Joint Agent Memory and Exploration Learning (JAMEL) framework trains memory and exploration policies together through novelty-driven interaction, enabling effective exploration in open-ended environments with reduced computational costs.</p>\n","updatedAt":"2026-06-02T05:03:10.244Z","author":{"_id":"6673a56dc1ccf20bff21ac54","avatarUrl":"/avatars/584ea67f520cff7dc66596263b49f37a.svg","fullname":"Rui Kong","name":"monster119120","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false,"primaryOrg":{"avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/64f187a2cc1c03340ac30498/TYYUxK8xD1AxExFMWqbZD.png","fullname":"BAIDU","name":"baidu","type":"org","isHf":false,"plan":"team"}}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8456374406814575},"editors":["monster119120"],"editorAvatarUrls":["/avatars/584ea67f520cff7dc66596263b49f37a.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.01528","authors":[{"_id":"6a1e63b6808ddbc3c7d43e3a","name":"Shizuo Tian","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e3b","name":"Xiaohong Weng","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e3c","name":"Rui Kong","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e3d","name":"Yuxuan Chen","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e3e","name":"Guohong Liu","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e3f","name":"Yuebing Song","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e40","name":"Jiacheng Liu","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e41","name":"Yuchen Li","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e42","name":"Dawei Yin","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e43","name":"Ting Cao","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e44","name":"Yunxin Liu","hidden":false},{"_id":"6a1e63b6808ddbc3c7d43e45","name":"Yuanchun Li","hidden":false}],"publishedAt":"2026-06-01T00:00:00.000Z","submittedOnDailyAt":"2026-06-02T00:00:00.000Z","title":"Joint Agent Memory and Exploration Learning via Novelty Signals","submittedOnDailyBy":{"_id":"6673a56dc1ccf20bff21ac54","avatarUrl":"/avatars/584ea67f520cff7dc66596263b49f37a.svg","isPro":false,"fullname":"Rui Kong","user":"monster119120","type":"user","name":"monster119120"},"summary":"In open-ended environments, exploration is fundamental for autonomous agents, yet current language model agents struggle with this. Effective exploration requires memory, but retaining raw interaction histories is computationally expensive over long trajectories. While latent memory offers a solution to compress interaction histories, its training lacks reliable supervisory signals. We introduce Joint Agent Memory and Exploration Learning (JAMEL), a framework that trains agentic memory and exploration policy together through novelty-driven interaction. We observe that memory and exploration form a mutually dependent loop: sustained exploration requires memory to distinguish exhausted behaviors from unseen ones, while novelty-seeking interaction provides the supervision needed to make memory useful for future exploration. By utilizing deterministic and persistent novelty signals such as code coverage in the GUI domain, we provide natural, annotation-free supervision for the memory module. Empirical evaluations demonstrate that \\ours successfully generalizes to unseen environments. Its exploration capability outperforms open-weight baselines and rivals the exploration depth of a closed-source model while reducing token consumption. Our code and model are open-sourced at https://github.com/MobileLLM/JAMEL.","upvotes":4,"discussionId":"6a1e63b7808ddbc3c7d43e46","projectPage":"https://github.com/MobileLLM/JAMEL","githubRepo":"https://github.com/MobileLLM/JAMEL","githubRepoAddedBy":"user","ai_summary":"Joint Agent Memory and Exploration Learning (JAMEL) framework trains memory and exploration policies together through novelty-driven interaction, enabling effective exploration in open-ended environments with reduced computational costs.","ai_keywords":["agent memory","exploration policy","novelty-driven interaction","latent memory","deterministic novelty signals","persistent novelty signals","code coverage","open-ended environments","token consumption","closed-source model"],"githubStars":2},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"6673a56dc1ccf20bff21ac54","avatarUrl":"/avatars/584ea67f520cff7dc66596263b49f37a.svg","isPro":false,"fullname":"Rui Kong","user":"monster119120","type":"user"},{"_id":"69bce174b0b4d685f7c101e6","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/noauth/XZhH6OyfcuVMU261TUoM_.jpeg","isPro":false,"fullname":"ZHANG Yutong","user":"wrodriguez509","type":"user"},{"_id":"61e52be53d6dbb1da842316a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/61e52be53d6dbb1da842316a/gx0WGPcOCClXPymoKglc4.jpeg","isPro":false,"fullname":"Börje Karlsson","user":"tellarin","type":"user"},{"_id":"63ca8e060609f1def7e6548a","avatarUrl":"/avatars/1da7947840cb87d5f77c0af9ee11f9c2.svg","isPro":true,"fullname":"Yi Jung","user":"YJ-142150","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.01528.md"}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.01528 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.01528 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.01528 in a Space README.md to link it from this page.

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