EventVLA：面向長程視覺-語言-行動策略的事件驅動視覺證據記憶

摘要

在長時間跨度的機器人操控任務中，記憶依然是關鍵瓶頸，因為標準的視覺-語言-動作（VLA）策略往往在任務相關線索隨時間被遮蔽或無法觀察時失效。現有的記憶增強方法雖利用了歷史背景，但它們要麼遭受嚴重的資訊瓶頸，要麼透過分離式雙系統導致高延遲，或依賴無選擇性的緩衝區而累積大量視覺冗餘。為了解決這些限制，我們提出 EventVLA——一個基於「稀疏視覺證據記憶」概念的端到端框架，包含兩項核心組件：用於保留初始與短期情境的基礎視覺錨點，以及動態關鍵幀證據記憶（KEM）模組。具體而言，KEM 直接從 VLA 的潛在表徵中預測未來的關鍵幀機率，以自主捕捉並儲存稀疏且任務關鍵的視覺事件。這種前瞻驅動機制使策略能夠動態評估當前觀察的未來因果效用，在其變得不可觀察之前保留瞬態的視覺證據。此外，我們提出 RoboTwin-MeM，一個專門設計用於評估具互動視覺證據的非馬可夫操控任務的診斷基準。廣泛的實驗顯示，在 17 項需要記憶的模擬任務與 4 項真實世界的雙手任務中，EventVLA 相較於最先進的記憶增強 VLA 平均成功率提升了 40%。

English

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.