Deprecated: The each() function is deprecated. This message will be suppressed on further calls in /home/zhenxiangba/zhenxiangba.com/public_html/phproxy-improved-master/index.php on line 456
Paper page - SWE-Universe: Scale Real-World Verifiable Environments to Millions
[go: Go Back, main page]

\"image\"

\n","updatedAt":"2026-02-03T05:41:46.667Z","author":{"_id":"66d58df54b87a685ccb8e4a0","avatarUrl":"/avatars/2566c20d79088ba761215b9a0197cb8e.svg","fullname":"Mouxiang Chen","name":"chenmouxiang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":5,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7476180791854858},"editors":["chenmouxiang"],"editorAvatarUrls":["/avatars/2566c20d79088ba761215b9a0197cb8e.svg"],"reactions":[{"reaction":"🔥","users":["AdinaY"],"count":1}],"isReport":false}},{"id":"6982a3d2291c5930ff3a78c5","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":317,"isUserFollowing":false},"createdAt":"2026-02-04T01:41:38.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"This is an automated message from the [Librarian Bot](https://huggingface.co/librarian-bots). I found the following papers similar to this paper. \n\nThe following papers were recommended by the Semantic Scholar API \n\n* [MEnvAgent: Scalable Polyglot Environment Construction for Verifiable Software Engineering](https://huggingface.co/papers/2601.22859) (2026)\n* [Training Versatile Coding Agents in Synthetic Environments](https://huggingface.co/papers/2512.12216) (2025)\n* [SWE-EVO: Benchmarking Coding Agents in Long-Horizon Software Evolution Scenarios](https://huggingface.co/papers/2512.18470) (2025)\n* [SWE-Lego: Pushing the Limits of Supervised Fine-tuning for Software Issue Resolving](https://huggingface.co/papers/2601.01426) (2026)\n* [Multi-Docker-Eval: A 'Shovel of the Gold Rush' Benchmark on Automatic Environment Building for Software Engineering](https://huggingface.co/papers/2512.06915) (2025)\n* [daVinci-Dev: Agent-native Mid-training for Software Engineering](https://huggingface.co/papers/2601.18418) (2026)\n* [DevOps-Gym: Benchmarking AI Agents in Software DevOps Cycle](https://huggingface.co/papers/2601.20882) (2026)\n\n\n Please give a thumbs up to this comment if you found it helpful!\n\n If you want recommendations for any Paper on Hugging Face checkout [this](https://huggingface.co/spaces/librarian-bots/recommend_similar_papers) Space\n\n You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: `@librarian-bot recommend`","html":"

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

\n

The following papers were recommended by the Semantic Scholar API

\n\n

Please give a thumbs up to this comment if you found it helpful!

\n

If you want recommendations for any Paper on Hugging Face checkout this Space

\n

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: \n\n@librarian-bot\n\t recommend

