{"slug": "remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous", "title": "Remember, Don't Re-read: Stateful ReAct Agents for Token-Efficient Autonomous Experimentation", "summary": "Researchers reformulated the autoresearch pattern as a stateful ReAct agent using LangGraph, achieving 90% fewer tokens on hyperparameter tuning and 52% fewer on code optimization compared to stateless designs. The stateful agent operates within a fixed-size conversation window, reducing token costs from O(n) to O(1) per iteration while maintaining optimization quality.", "body_md": "arXiv:2606.14945v1 Announce Type: new\nAbstract: The autoresearch pattern enables autonomous experimentation by having a large language model (LLM) iteratively modify code to optimize a target metric. Its stateless design, however, reconstructs experimental context from scratch at every iteration, incurring $O(n)$ token cost per iteration and $O(n^{2})$ total. This work reformulates the pattern as a stateful ReAct agent using LangGraph, where typed persistent state carries experimental history across iterations via a tool-calling interface. Two benchmarks are evaluated: hyperparameter tuning (15 iterations, small per-iteration observations) and code performance optimization (40 iterations, large per-iteration observations containing full source code and benchmark results). On hyperparameter tuning, the stateful agent consumes 90\\% fewer tokens (2{,}492 vs.\\ 24{,}465). On code optimization, the stateful agent consumes 52\\% fewer tokens (627K vs.\\ 1{,}275K) while achieving comparable optimization quality on both tasks. The token reduction is structural: the stateless agent re-reads the full history at $O(n)$ cost per iteration, while the stateful agent operates within a fixed-size conversation window at $O(1)$ cost. This paper describes the architecture in sufficient detail for practitioners to implement a stateful autoresearch agent for their own workflows.", "url": "https://wpnews.pro/news/remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous", "canonical_source": "https://arxiv.org/abs/2606.14945", "published_at": "2026-06-16 04:00:00+00:00", "updated_at": "2026-06-16 04:26:13.739216+00:00", "lang": "en", "topics": ["large-language-models", "ai-agents", "ai-research"], "entities": ["LangGraph", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous", "markdown": "https://wpnews.pro/news/remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous.md", "text": "https://wpnews.pro/news/remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous.txt", "jsonld": "https://wpnews.pro/news/remember-don-t-re-read-stateful-react-agents-for-token-efficient-autonomous.jsonld"}}