# Asynchronous Is Nearly Free for Evolution Strategies on Long-Horizon Agentic Tasks

arXiv: https://arxiv.org/abs/2610.04196 · HTML: https://mlsanity.com/paper/2610.04196/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use.
- Builds on 18 of its 18 references that are in the map.
- First posted on arXiv 3 Oct 2026.

## Placed in

- LLM agents and tool use / [Reinforcement learning for LLM agents](https://mlsanity.com/agents/reinforcement-learning-for-llm-agents/)

## Builds on (most cited first, 18 of 18)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Asynchronous Methods for Deep Reinforcement Learning](https://mlsanity.com/paper/1602.01783/) (2016)
- [IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures](https://mlsanity.com/paper/1802.01561/) (2018)
- [Evolution Strategies as a Scalable Alternative to Reinforcement Learning](https://mlsanity.com/paper/1703.03864/) (2017)
- [AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning](https://mlsanity.com/paper/2505.24298/) (2025)
- [Beyond Ten Turns: Unlocking Long-Horizon Agentic Search with Large-Scale Asynchronous RL](https://mlsanity.com/paper/2508.07976/) (2025)
- [Massively Parallel Methods for Deep Reinforcement Learning](https://mlsanity.com/paper/1507.04296/) (2015)
- [Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models](https://mlsanity.com/paper/2410.18252/) (2024)
- [Endless Terminals: Scaling RL Environments for Terminal Agents](https://mlsanity.com/paper/2601.16443/) (2026)
- [Prosperity before Collapse: How Far Can Off-Policy RL Reach with Stale Data on LLMs?](https://mlsanity.com/paper/2510.01161/) (2025)
- [AsyncFlow: An Asynchronous Streaming RL Framework for Efficient LLM Post-Training](https://mlsanity.com/paper/2507.01663/) (2025)
- [Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning](https://mlsanity.com/paper/2607.07508/) (2026)
- [Evolution Strategies at Scale: LLM Fine-Tuning Beyond Reinforcement Learning](https://mlsanity.com/paper/2509.24372/) (2025)
- [Part II: ROLL Flash -- Accelerating RLVR and Agentic Training with Asynchrony](https://mlsanity.com/paper/2510.11345/) (2025)
- [Matching Accuracy, Different Geometry: Evolution Strategies vs GRPO in LLM Post-Training](https://mlsanity.com/paper/2604.01499/) (2026)
- [Evolutionary Strategies lead to Catastrophic Forgetting in LLMs](https://mlsanity.com/paper/2601.20861/) (2026)
- [Overcoming Forgetting in LLM Fine-Tuning with Evolution Strategies](https://mlsanity.com/paper/2605.30148/) (2026)
- [Agentic ESOpt: Fine-Tuning Long-Horizon LLM Agents with Minimal GPU Requirements](https://mlsanity.com/paper/2608.17310/) (2026)

## Abstract (arXiv)

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 substantially, and synchronous update barriers leave faster workers idling, wasting compute resources. Asynchronous reinforcement learning addresses this inefficiency but introduces policy lag and off-policy optimization. Evolution strategies (ES) are a backpropagation-free alternative to gradient-based training, yet require many perturbed-policy evaluations and have remained largely synchronous in current LLM post-training. We introduce bounded-staleness asynchronous ES and investigate its training behavior under stale fitness measurements. On WebShop, we systematically vary the lag and fraction of stale fitness measurements and additionally introduce two stress tests to probe extreme staleness cases. Natural asynchrony closely tracks synchronous ES, and degradation in evaluation accuracy remains small even under extreme staleness, including when all fitness measurements are stale. These findings suggest that ES can tolerate substantial policy staleness without explicit off-policy correction, supporting its use in asynchronous post-training for long-horizon LLM agents.