Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.10921 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910209567358976 |
|---|---|
| author | Lei, Huashuo Song, Wenxuan Zhang, Huarui Pei, Jieyuan Chen, Jiayi Yan, Haodong Zhao, Han Ding, Pengxiang Zhang, Zhipeng Huang, Lida Wang, Donglin Wang, Yan Li, Haoang |
| author_facet | Lei, Huashuo Song, Wenxuan Zhang, Huarui Pei, Jieyuan Chen, Jiayi Yan, Haodong Zhao, Han Ding, Pengxiang Zhang, Zhipeng Huang, Lida Wang, Donglin Wang, Yan Li, Haoang |
| contents | Memory is a critical component of robotic intelligence, as robots must rely on past observations and actions to accomplish long-horizon tasks in partially observable environments. However, existing robotic memory benchmarks still lack multimodal annotations for memory formation, provide limited task coverage and structural complexity, and remain restricted to simulation without real-world evaluation. We address this gap with RoboMemArena, a large-scale benchmark of 26 tasks, with average trajectory lengths exceeding 1,000 steps per task and 68.9% of subtasks being memory-dependent. The generation pipeline leverages a vision-language model (VLM) to design and compose subtasks, generates full trajectories through atomic functions, and provides memory-related annotations, including subtask instructions and native keyframe annotations, while paired real-world memory tasks support physical evaluation. We further design PrediMem, a dual-system VLA in which a high-level VLM planner manages a memory bank with recent and keyframe buffers and uses a predictive coding head to improve sensitivity to task dynamics. Extensive experiments on RoboMemArena show that PrediMem outperforms all baselines and provides insights into memory management, model architecture, and scaling laws for complex memory systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_10921 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RoboMemArena: A Comprehensive and Challenging Robotic Memory Benchmark Lei, Huashuo Song, Wenxuan Zhang, Huarui Pei, Jieyuan Chen, Jiayi Yan, Haodong Zhao, Han Ding, Pengxiang Zhang, Zhipeng Huang, Lida Wang, Donglin Wang, Yan Li, Haoang Robotics Memory is a critical component of robotic intelligence, as robots must rely on past observations and actions to accomplish long-horizon tasks in partially observable environments. However, existing robotic memory benchmarks still lack multimodal annotations for memory formation, provide limited task coverage and structural complexity, and remain restricted to simulation without real-world evaluation. We address this gap with RoboMemArena, a large-scale benchmark of 26 tasks, with average trajectory lengths exceeding 1,000 steps per task and 68.9% of subtasks being memory-dependent. The generation pipeline leverages a vision-language model (VLM) to design and compose subtasks, generates full trajectories through atomic functions, and provides memory-related annotations, including subtask instructions and native keyframe annotations, while paired real-world memory tasks support physical evaluation. We further design PrediMem, a dual-system VLA in which a high-level VLM planner manages a memory bank with recent and keyframe buffers and uses a predictive coding head to improve sensitivity to task dynamics. Extensive experiments on RoboMemArena show that PrediMem outperforms all baselines and provides insights into memory management, model architecture, and scaling laws for complex memory systems. |
| title | RoboMemArena: A Comprehensive and Challenging Robotic Memory Benchmark |
| topic | Robotics |
| url | https://arxiv.org/abs/2605.10921 |