lynx   »   [go: up one dir, main page]

https://arxivexplained.com/papers/agent-kb-leveraging-cross-domain-experience-for-agentic-problem-solving

\n","updatedAt":"2025-07-10T03:03:40.891Z","author":{"_id":"65d9fc2a0e6ad24551d87a1e","avatarUrl":"/avatars/3aedb9522cc3cd08349d654f523fd792.svg","fullname":"Grant Singleton","name":"grantsing","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":1}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7271271347999573},"editors":["grantsing"],"editorAvatarUrls":["/avatars/3aedb9522cc3cd08349d654f523fd792.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2507.06229","authors":[{"_id":"686dc18acb5725779c60b326","user":{"_id":"63357c608adfa81faf2ac180","avatarUrl":"/avatars/ae0314c644f882251baf59b9134fd36f.svg","isPro":false,"fullname":"Xiangru Tang","user":"RTT1","type":"user"},"name":"Xiangru Tang","status":"claimed_verified","statusLastChangedAt":"2025-07-09T08:50:46.438Z","hidden":false},{"_id":"686dc18acb5725779c60b327","user":{"_id":"64301abe450c0de9a1d3d18e","avatarUrl":"/avatars/01b284874dadc7d21d656c53dcb77e42.svg","isPro":false,"fullname":"tianrui","user":"tianyue818","type":"user"},"name":"Tianrui Qin","status":"claimed_verified","statusLastChangedAt":"2025-08-20T08:53:40.671Z","hidden":false},{"_id":"686dc18acb5725779c60b328","name":"Tianhao Peng","hidden":false},{"_id":"686dc18acb5725779c60b329","user":{"_id":"67930201aad25d3eecab81cc","avatarUrl":"/avatars/afa8e19ccd5214979e405caf462d7a72.svg","isPro":false,"fullname":"ZiyangZhou","user":"AzHouangy","type":"user"},"name":"Ziyang Zhou","status":"claimed_verified","statusLastChangedAt":"2025-07-09T08:50:48.624Z","hidden":false},{"_id":"686dc18acb5725779c60b32a","name":"Daniel Shao","hidden":false},{"_id":"686dc18acb5725779c60b32b","name":"Tingting Du","hidden":false},{"_id":"686dc18acb5725779c60b32c","name":"Xinming Wei","hidden":false},{"_id":"686dc18acb5725779c60b32d","user":{"_id":"643e9ee6f6bb3c31a26e7bc4","avatarUrl":"/avatars/acfaa7d6a23dada24c86b954c3be116a.svg","isPro":false,"fullname":"Peng Xia","user":"richardxp888","type":"user"},"name":"Peng Xia","status":"claimed_verified","statusLastChangedAt":"2025-07-09T08:50:44.151Z","hidden":false},{"_id":"686dc18acb5725779c60b32e","user":{"_id":"675e0d5cdd3e9eeed6954f5a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/7oMEoBmaFiCR9K2q9Z_7q.png","isPro":false,"fullname":"Fang Wu","user":"fangwu97","type":"user"},"name":"Fang Wu","status":"claimed_verified","statusLastChangedAt":"2025-07-23T08:37:25.015Z","hidden":false},{"_id":"686dc18acb5725779c60b32f","name":"He Zhu","hidden":false},{"_id":"686dc18acb5725779c60b330","user":{"_id":"638efcf4c67af472d316d424","avatarUrl":"/avatars/97a57859d7d87a3a8f1bb41d32a72bc2.svg","isPro":false,"fullname":"Ge Zhang","user":"zhangysk","type":"user"},"name":"Ge Zhang","status":"claimed_verified","statusLastChangedAt":"2025-07-09T08:50:41.208Z","hidden":false},{"_id":"686dc18acb5725779c60b331","name":"Jiaheng Liu","hidden":false},{"_id":"686dc18acb5725779c60b332","name":"Xingyao Wang","hidden":false},{"_id":"686dc18acb5725779c60b333","name":"Sirui Hong","hidden":false},{"_id":"686dc18acb5725779c60b334","name":"Chenglin Wu","hidden":false},{"_id":"686dc18acb5725779c60b335","user":{"_id":"608c49299d7c9519b4adae25","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/608c49299d7c9519b4adae25/pWSW52_QU7VD9hhDUHbmq.jpeg","isPro":false,"fullname":"Hao Cheng","user":"kelvinih","type":"user"},"name":"Hao Cheng","status":"claimed_verified","statusLastChangedAt":"2025-09-09T13:51:52.706Z","hidden":false},{"_id":"686dc18acb5725779c60b336","name":"Chi Wang","hidden":false},{"_id":"686dc18acb5725779c60b337","user":{"_id":"628c8598ef14f971b698107f","avatarUrl":"/avatars/3a4ad87e6b5f9e836a1160d869df1447.svg","isPro":false,"fullname":"Zhou","user":"Wangchunshu","type":"user"},"name":"Wangchunshu Zhou","status":"claimed_verified","statusLastChangedAt":"2025-07-29T12:51:17.995Z","hidden":false}],"publishedAt":"2025-07-08T17:59:22.000Z","submittedOnDailyAt":"2025-07-09T17:50:12.079Z","title":"Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving","submittedOnDailyBy":{"_id":"63357c608adfa81faf2ac180","avatarUrl":"/avatars/ae0314c644f882251baf59b9134fd36f.svg","isPro":false,"fullname":"Xiangru Tang","user":"RTT1","type":"user"},"summary":"As language agents tackle increasingly complex tasks, they struggle with\neffective error correction and experience reuse across domains. We introduce\nAgent KB, a hierarchical experience framework that enables complex agentic\nproblem solving via a novel Reason-Retrieve-Refine pipeline. Agent KB addresses\na core limitation: agents traditionally cannot learn from each other's\nexperiences. By capturing both high-level strategies and detailed execution\nlogs, Agent KB creates a shared knowledge base that enables cross-agent\nknowledge transfer. Evaluated on the GAIA benchmark, Agent KB improves success\nrates by up to 16.28 percentage points. On the most challenging tasks, Claude-3\nimproves from 38.46% to 57.69%, while GPT-4 improves from 53.49% to 73.26% on\nintermediate tasks. On SWE-bench code repair, Agent KB enables Claude-3 to\nimprove from 41.33% to 53.33%. Our results suggest that Agent KB provides a\nmodular, framework-agnostic infrastructure for enabling agents to learn from\npast experiences and generalize successful strategies to new tasks.","upvotes":74,"discussionId":"686dc18acb5725779c60b338","githubRepo":"https://github.com/OPPO-PersonalAI/Agent-KB","ai_summary":"Agent KB, a hierarchical experience framework, enhances problem-solving success rates across different agents by enabling cross-agent knowledge transfer through a Reason-Retrieve-Refine pipeline.","ai_keywords":["Agent KB","hierarchical experience framework","Reason-Retrieve-Refine pipeline","cross-agent knowledge transfer","GAIA benchmark","SWE-bench code repair"],"githubStars":367},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"63357c608adfa81faf2ac180","avatarUrl":"/avatars/ae0314c644f882251baf59b9134fd36f.svg","isPro":false,"fullname":"Xiangru Tang","user":"RTT1","type":"user"},{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"65377c30e48353201e6fdda0","avatarUrl":"/avatars/a8f803b6f2e598eaee9c52c0d2ddfc16.svg","isPro":false,"fullname":"Jiaheng Liu","user":"CheeryLJH","type":"user"},{"_id":"67f5982ee6293b1b2a3c39ff","avatarUrl":"/avatars/bbd2c3858131876536c008d1dbdde023.svg","isPro":false,"fullname":"Jiapeng Chen","user":"GabbyChen","type":"user"},{"_id":"62970df979f193515da13dc0","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62970df979f193515da13dc0/A-mgKIcgTXRJ54GCHswTq.jpeg","isPro":false,"fullname":"Yanjun Shao","user":"super-dainiu","type":"user"},{"_id":"6578265ddea7e2122d02f6ba","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6578265ddea7e2122d02f6ba/Bh6JjoVF5ceLSjV7Z7nTk.jpeg","isPro":false,"fullname":"kang zhu","user":"kangz","type":"user"},{"_id":"6750142023e47fa7ff42fabc","avatarUrl":"/avatars/2a7ebdcf79907dd39cbad40d03085b71.svg","isPro":false,"fullname":"Linabear","user":"xyz2021314","type":"user"},{"_id":"67930201aad25d3eecab81cc","avatarUrl":"/avatars/afa8e19ccd5214979e405caf462d7a72.svg","isPro":false,"fullname":"ZiyangZhou","user":"AzHouangy","type":"user"},{"_id":"64301abe450c0de9a1d3d18e","avatarUrl":"/avatars/01b284874dadc7d21d656c53dcb77e42.svg","isPro":false,"fullname":"tianrui","user":"tianyue818","type":"user"},{"_id":"64b15284372d4340772a3dca","avatarUrl":"/avatars/417d5f1bc1bcb5e4d5de6169673c2cf7.svg","isPro":false,"fullname":"Zefan Cai","user":"ZefanCai","type":"user"},{"_id":"65cae89119683f9817c049ea","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65cae89119683f9817c049ea/A0XxjmaJldu28JhFvWmpP.jpeg","isPro":false,"fullname":"Wenqi Shi","user":"wshi83","type":"user"},{"_id":"64ba096e760936217a3ad2e2","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64ba096e760936217a3ad2e2/aNQK83Jg5PsBkY0UDg-RA.jpeg","isPro":false,"fullname":"Linzheng Chai","user":"Challenging666","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":0}">
Papers
arxiv:2507.06229

Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving

Published on Jul 8
· Submitted by Xiangru Tang on Jul 9
Authors:
,
,
,
,
,
,
,
,
,
,

Abstract

Agent KB, a hierarchical experience framework, enhances problem-solving success rates across different agents by enabling cross-agent knowledge transfer through a Reason-Retrieve-Refine pipeline.

AI-generated summary

As language agents tackle increasingly complex tasks, they struggle with effective error correction and experience reuse across domains. We introduce Agent KB, a hierarchical experience framework that enables complex agentic problem solving via a novel Reason-Retrieve-Refine pipeline. Agent KB addresses a core limitation: agents traditionally cannot learn from each other's experiences. By capturing both high-level strategies and detailed execution logs, Agent KB creates a shared knowledge base that enables cross-agent knowledge transfer. Evaluated on the GAIA benchmark, Agent KB improves success rates by up to 16.28 percentage points. On the most challenging tasks, Claude-3 improves from 38.46% to 57.69%, while GPT-4 improves from 53.49% to 73.26% on intermediate tasks. On SWE-bench code repair, Agent KB enables Claude-3 to improve from 41.33% to 53.33%. Our results suggest that Agent KB provides a modular, framework-agnostic infrastructure for enabling agents to learn from past experiences and generalize successful strategies to new tasks.

Community

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2507.06229 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/2507.06229 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/2507.06229 in a Space README.md to link it from this page.

Collections including this paper 5

Лучший частный хостинг