-free alternative to reinforcement learning based LLM fine-tuning, but it remains unclear whether comparable task perfo…
STABLE: Gated Continual Learning for Large Language Models
William Hoy, Nurcin Celik
Large language models (LLMs) increasingly require mechanisms for continual adaptation without full retraining. However, sequential updates can lead to catastrophic forgetting, wher…