arXiv：cs.AI· Soumil Rathi·· 13 小时前
RoboMME-Interference：干扰环境下的机器人记忆基准测试
RoboMME-Interference: Benchmarking Robot Memory Under Interference
AI 导读
研究者提出跨会话基准 RoboMME-Interference，在 RoboMME 基础上为每次查询构造包含相关示范与若干无关会话的历史，用于测试 VLA 的长上下文记忆能力。运行 RoboMME 已发布的记忆增强 π₀.₅ 变体发现，感知记忆在无干扰时提升成功率，但随无关会话累积而持续衰减；子目标记忆提升较小却更稳健，为最强感知变体加入检索步骤后可在各干扰水平恢复无干扰成功率。
正文
Abstract:Robots deployed in realistic settings will accumulate experience across many sessions and tasks over their deployment. The robot's tasks may often require it to remember information from multiple sessions ago, making long-context robot memory important for real-world deployments. However, most robot-memory benchmarks today are based on single episodes or a short context. To measure how current robot memory systems perform on longer sessions with more distractions, we introduce RoboMME-Interference, a cross-session benchmark built on RoboMME (Dai et al., 2026). For each query episode, we construct a session history using the query's relevant prior demonstration followed by a controlled number of unrelated sessions, which we provide to the VLA as memory and measure accuracy. Running RoboMME's released memory-augmented $\pi_{0.5}$ variants unmodified through this benchmark, we find that while perceptual memory variants improve success when given the history without any distractors, they decay strongly and steadily as unrelated sessions accumulate. The subgoal variants, which read the history with a vision-language model and pass written subgoals to the policy, improve less at their best but hold more of that improvement as distractors accumulate. Adding a retrieval step to the strongest perceptual variant, which selects the section of history most visually similar to the robot's current view and passes only that section to the policy, restores its no-distractor success rate at every interference level. With this release, we emphasize the importance of long-context memory and robustness to interference and show that current systems largely fail on such capabilities. The project page, videos, code, and data are at this https URL.
| Comments: | 9 pages, 5 figures. Updated results; added subgoal-memory systems |
| Subjects: | Robotics (cs.RO); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2606.22338 [cs.RO] |
| (or arXiv:2606.22338v4 [cs.RO] for this version) | |
| https://doi.org/10.48550/arXiv.2606.22338 arXiv-issued DOI via DataCite |
Submission history
From: Soumil Rathi [view email]
[v1]
Sun, 21 Jun 2026 05:06:16 UTC (54 KB)
[v2]
Tue, 28 Jul 2026 00:23:57 UTC (62 KB)
[v3]
Wed, 26 Aug 2026 09:52:30 UTC (72 KB)
[v4]
Fri, 9 Oct 2026 16:43:07 UTC (68 KB)
来源：arXiv：cs.AI · arxiv.org