Robocasa-Mem: Benchmarking Environment and Process Memory in Long-Horizon Mobile Manipulation
Robocasa-Mem: Benchmarking Environment and Process Memory in Long-Horizon Mobile Manipulation
This paper introduces RoboCasa-Mem, a new benchmark for mobile manipulation that evaluates environment memory (spatial locations, object attributes) and process memory (task progress, scheduling). It features Behavior-Conditioned Evaluation to assess consistency with interaction history rather than fixed targets. The release includes 1,800 successful demonstrations and the HMAP baseline policy, which uses hierarchical memory-augmented planning and VLA controllers incorporating geometry data.
The authors introduce RoboCasa-Mem, a benchmark for mobile manipulation that evaluates environment and process memory capabilities, accompanied by the HMAP baseline policy and 1,800 demonstration trajectories.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
Reported numbers
successful demonstrations
1,800
View original evidence
nditioned Evaluation. For tasks with multiple valid execution paths, instead of fixing a single target or trajectory, the evaluator derives later task constraints from the robot's earlier actions. This allows us to evaluate whether later behavior is consistent with the robot's own interaction history while reducing shortcuts based on fixed target assignments. We provide 1,800 successful demonstrations for policy trai
Open source S4mobile manipulation tasks
9
View original evidence
ark comprising nine mobile manipulation tasks that require agents to use interaction history at key decision points. Specifically, the tasks are grouped into two categories of memory: One is environment memory, which captures spatial locations, object attributes, and object relationships. The other is process memory, which captures single-task progress and multi-task scheduling. Particularly, we introduce Behavior-Co
Open source S3
- paper: not_reported
What remains unknown
Dataset licensing and commercial-use rights are not established here.
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
nditioned Evaluation. For tasks with multiple valid execution paths, instead of fixing a single target or trajectory, the evaluator derives later task constraints from the robot's earlier actions. This allows us to evaluate whether later behavior is consistent with the robot's own interaction history while reducing shortcuts based on fixed target assignments. We provide 1,800 successful demonstrations for policy trai
Open source S4
ark comprising nine mobile manipulation tasks that require agents to use interaction history at key decision points. Specifically, the tasks are grouped into two categories of memory: One is environment memory, which captures spatial locations, object attributes, and object relationships. The other is process memory, which captures single-task progress and multi-task scheduling. Particularly, we introduce Behavior-Co
Open source S3
Robocasa-Mem: Benchmarking Environment and Process Memory in Long-Horizon Mobile Manipulation
Open source S1
arXiv:2611.06475v1 Announce Type: new Abstract: In the real world, mobile manipulation robots must perform complex object interactions and make multi-step decisions, often relying on task-relevant information that is no longer observable. However, existing benchmarks provide limited evaluation of such memory-dependent decision-making in mobile manipulation. To address this gap, we introduce RoboCasa-Mem, a new benchm
Open source S2
ning. We also develop a baseline on our benchmark, namely Hierarchical Memory-Augmented Policy(HMAP). Its high-level planner uses structured entity and process memories to generate subtasks. Its low-level VLA controller uses subgoal-conditioned queries to incorporate VGGT-Omega geometry into action generation. Together, RoboCasa-Mem and HMAP baseline support systematic evaluation of memory-dependent behavior and the
Open source S5
development of memory-aware mobile manipulation policies.
Open source S6
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Confirm the required data type and collection setting with the buyer; this source does not establish a complete collection specification.
- Compare the reported units and scope before using these quantities in a budget. Recording hours, sensor-hours and trajectories are different measures.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org