Evolving Agent Teams

A paper submitted to arXiv on 27 Oct 2026 introduces Evolving Agent Teams, a multi-agent framework whose central supervisor diagnoses local optima and restructures agent teams, backed by a persistent file system holding a learning document, evolution history and per-team artifacts. Under an extended token budget, Evolving Agent Teams generated a CUDA kernel from scratch reaching 1.2x mean speedup over the production FlashInfer MLA paged decode kernel and winning on 36 of 47 FlashInfer-Bench production workloads (0.86-0.99x on the remaining 11). The framework also reported 3.01x over vLLM on KDA prefill on AMD MI350X, 3.44x over FlashKDA on NVIDIA B200, 30.3 tokens/s decode for Qwen3.8-Flash-Next in MLX 4-bit on a 48 GB M4 Pro, and 62.1 on held-out RoboDojo robot manipulation seeds versus 55.3 for the best leaderboard entry, with code released on GitHub.

Computer Science Artificial Intelligence Submitted on 27 Oct 2026 Title:Evolving Agent Teams View PDF https://arxiv.org/pdf/2610.36848 HTML experimental https://arxiv.org/html/2610.36848v1 Abstract:The rapid progress of software engineering agents drives interest in applying them to open-ended problems such as GPU kernel optimization. However, we find that existing coding agents converge to local optima on real-world, expert-written kernels e.g., FlashInfer . We introduce Evolving Agent Teams, a multi-agent framework that escapes such local optima. It features 1 a central supervisor that diagnoses each local optimum and restructures a team of agents; and 2 a persistent file system that comprises a learning document, an evolution history, and per-team artifacts, grounding restructuring with accumulated experience. Under a matched token budget, Evolving Agent Teams consistently outperforms strong coding-agent baselines. With an extended token budget, Evolving Agent Teams generates a CUDA kernel that reaches 1.2x mean speedup over the production FlashInfer MLA paged decode kernel from scratch and wins on 36 of 47 production workloads from FlashInfer-Bench 0.86-0.99x on the remaining 11 . It also demonstrates superior performance on tasks such as KDA prefill on AMD MI350X 3.01x over vLLM and NVIDIA B200 3.44x over FlashKDA , MoE expert-offload serving on Apple silicon 30.3 tokens/s decode for Qwen3.8-Flash-Next in MLX 4-bit on a 48 GB M4 Pro , and RoboDojo robot manipulation 62.1 on held-out seeds vs. 55.3 for the best leaderboard entry . Code is available at this https URL https://github.com/caoshiyi/EAT . Additional Features References & Citations Loading... Bibliographic and Citation Tools Bibliographic Explorer What is the Explorer? https://info.arxiv.org/labs/showcase.html arxiv-bibliographic-explorer Connected Papers What is Connected Papers? https://www.connectedpapers.com/about Litmaps What is Litmaps? https://www.litmaps.co/ scite Smart Citations What are Smart Citations? https://www.scite.ai/ Code, Data and Media Associated with this Article alphaXiv What is alphaXiv? https://alphaxiv.org/ CatalyzeX Code Finder for Papers What is CatalyzeX? https://www.catalyzex.com DagsHub What is DagsHub? https://dagshub.com/ Gotit.pub What is GotitPub? http://gotit.pub/faq Hugging Face What is Huggingface? https://huggingface.co/huggingface ScienceCast What is ScienceCast? https://sciencecast.org/welcome Demos Recommenders and Search Tools Influence Flower What are Influence Flowers? https://influencemap.cmlab.dev/ CORE Recommender What is CORE? https://core.ac.uk/services/recommender arXivLabs: experimental projects with community collaborators arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website. Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them. Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs https://info.arxiv.org/labs/index.html .