https://github.com/Tencent/CognitiveKernel-Pro

\n","updatedAt":"2025-08-04T17:31:44.766Z","author":{"_id":"657cd228138b7e391444a65d","avatarUrl":"/avatars/c7c984ae483144fab627aa2c54d91d0f.svg","fullname":"Xiaoyang Wang","name":"xywang1","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":12,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7140704393386841},"editors":["xywang1"],"editorAvatarUrls":["/avatars/c7c984ae483144fab627aa2c54d91d0f.svg"],"reactions":[{"reaction":"🔥","users":["tqfang229","Forbu14"],"count":2}],"isReport":false}},{"id":"68916027cef8b00d9f8b967e","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":"2025-08-05T01:36:39.000Z","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* [WebSailor: Navigating Super-human Reasoning for Web Agent](https://huggingface.co/papers/2507.02592) (2025)\n* [MetaAgent: Toward Self-Evolving Agent via Tool Meta-Learning](https://huggingface.co/papers/2508.00271) (2025)\n* [Decoupled Planning and Execution: A Hierarchical Reasoning Framework for Deep Search](https://huggingface.co/papers/2507.02652) (2025)\n* [MMAT-1M: A Large Reasoning Dataset for Multimodal Agent Tuning](https://huggingface.co/papers/2507.21924) (2025)\n* [SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam?](https://huggingface.co/papers/2507.05241) (2025)\n* [Deep Research Agents: A Systematic Examination And Roadmap](https://huggingface.co/papers/2506.18096) (2025)\n* [AgentOrchestra: A Hierarchical Multi-Agent Framework for General-Purpose Task Solving](https://huggingface.co/papers/2506.12508) (2025)\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
• WebSailor: Navigating Super-human Reasoning for Web Agent (2025)
\n
• MetaAgent: Toward Self-Evolving Agent via Tool Meta-Learning (2025)
\n
• Decoupled Planning and Execution: A Hierarchical Reasoning Framework for Deep Search (2025)
\n
• MMAT-1M: A Large Reasoning Dataset for Multimodal Agent Tuning (2025)
\n
• SciMaster: Towards General-Purpose Scientific AI Agents, Part I. X-Master as Foundation: Can We Lead on Humanity's Last Exam? (2025)
\n
• Deep Research Agents: A Systematic Examination And Roadmap (2025)
\n
• AgentOrchestra: A Hierarchical Multi-Agent Framework for General-Purpose Task Solving (2025)
\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":"2025-08-05T01:36:39.975Z","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":0.71593177318573},"editors":["librarian-bot"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/1674830754237-63d3e0e8ff1384ce6c5dd17d.jpeg"],"reactions":[{"reaction":"👍","users":["ako-oak","justiin"],"count":2}],"isReport":false}},{"id":"689c2afd2466b47e68bfa237","author":{"_id":"65d9fc2a0e6ad24551d87a1e","avatarUrl":"/avatars/3aedb9522cc3cd08349d654f523fd792.svg","fullname":"Grant Singleton","name":"grantsing","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false},"createdAt":"2025-08-13T06:04:45.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/cognitive-kernel-pro-a-framework-for-deep-research-agents-and-agent-foundation-models-training","html":"

arXiv explained breakdown of this paper 👉 https://arxivexplained.com/papers/cognitive-kernel-pro-a-framework-for-deep-research-agents-and-agent-foundation-models-training

\n","updatedAt":"2025-08-13T06:04:45.065Z","author":{"_id":"65d9fc2a0e6ad24551d87a1e","avatarUrl":"/avatars/3aedb9522cc3cd08349d654f523fd792.svg","fullname":"Grant Singleton","name":"grantsing","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":4,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.6863317489624023},"editors":["grantsing"],"editorAvatarUrls":["/avatars/3aedb9522cc3cd08349d654f523fd792.svg"],"reactions":[],"isReport":false}},{"id":"68a094584a09a88da53307ca","author":{"_id":"65d02414af9670557423df10","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65d02414af9670557423df10/5SNGUGEEdkliGKNceo_gS.jpeg","fullname":"Pranav Pawar","name":"prnvpwr2612","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":9,"isUserFollowing":false},"createdAt":"2025-08-16T14:23:20.000Z","type":"comment","data":{"edited":false,"hidden":false,"latest":{"raw":"Glad to see high performance deep research open source frameworks!\nThank you, would surely learn a lot from this!","html":"

Glad to see high performance deep research open source frameworks!
Thank you, would surely learn a lot from this!