\n","updatedAt":"2026-02-04T01:41:38.059Z","author":{"_id":"63d3e0e8ff1384ce6c5dd17d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg","fullname":"Librarian Bot (Bot)","name":"librarian-bot","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":317,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7421538233757019},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2602.02361","authors":[{"_id":"69818982ce18b18628096375","user":{"_id":"66d58df54b87a685ccb8e4a0","avatarUrl":"/avatars/2566c20d79088ba761215b9a0197cb8e.svg","isPro":false,"fullname":"Mouxiang Chen","user":"chenmouxiang","type":"user"},"name":"Mouxiang Chen","status":"claimed_verified","statusLastChangedAt":"2026-02-03T10:03:16.116Z","hidden":false},{"_id":"69818982ce18b18628096376","user":{"_id":"64c38871f9cd765462fa1a17","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64c38871f9cd765462fa1a17/yuIlVcqeDlQVKsUF8uEl3.jpeg","isPro":false,"fullname":"Lei Zhang","user":"Lemoncoke","type":"user"},"name":"Lei Zhang","status":"claimed_verified","statusLastChangedAt":"2026-02-03T10:03:13.717Z","hidden":false},{"_id":"69818982ce18b18628096377","user":{"_id":"5df83428da6d0311fd3d5404","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1613644244177-5df83428da6d0311fd3d5404.jpeg","isPro":false,"fullname":"Feng YunLong","user":"ylfeng","type":"user"},"name":"Yunlong Feng","status":"claimed_verified","statusLastChangedAt":"2026-02-03T10:03:11.567Z","hidden":false},{"_id":"69818982ce18b18628096378","name":"Xuwu Wang","hidden":false},{"_id":"69818982ce18b18628096379","name":"Wenting Zhao","hidden":false},{"_id":"69818982ce18b1862809637a","user":{"_id":"5fbcf0d28f35b82700205fd7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5fbcf0d28f35b82700205fd7/KZPwcaIymBAPs0UHXe5fa.jpeg","isPro":false,"fullname":"Ruisheng Cao","user":"rshcao","type":"user"},"name":"Ruisheng Cao","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:07:06.585Z","hidden":false},{"_id":"69818982ce18b1862809637b","user":{"_id":"646df403ad20c6fa4f30b7ec","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/646df403ad20c6fa4f30b7ec/Q64-XMghOcBoo3itZDGYA.jpeg","isPro":false,"fullname":"Jiaxi Yang","user":"jx-yang","type":"user"},"name":"Jiaxi Yang","status":"claimed_verified","statusLastChangedAt":"2026-02-03T10:03:08.705Z","hidden":false},{"_id":"69818982ce18b1862809637c","name":"Jiawei Chen","hidden":false},{"_id":"69818982ce18b1862809637d","user":{"_id":"6923fc28c3a5838b88458179","avatarUrl":"/avatars/229ec56ce5d019223ad625a41b471aef.svg","isPro":false,"fullname":"Mingze Li","user":"Crystalze36","type":"user"},"name":"Mingze Li","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:06:52.286Z","hidden":false},{"_id":"69818982ce18b1862809637e","user":{"_id":"6384c07fdfffab4824ff45fb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1669644372381-noauth.jpeg","isPro":false,"fullname":"Zeyao Ma","user":"KAKA22","type":"user"},"name":"Zeyao Ma","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:06:44.545Z","hidden":false},{"_id":"69818982ce18b1862809637f","name":"Hao Ge","hidden":false},{"_id":"69818982ce18b18628096380","name":"Zongmeng Zhang","hidden":false},{"_id":"69818982ce18b18628096381","name":"Zeyu Cui","hidden":false},{"_id":"69818982ce18b18628096382","user":{"_id":"6434d4989bd5a84b5dd0b0f5","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6434d4989bd5a84b5dd0b0f5/0Elf9qbfG9Hkgypm9pTGm.jpeg","isPro":false,"fullname":"Dayiheng Liu","user":"Losin94","type":"user"},"name":"Dayiheng Liu","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:06:31.666Z","hidden":false},{"_id":"69818982ce18b18628096383","name":"Jingren Zhou","hidden":false},{"_id":"69818982ce18b18628096384","name":"Jianling Sun","hidden":false},{"_id":"69818982ce18b18628096385","user":{"_id":"620760a26e3b7210c2ff1943","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620760a26e3b7210c2ff1943/VC-rKqimF6yxGESNVlPoR.jpeg","isPro":false,"fullname":"Junyang Lin","user":"JustinLin610","type":"user"},"name":"Junyang Lin","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:06:14.151Z","hidden":false},{"_id":"69818982ce18b18628096386","user":{"_id":"61e4c4ca1ab24785ac11ba69","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61e4c4ca1ab24785ac11ba69/1Q1zhhyGSJ9RJG9MzwxVv.jpeg","isPro":false,"fullname":"Binyuan Hui","user":"huybery","type":"user"},"name":"Binyuan Hui","status":"admin_assigned","statusLastChangedAt":"2026-02-04T19:06:05.981Z","hidden":false}],"publishedAt":"2026-02-02T17:20:30.000Z","submittedOnDailyAt":"2026-02-03T03:11:46.660Z","title":"SWE-Universe: Scale Real-World Verifiable Environments to Millions","submittedOnDailyBy":{"_id":"66d58df54b87a685ccb8e4a0","avatarUrl":"/avatars/2566c20d79088ba761215b9a0197cb8e.svg","isPro":false,"fullname":"Mouxiang Chen","user":"chenmouxiang","type":"user"},"summary":"We propose SWE-Universe, a scalable and efficient framework for automatically constructing real-world software engineering (SWE) verifiable environments from GitHub pull requests (PRs). To overcome the prevalent challenges of automatic building, such as low production yield, weak verifiers, and prohibitive cost, our framework utilizes a building agent powered by an efficient custom-trained model. This agent employs iterative self-verification and in-loop hacking detection to ensure the reliable generation of high-fidelity, verifiable tasks. Using this method, we scale the number of real-world multilingual SWE environments to a million scale (807,693). We demonstrate the profound value of our environments through large-scale agentic mid-training and reinforcement learning. Finally, we applied this technique to Qwen3-Max-Thinking and achieved a score of 75.3% on SWE-Bench Verified. Our work provides both a critical resource and a robust methodology to advance the next generation of coding agents.","upvotes":60,"discussionId":"69818983ce18b18628096387","ai_summary":"A scalable framework for constructing real-world software engineering environments from GitHub pull requests using an efficient building agent with self-verification and hacking detection capabilities.","ai_keywords":["building agent","self-verification","in-loop hacking detection","real-world software engineering","GitHub pull requests","custom-trained model","verifiable environments","agentic mid-training","reinforcement learning","SWE-Bench Verified"],"organization":{"_id":"64c8b5837fe12ecd0a7e92eb","name":"Qwen","fullname":"Qwen","avatar":"https://cdn-uploads.huggingface.co/production/uploads/620760a26e3b7210c2ff1943/-s1gyJfvbE1RgO5iBeNOi.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63ffdbaab09f82a81a222c27","avatarUrl":"/avatars/3aca53f6555f4548489844902fe4a80a.svg","isPro":false,"fullname":"Wenting Zhao","user":"wentingzhao","type":"user"},{"_id":"64c38871f9cd765462fa1a17","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64c38871f9cd765462fa1a17/yuIlVcqeDlQVKsUF8uEl3.jpeg","isPro":false,"fullname":"Lei Zhang","user":"Lemoncoke","type":"user"},{"_id":"6384c07fdfffab4824ff45fb","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1669644372381-noauth.jpeg","isPro":false,"fullname":"Zeyao Ma","user":"KAKA22","type":"user"},{"_id":"641c9662043963b1c0a1df52","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/641c9662043963b1c0a1df52/L1o85EHztv_xP9r6ppljf.jpeg","isPro":false,"fullname":"KaShun SHUM","user":"ksshumab","type":"user"},{"_id":"63eb30275c837d9968f3a2c7","avatarUrl":"/avatars/f0da56bca6760b9a79133bba6eb4379d.svg","isPro":false,"fullname":"Jiawei Chen","user":"jiawei1998","type":"user"},{"_id":"5df83428da6d0311fd3d5404","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1613644244177-5df83428da6d0311fd3d5404.jpeg","isPro":false,"fullname":"Feng YunLong","user":"ylfeng","type":"user"},{"_id":"64b63a26fafb5319dc9458e9","avatarUrl":"/avatars/2a893215d61a176766ba77c2a203ede1.svg","isPro":false,"fullname":"wxw wang","user":"0wxw0","type":"user"},{"_id":"66b34647a29e5c00011d34c3","avatarUrl":"/avatars/ae39f9b84f9379423fa3a8509fbcc94e.svg","isPro":false,"fullname":"Zongmeng Zhang","user":"ustc-zhangzm","type":"user"},{"_id":"6323f399462470712720c155","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6323f399462470712720c155/SWsMNa7vETUSrOt9Qf-oe.png","isPro":false,"fullname":"Yinxu Pan","user":"cppowboy","type":"user"},{"_id":"65897684f8b453e1f57cdb26","avatarUrl":"/avatars/80096d6c808805e1a84a68fb6194a7d4.svg","isPro":false,"fullname":"huxueyu","user":"huxueyu","type":"user"},{"_id":"65192e8a107446b24ca96680","avatarUrl":"/avatars/b202fb605d4e14b996a59d6966e95218.svg","isPro":false,"fullname":"Jiang","user":"LiangJiang","type":"user"},{"_id":"6434d4989bd5a84b5dd0b0f5","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6434d4989bd5a84b5dd0b0f5/0Elf9qbfG9Hkgypm9pTGm.jpeg","isPro":false,"fullname":"Dayiheng Liu","user":"Losin94","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0,"organization":{"_id":"64c8b5837fe12ecd0a7e92eb","name":"Qwen","fullname":"Qwen","avatar":"https://cdn-uploads.huggingface.co/production/uploads/620760a26e3b7210c2ff1943/-s1gyJfvbE1RgO5iBeNOi.png"}}">
Papers
arxiv:2602.02361

