Hi everyone, please see our latest work, EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policies</p>\n","updatedAt":"2026-06-24T08:55:02.750Z","author":{"_id":"6565d7149afd51867e55520b","avatarUrl":"/avatars/027b17651e61df598af53f69b92e7771.svg","fullname":"Ganlin Yang","name":"ganlinyang","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":3,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.5420696139335632},"editors":["ganlinyang"],"editorAvatarUrls":["/avatars/027b17651e61df598af53f69b92e7771.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.20092","authors":[{"_id":"6a38a0b3db23715e9da13901","name":"Ganlin Yang","hidden":false},{"_id":"6a38a0b3db23715e9da13902","name":"Zhangzheng Tu","hidden":false},{"_id":"6a38a0b3db23715e9da13903","name":"Yuqiang Yang","hidden":false},{"_id":"6a38a0b3db23715e9da13904","name":"Sitong Mao","hidden":false},{"_id":"6a38a0b3db23715e9da13905","name":"Junyi Dong","hidden":false},{"_id":"6a38a0b3db23715e9da13906","name":"Tianxing Chen","hidden":false},{"_id":"6a38a0b3db23715e9da13907","name":"Jiaqi Peng","hidden":false},{"_id":"6a38a0b3db23715e9da13908","name":"Jing Xiong","hidden":false},{"_id":"6a38a0b3db23715e9da13909","name":"Jiafei Cao","hidden":false},{"_id":"6a38a0b3db23715e9da1390a","name":"Jifeng Dai","hidden":false},{"_id":"6a38a0b3db23715e9da1390b","name":"Wengang Zhou","hidden":false},{"_id":"6a38a0b3db23715e9da1390c","name":"Yao Mu","hidden":false},{"_id":"6a38a0b3db23715e9da1390d","name":"Tai Wang","hidden":false}],"publishedAt":"2026-06-18T00:00:00.000Z","submittedOnDailyAt":"2026-06-24T00:00:00.000Z","title":"EventVLA: Event-Driven Visual Evidence Memory for Long-Horizon Vision-Language-Action Policies","submittedOnDailyBy":{"_id":"6565d7149afd51867e55520b","avatarUrl":"/avatars/027b17651e61df598af53f69b92e7771.svg","isPro":false,"fullname":"Ganlin Yang","user":"ganlinyang","type":"user","name":"ganlinyang"},"summary":"Memory remains a critical bottleneck for long-horizon robotic manipulation, as standard Vision-Language-Action (VLA) policies often fail when task-relevant cues become occluded or unobservable over time. While existing memory-augmented methods utilize historical context, they either suffer from severe information bottlenecks, incur high latency via decoupled dual systems, or rely on unselective buffers that accumulate massive visual redundancies. To address these limitations, we introduce EventVLA, an end-to-end framework founded on the concept of sparse visual evidence memory that comprises two core components: foundational visual anchors to retain initial and short-term contexts, and a dynamic Keyframe Evidence Memory (KEM) module. Specifically, KEM directly predicts future keyframe probabilities from the VLA's latent embeddings to autonomously capture and store sparse, task-critical visual events. This foresight-driven mechanism empowers the policy to dynamically evaluate the future causal utility of current observations, preserving transient visual evidence before it becomes unobservable. Furthermore, we propose RoboTwin-MeM, a diagnostic benchmark specifically designed to evaluate non-Markovian manipulation tasks with interactive visual evidence. Extensive evaluations show that across 17 memory-requiring simulation tasks and 4 real-world bimanual tasks, EventVLA achieves an average success rate improvement of +40% over state-of-the-art memory-augmented VLAs.","upvotes":1,"discussionId":"6a38a0b4db23715e9da1390e","projectPage":"https://ganlin-yang.github.io/EventVLA.github.io/","githubRepo":"https://github.com/InternRobotics/EventVLA","githubRepoAddedBy":"user","ai_summary":"EventVLA addresses long-horizon robotic manipulation challenges by introducing a sparse visual evidence memory framework with visual anchors and dynamic Keyframe Evidence Memory module for improved task performance.","ai_keywords":["Vision-Language-Action","memory-augmented methods","visual anchors","Keyframe Evidence Memory","latent embeddings","causal utility","visual evidence","non-Markovian manipulation tasks","diagnostic benchmark"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":17,"organization":{"_id":"6747ee5decec679eafb90450","name":"ShanghaiAiLab","fullname":"shanghai ailab ","avatar":"https://www.gravatar.com/avatar/6cd2acf412ad103653d9ce14a1aacc19?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"organization":{"_id":"6747ee5decec679eafb90450","name":"ShanghaiAiLab","fullname":"shanghai ailab ","avatar":"https://www.gravatar.com/avatar/6cd2acf412ad103653d9ce14a1aacc19?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.20092.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.20092 in a model README.md to link it from this page.

Cite arxiv.org/abs/2606.20092 in a dataset README.md to link it from this page.

Cite arxiv.org/abs/2606.20092 in a Space README.md to link it from this page.