\n","updatedAt":"2025-08-16T14:23:20.684Z","author":{"_id":"65d02414af9670557423df10","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/65d02414af9670557423df10/5SNGUGEEdkliGKNceo_gS.jpeg","fullname":"Pranav Pawar","name":"prnvpwr2612","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":9,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.9321502447128296},"editors":["prnvpwr2612"],"editorAvatarUrls":["https://cdn-avatars.huggingface.co/v1/production/uploads/65d02414af9670557423df10/5SNGUGEEdkliGKNceo_gS.jpeg"],"reactions":[{"reaction":"❤️","users":["xywang1"],"count":1}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2508.00414","authors":[{"_id":"6890eddff01a094725f833f8","user":{"_id":"641129818573c51c0458b793","avatarUrl":"/avatars/d4bc67c160a07146cf41c614678aa36b.svg","isPro":false,"fullname":"Tianqing Fang","user":"tqfang229","type":"user"},"name":"Tianqing Fang","status":"claimed_verified","statusLastChangedAt":"2025-08-06T19:23:18.239Z","hidden":false},{"_id":"6890eddff01a094725f833f9","name":"Zhisong Zhang","hidden":false},{"_id":"6890eddff01a094725f833fa","user":{"_id":"657cd228138b7e391444a65d","avatarUrl":"/avatars/c7c984ae483144fab627aa2c54d91d0f.svg","isPro":false,"fullname":"Xiaoyang Wang","user":"xywang1","type":"user"},"name":"Xiaoyang Wang","status":"claimed_verified","statusLastChangedAt":"2025-08-06T19:23:25.761Z","hidden":false},{"_id":"6890eddff01a094725f833fb","name":"Rui Wang","hidden":false},{"_id":"6890eddff01a094725f833fc","name":"Can Qin","hidden":false},{"_id":"6890eddff01a094725f833fd","user":{"_id":"64c0f43732dd2d752e780322","avatarUrl":"/avatars/b2be921b3d5744b2615f03bc4de47b40.svg","isPro":false,"fullname":"Yuxuan Wan","user":"iforgott","type":"user"},"name":"Yuxuan Wan","status":"claimed_verified","statusLastChangedAt":"2026-01-08T08:34:25.617Z","hidden":false},{"_id":"6890eddff01a094725f833fe","name":"Jun-Yu Ma","hidden":false},{"_id":"6890eddff01a094725f833ff","name":"Ce Zhang","hidden":false},{"_id":"6890eddff01a094725f83400","name":"Jiaqi Chen","hidden":false},{"_id":"6890eddff01a094725f83401","user":{"_id":"66a3b73ac44ffdaf67aa0a74","avatarUrl":"/avatars/ed73f39ea541ddcc35b7405af42485e5.svg","isPro":false,"fullname":"Xiyun Li","user":"xiyun98","type":"user"},"name":"Xiyun Li","status":"claimed_verified","statusLastChangedAt":"2025-08-06T19:23:21.773Z","hidden":false},{"_id":"6890eddff01a094725f83402","name":"Hongming Zhang","hidden":false},{"_id":"6890eddff01a094725f83403","name":"Haitao Mi","hidden":false},{"_id":"6890eddff01a094725f83404","name":"Dong Yu","hidden":false}],"publishedAt":"2025-08-01T08:11:31.000Z","submittedOnDailyAt":"2025-08-04T16:01:44.753Z","title":"Cognitive Kernel-Pro: A Framework for Deep Research Agents and Agent\n Foundation Models Training","submittedOnDailyBy":{"_id":"657cd228138b7e391444a65d","avatarUrl":"/avatars/c7c984ae483144fab627aa2c54d91d0f.svg","isPro":false,"fullname":"Xiaoyang Wang","user":"xywang1","type":"user"},"summary":"General AI Agents are increasingly recognized as foundational frameworks for\nthe next generation of artificial intelligence, enabling complex reasoning, web\ninteraction, coding, and autonomous research capabilities. However, current\nagent systems are either closed-source or heavily reliant on a variety of paid\nAPIs and proprietary tools, limiting accessibility and reproducibility for the\nresearch community. In this work, we present Cognitive Kernel-Pro, a\nfully open-source and (to the maximum extent) free multi-module agent framework\ndesigned to democratize the development and evaluation of advanced AI agents.