Prunes tool outputs in coding agents using internal representations for efficient context management without external classifiers.</p>\n","updatedAt":"2026-07-21T03:15:12.533Z","author":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","fullname":"taesiri","name":"taesiri","type":"user","isPro":true,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":339,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6396944522857666},"editors":["taesiri"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg"],"reactions":[],"isReport":false}},{"id":"6a5f182c197d7fe699af5008","author":{"_id":"679cbc024ee3403814ca0e74","avatarUrl":"/avatars/0fc11476c642121eb264a891b072bb0e.svg","fullname":"Yuhang Wang","name":"ayanami-kitasan","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false},"createdAt":"2026-07-21T06:56:44.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"See the code in https://github.com/Ayanami1314/swe-pruner-pro!","html":"<p>See the code in <a href=\"https://github.com/Ayanami1314/swe-pruner-pro\" rel=\"nofollow\">https://github.com/Ayanami1314/swe-pruner-pro</a>!</p>\n","updatedAt":"2026-07-21T06:56:44.067Z","author":{"_id":"679cbc024ee3403814ca0e74","avatarUrl":"/avatars/0fc11476c642121eb264a891b072bb0e.svg","fullname":"Yuhang Wang","name":"ayanami-kitasan","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6347616910934448},"editors":["ayanami-kitasan"],"editorAvatarUrls":["/avatars/0fc11476c642121eb264a891b072bb0e.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2607.18213","authors":[{"_id":"6a5ee42d4fe5d1d13e84ab85","name":"Yuhang Wang","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab86","name":"Yuling Shi","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab87","name":"Shaoqiu Zhang","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab88","name":"Jialiang Liang","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab89","name":"Shilin He","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab8a","name":"Siyu Ye","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab8b","name":"Yuting Chen","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab8c","name":"Kai Cai","hidden":false},{"_id":"6a5ee42d4fe5d1d13e84ab8d","name":"Xiaodong Gu","hidden":false}],"publishedAt":"2026-07-20T00:00:00.000Z","submittedOnDailyAt":"2026-07-21T00:00:00.000Z","title":"SWE-Pruner Pro: The Coder LLM Already Knows What to Prune","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user","name":"taesiri"},"summary":"Pruning long context for coding agents has been a vital technology for efficient context management. While existing context pruning methods such as SWE-Pruner realize this by attaching a separate code classifier, we find the agent itself encodes internal representations indicating the relevance of code context when reading tool output. Based on this finding, we propose SWE-Pruner Pro, which prunes tool outputs directly inside the agent. Concretely, a small head turns the agent's own internal representations into a keep-or-prune label for each line, with a length-aware embedding keyed to each tool output's line count. Across two open-weight backbones and four multi-turn benchmarks, SWE-Pruner Pro saves up to 39% of prompt and completion tokens while preserving task quality, with bounded inference overhead. Notably, on MiMo-V2-Flash SWE-Pruner Pro additionally raises the SWE-Bench Verified resolve rate by +3.8% and the long-context Oolong accuracy by +2.2 points.","upvotes":41,"discussionId":"6a5ee42e4fe5d1d13e84ab8e","organization":{"_id":"653b817d32c97d0655575872","name":"ByteDance","fullname":"ByteDance","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6535c9e88bde2fae19b6fb25/0clr54wj5Ly-RkYU9OXPp.png"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6a2da6c8ca070ee12c6e396c","avatarUrl":"/avatars/0355287dcabaa67dbc7f0b10b87451f9.svg","isPro":false,"fullname":"Joe Mama","user":"JoeMama123123123","type":"user"},{"_id":"63ac5701c21e60a3e9b58aa7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/63ac5701c21e60a3e9b58aa7/g6EX7diOpuA94R2ab-rZC.png","isPro":true,"fullname":"Dipankar Sarkar","user":"dipankarsarkar","type":"user"},{"_id":"645b0c3ec35da9c7afd95421","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/645b0c3ec35da9c7afd95421/vYBrCDagHsXAo6J2p-uG0.jpeg","isPro":false,"fullname":"Yuling","user":"YerbaPage","type":"user"},{"_id":"68831681240aa3d8ce43e1bf","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/iSjQ3OIkIvqm_0Wxv4Qib.png","isPro":false,"fullname":"Azzz","user":"azzzacs","type":"user"},{"_id":"68e2570e6ecc8f79ab4577ec","avatarUrl":"/avatars/408e8683b6947b1315c9eb41da5b4a34.svg","isPro":false,"fullname":"aa","user":"Disaaad","type":"user"},{"_id":"679cbc024ee3403814ca0e74","avatarUrl":"/avatars/0fc11476c642121eb264a891b072bb0e.svg","isPro":false,"fullname":"Yuhang Wang","user":"ayanami-kitasan","type":"user"},{"_id":"688c72c011ef3399b561dee7","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/688c72c011ef3399b561dee7/puhgnTOAfZYetsC46hqGm.jpeg","isPro":false,"fullname":"BoxueYang","user":"Boxue","type":"user"},{"_id":"68a137c3846db9d4bad0995f","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/cFRgODYfk8IGxv8jjYgBL.png","isPro":false,"fullname":"黄雨辰","user":"zhaohai","type":"user"},{"_id":"6715b493d54796e4b99d90e8","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6715b493d54796e4b99d90e8/X-VqGsRrjPlfb94GRn13x.jpeg","isPro":false,"fullname":"黄炜锴","user":"tsrigo","type":"user"},{"_id":"68f2120d45cc9162fb3ec64c","avatarUrl":"/avatars/cb5b211a06a8f48d397419d4bccd1766.svg","isPro":false,"fullname":"caixintong","user":"Stella3029","type":"user"},{"_id":"697c50df4362f3a23b63444b","avatarUrl":"/avatars/d0a959fa4cc1c8361e318451a1b4bd45.svg","isPro":false,"fullname":"Cody Tang","user":"AMCody","type":"user"},{"_id":"67990449a5da36c49b9cf4ae","avatarUrl":"/avatars/85fd1dbd65a07175528f808ee3c927b1.svg","isPro":false,"fullname":"YM","user":"Yuanmoo","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":1,"organization":{"_id":"653b817d32c97d0655575872","name":"ByteDance","fullname":"ByteDance","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6535c9e88bde2fae19b6fb25/0clr54wj5Ly-RkYU9OXPp.png"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2607/2607.18213.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2607.18213 in a model README.md to link it from this page.

Cite arxiv.org/abs/2607.18213 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2607.18213 in a Space README.md to link it from this page.

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