We propose Decentralized Language Models (DeLM), a MAS framework that decentralizes coordination through parallel agents, a shared verified context, and a task queue.</p>\n","updatedAt":"2026-06-10T16:25:04.493Z","author":{"_id":"69d1ff76196e4c5457605da7","avatarUrl":"/avatars/66969ac38f7a66e02d08d00e5d0121be.svg","fullname":"Yuzhen Mao","name":"gist-sparse-attention","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8197077512741089},"editors":["gist-sparse-attention"],"editorAvatarUrls":["/avatars/66969ac38f7a66e02d08d00e5d0121be.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.10662","authors":[{"_id":"6a298fa06ae15f2243580974","name":"Yuzhen Mao","hidden":false},{"_id":"6a298fa06ae15f2243580975","name":"Azalia Mirhoseini","hidden":false}],"publishedAt":"2026-06-09T00:00:00.000Z","submittedOnDailyAt":"2026-06-10T00:00:00.000Z","title":"Decentralized Multi-Agent Systems with Shared Context","submittedOnDailyBy":{"_id":"69d1ff76196e4c5457605da7","avatarUrl":"/avatars/66969ac38f7a66e02d08d00e5d0121be.svg","isPro":false,"fullname":"Yuzhen Mao","user":"gist-sparse-attention","type":"user","name":"gist-sparse-attention"},"summary":"Multi-agent systems (MAS) can scale large language model reasoning at test time by decomposing complex problems into parallel subtasks. However, most existing MAS rely on centralized orchestration, where a main agent assigns work, collects outputs, and merges results. As the number of subtasks grows, this controller becomes a communication and integration bottleneck. We propose Decentralized Language Models (DeLM), a MAS framework that decentralizes coordination through parallel agents, a shared verified context, and a task queue. Agents asynchronously claim subtasks, read accumulated progress, perform local reasoning, and write back compact verified updates. The shared context acts as a common communication substrate, enabling agents to build on one another's verified progress without routing every update through a central controller. Empirically, DeLM improves both software-engineering test-time scaling and long-context reasoning. On SWE-bench Verified, DeLM achieves the best performance across Avg.@1, Pass@2, and Pass@4, with gains of up to 10.5 percentage points over the strongest baseline, while reducing cost per task by roughly 50%. On LongBench-v2 Multi-Doc QA, DeLM achieves the highest average accuracy across four frontier model families, improving over the strongest baseline by up to 5.7 percentage points. The code is available on our project website at https://yuzhenmao.github.io/DeLM/.","upvotes":1,"discussionId":"6a298fa16ae15f2243580976","projectPage":"https://yuzhenmao.github.io/DeLM/","githubRepo":"https://github.com/yuzhenmao/DeLM","githubRepoAddedBy":"user","ai_summary":"Decentralized Language Models (DeLM) framework enables scalable large language model reasoning through parallel agents that asynchronously coordinate via a shared verified context, improving performance and efficiency over centralized approaches.","ai_keywords":["multi-agent systems","large language models","test-time scaling","parallel subtasks","decentralized coordination","shared verified context","task queue","asynchronous execution","software-engineering","long-context reasoning","SWE-bench Verified","LongBench-v2"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":1,"organization":{"_id":"672c672dcf09d152f4da04c4","name":"StanfordUniversity","fullname":"Stanford University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e396f2b5bb631e9b2fac9a/vJI0POlzGMXL2878t1vz2.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"69d1ff76196e4c5457605da7","avatarUrl":"/avatars/66969ac38f7a66e02d08d00e5d0121be.svg","isPro":false,"fullname":"Yuzhen Mao","user":"gist-sparse-attention","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"672c672dcf09d152f4da04c4","name":"StanfordUniversity","fullname":"Stanford University","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/68e396f2b5bb631e9b2fac9a/vJI0POlzGMXL2878t1vz2.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.10662.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.10662 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.10662 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.10662 in a Space README.md to link it from this page.

Add this paper to a collection to link it from this page.