\nWithin Cognitive Kernel-Pro, we systematically investigate the curation of\nhigh-quality training data for Agent Foundation Models, focusing on the\nconstruction of queries, trajectories, and verifiable answers across four key\ndomains: web, file, code, and general reasoning. Furthermore, we explore novel\nstrategies for agent test-time reflection and voting to enhance agent\nrobustness and performance. We evaluate Cognitive Kernel-Pro on GAIA, achieving\nstate-of-the-art results among open-source and free agents. Notably, our\n8B-parameter open-source model surpasses previous leading systems such as\nWebDancer and WebSailor, establishing a new performance standard for\naccessible, high-capability AI agents. Code is available at\nhttps://github.com/Tencent/CognitiveKernel-Pro","upvotes":94,"discussionId":"6890eddff01a094725f83405","githubRepo":"https://github.com/Tencent/CognitiveKernel-Pro","githubRepoAddedBy":"user","ai_summary":"Cognitive Kernel-Pro is an open-source multi-module agent framework that enhances AI agent robustness and performance through data curation and novel test-time strategies, achieving state-of-the-art results.","ai_keywords":["Agent Foundation Models","queries","trajectories","verifiable answers","agent test-time reflection","agent voting","GAIA","WebDancer","WebSailor"],"githubStars":493},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"657cd228138b7e391444a65d","avatarUrl":"/avatars/c7c984ae483144fab627aa2c54d91d0f.svg","isPro":false,"fullname":"Xiaoyang Wang","user":"xywang1","type":"user"},{"_id":"61b859ddbdf1fac5ed499992","avatarUrl":"/avatars/2387fb9b8a46840bfc75248462f0a410.svg","isPro":false,"fullname":"Jiaqi Chen","user":"judge","type":"user"},{"_id":"65147a1426fbd558dbd08f1b","avatarUrl":"/avatars/86574ee2d5c22e940be1c4e50be88675.svg","isPro":false,"fullname":"Haitao Mi","user":"haitaominlp","type":"user"},{"_id":"641129818573c51c0458b793","avatarUrl":"/avatars/d4bc67c160a07146cf41c614678aa36b.svg","isPro":false,"fullname":"Tianqing Fang","user":"tqfang229","type":"user"},{"_id":"625aa04e535747b1a15cc14b","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1650106391571-noauth.jpeg","isPro":false,"fullname":"Weiqi Wang","user":"mightyweaver","type":"user"},{"_id":"626fbdcca8dcec0066f4ae12","avatarUrl":"/avatars/1e4e174c425943b7dde29c140d322478.svg","isPro":true,"fullname":"Tianshi ZHENG","user":"StoneTZHENG","type":"user"},{"_id":"64670b0e0ed2f7a8cba98aa3","avatarUrl":"/avatars/e8d29d20a149b85850deb4ba781a5ebd.svg","isPro":false,"fullname":"Jiaxin","user":"jbai0318","type":"user"},{"_id":"653e26e09107029eb030c5a2","avatarUrl":"/avatars/5aab2528523db0730dbea38fc228b38a.svg","isPro":false,"fullname":"Baixuan Xu","user":"Moonlight12138","type":"user"},{"_id":"6389b17678e0e042448fcd29","avatarUrl":"/avatars/0bc902ca85593a1c174cf3dcf4305b76.svg","isPro":false,"fullname":"chunyang li","user":"lcy2723","type":"user"},{"_id":"653882396d6743c014db1fe5","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/saCRVks8jnS2Ifs3mT7QW.png","isPro":false,"fullname":"FAN WEI","user":"AlexFanWei","type":"user"},{"_id":"6539dcd629f8a911552b64e5","avatarUrl":"/avatars/983c99a329ac180b196ff0f753039790.svg","isPro":false,"fullname":"Zheye Deng","user":"dzy97","type":"user"},{"_id":"68313679e85918b23d181d47","avatarUrl":"/avatars/85a7b26564365b88469357ed4adb1b40.svg","isPro":false,"fullname":"Liu Yuxuan","user":"xuansenpai","type":"user"}],"acceptLanguages":["*"],"dailyPaperRank":1}">