SWE-Universe: Scale Real-World Verifiable Environments to Millions

Published on Feb 2
· Submitted by
Mouxiang Chen
on Feb 3
· Qwen Qwen
Authors:
,
,
,
,
,
,
,
,

Abstract

A scalable framework for constructing real-world software engineering environments from GitHub pull requests using an efficient building agent with self-verification and hacking detection capabilities.

AI-generated summary

We propose SWE-Universe, a scalable and efficient framework for automatically constructing real-world software engineering (SWE) verifiable environments from GitHub pull requests (PRs). To overcome the prevalent challenges of automatic building, such as low production yield, weak verifiers, and prohibitive cost, our framework utilizes a building agent powered by an efficient custom-trained model. This agent employs iterative self-verification and in-loop hacking detection to ensure the reliable generation of high-fidelity, verifiable tasks. Using this method, we scale the number of real-world multilingual SWE environments to a million scale (807,693). We demonstrate the profound value of our environments through large-scale agentic mid-training and reinforcement learning. Finally, we applied this technique to Qwen3-Max-Thinking and achieved a score of 75.3% on SWE-Bench Verified. Our work provides both a critical resource and a robust methodology to advance the next generation of coding agents.

Community

Paper author Paper submitter

We propose SWE-Universe, a scalable and efficient framework for automatically constructing real-world software engineering (SWE) verifiable environments from GitHub pull requests (PRs). Using this method, we scale the number of real-world multilingual SWE environments to a million scale (807,693). Finally, we applied this technique to Qwen3-Max-Thinking and achieved a score of 75.3% on SWE-Bench Verified.

image

This is an automated message from the Librarian Bot. I found the following papers similar to this paper.

The following papers were recommended by the Semantic Scholar API

Please give a thumbs up to this comment if you found it helpful!

If you want recommendations for any Paper on Hugging Face checkout this Space

You can directly ask Librarian Bot for paper recommendations by tagging it in a comment: @librarian-bot recommend

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2602.02361 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2602.02361 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2602.02361 in a Space README.md to link it from this page.

Collections including this paper 6