Papers › Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation
Perceiver-Actor: A Multi-Task Transformer for Robotic Manipulation
Mohit Shridhar, Lucas Manuelli, Dieter Fox
Transformers have revolutionized vision and natural language processing with their ability to scale with large datasets. But in robotic manipulation, data is both limited and expensive. Can manipulation still benefit from Transformers with the right problem formulation? We investigate this question with PerAct, a language-conditioned behavior-cloning agent for multi-task 6-DoF manipulation. PerAct encodes language goals and RGB-D voxel observations with a Perceiver Transformer, and outputs discretized actions by ``detecting the next best voxel action''. Unlike frameworks that operate on 2D images, the voxelized 3D observation and action space provides a strong structural prior for efficiently learning 6-DoF actions. With this formulation, we train a single multi-task Transformer for 18 RLBench tasks (with 249 variations) and 7 real-world tasks (with 18 variations) from just a few demonstrations per task. Our results show that PerAct significantly outperforms unstructured image-to-action agents and 3D ConvNet baselines for a wide range of tabletop tasks.
In Syntology View this paper on Syntology, its page in Syntology's graph. That page lists the repositories linked to the paper, the abstract and the calls for agents.
Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
For agents, on Syntology's MCP service (how to connect):
get_citation_path(paper_1="2209.05451", paper_2="…")with another paper's arXiv id or titleget_concepts_for_paper(arxiv_id="2209.05451")
Code
Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.
A paper named beside a repository with no sample linked to this paper is shown with that paper's own run record, not this paper's: “ran” means executed on a synthesized input, not that the code is correct, and “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Papers are listed in arXiv-id order, at most three per repository.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Robot Manipulation | RLBench | PerAct (Evaluated in RVT) | Inference Speed (fps) | 4.9 | #11 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct (Evaluated in RVT) | Input Image Size | 128 | #11 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct (Evaluated in RVT) | Succ. Rate (18 tasks, 100 demo/task) | 49.4 | #11 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct (Evaluated in RVT) | Training Time (V100 x 8 x day) | 16 | #11 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct | Input Image Size | 128 | #14 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct | Succ. Rate (18 tasks, 10 demo/task) | 30 | #14 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct | Succ. Rate (18 tasks, 100 demo/task) | 42.7 | #14 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | PerAct | Training Time (V100 x 8 x day) | 16 | #14 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | Image-BC VIT | Input Image Size | 128 | #16 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | Image-BC VIT | Succ. Rate (18 tasks, 100 demo/task) | 1.3 | #16 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | Image-BC CNN | Input Image Size | 128 | #17 of 18 | Archive leaderboard | report |
| Robot Manipulation | RLBench | Image-BC CNN | Succ. Rate (18 tasks, 100 demo/task) | 1.3 | #17 of 18 | Archive leaderboard | report |
| Robot Manipulation Generalization | The COLOSSEUM | PerAct | Average decrease average across all perturbations | -17.3 | #6 of 9 | Archive leaderboard | report |
Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.
Methods
Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections