Source note

ManipArena: Comprehensive Real-world Evaluation of Reasoning-Oriented Generalist Robot Manipulation

Robot BenchmarkVision Language ActionGeneralist Robot PolicyMobile ManipulationReal To Sim

ManipArena is a real-world benchmark for generalist robot manipulation, aimed at evaluating reasoning-heavy Vision-Language-Action and world-model systems under standardized physical conditions. It builds a shared evaluation setup with diverse tasks, controlled out-of-distribution tests, rich sensor streams, and matched simulation counterparts.

  • Existing robot benchmarks lean on simulation, so they miss real deployment issues such as perception noise, contact dynamics, latency, and hardware limits.
  • Real-world evaluations are often run on different robots and lab setups, which makes comparisons hard to reproduce and hard to interpret.
  • Generalist robot models need tests for semantic reasoning, spatial generalization, and long-horizon mobile manipulation, but prior benchmarks only cover parts of that space.
  • The benchmark defines 20 real-world tasks across execution reasoning, semantic reasoning, and mobile manipulation, backed by 10,812 expert trajectories collected over about 188 hours.
  • It uses a single shared robot embodiment and a server-side inference protocol where each participant submits one model endpoint for all tasks, so scores reflect policy quality rather than custom hardware or per-task specialization.
  • Evaluation runs inside a green-screen enclosed booth with fixed lighting to isolate controlled variables, and uses a stratified 10-trial design per task: T1-T4 in-domain, T5-T8 shifted but in-distribution, T9-T10 semantic OOD when available.
  • Training diversity is designed along three levels: physical attributes, spatial layouts, and semantic composition, with explicit train/test separation for OOD objects.
  • The benchmark also includes rich observations such as joint velocity and motor current, plus Real2Sim environments built with 3D Gaussian Splatting, Hunyuan3D assets, and IsaacLab replay alignment.
  • The excerpt does not provide benchmark scores or baseline performance numbers, so there are no quantitative model comparisons to report.
  • The paper claims broader coverage than prior benchmarks: 20 tasks total, including 10 execution, 5 semantic, and 5 mobile tasks.
  • The dataset contains 10,812 trajectories and about 188 hours of data.
  • Mobile tasks are much longer than tabletop tasks: 2,878 vs. 665 average frames at 20 fps, or about 4.3× longer, and they make up 60.6% of total dataset frames while accounting for 26.7% of trajectories.
  • Tabletop tasks use 56D state/action per frame; mobile tasks use 62D. Full collection records 112D per frame before release-time filtering.
  • Tabletop evaluation uses 15 tasks scored over 10 trials each, with 0-10 points per trial, for up to 100 points per task and 1,500 total points across those evaluated tabletop tasks.