Papers › RVT-2: Learning Precise Manipulation from Few Demonstrations

RVT-2: Learning Precise Manipulation from Few Demonstrations

12 Jun 2024arXiv:2406.08545archive 2025-07-28

Ankit Goyal, Valts Blukis, Jie Xu, Yijie Guo, Yu-Wei Chao, Dieter Fox

In this work, we study how to build a robotic system that can solve multiple 3D manipulation tasks given language instructions. To be useful in industrial and household domains, such a system should be capable of learning new tasks with few demonstrations and solving them precisely. Prior works, like PerAct and RVT, have studied this problem, however, they often struggle with tasks requiring high precision. We study how to make them more effective, precise, and fast. Using a combination of architectural and system-level improvements, we propose RVT-2, a multitask 3D manipulation model that is 6X faster in training and 2X faster in inference than its predecessor RVT. RVT-2 achieves a new state-of-the-art on RLBench, improving the success rate from 65% to 82%. RVT-2 is also effective in the real world, where it can learn tasks requiring high precision, like picking up and inserting plugs, with just 10 demonstrations. Visual results, code, and trained model are provided at: https://robotic-view-transformer-2.github.io/.

PaperPDFCodeCode Syntology ran

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, Syntology's MCP tool lists every function and class Syntology harvested from this paper and whether it ran (how to connect): get_harvested_code_for_paper(arxiv_id="2406.08545")

Code

Syntology Ran 15 of 15 code samples harvested from 1 repository linked to this paper; 0 have no recorded run. Of those that ran: 2 ran · violated contract; 1 ran · our draft was wrong; 12 ran with no contract checked.

By repository: official repository: 15 samples from 1 repository, 15 ran. The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

NVlabs/RVT officialpytorchNOASSERTION report

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.

Code Syntology ran Syntology

15 samples harvested; 15 ran; 0 honoured the contract we drafted; 0 have no recorded run. Read from Syntology's graph 2026-09-24; that is when this build read the record, not when the samples ran.

2ran · violated contract
1ran · our draft was wrong
12ran

Licence: 15 of the 15 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from NVlabs/RVT. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself; “our draft was wrong” and “fixture could not drive it” are failures of Syntology's instrument, not of the code.

Each sample ends with its code_sha256, Syntology's identity for that exact code. An agent fetches the stored sample with Syntology's MCP tool get_code(code_sha256="…") (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples from a repository marked official are listed first. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of four reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; or the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar). Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

add_uni_noi NVlabs/RVT/rvt/mvt/utils.py official repository ran fingerprinted licence not identified · pointer only · 552ffd16f22068d8 · report
batched_index_select NVlabs/RVT/rvt/mvt/renderer.py official repository ran licence not identified · pointer only · a01525ed76586b6b · report
cache_fn NVlabs/RVT/rvt/mvt/attn.py official repository ran · our draft was wrong no licence file found · pointer only · 8458f9ba0361ff43 · report
default NVlabs/RVT/rvt/mvt/attn.py official repository ran · violated contract fingerprinted no licence file found · pointer only · 60fff7c3c400d7ff · report
eval_cls NVlabs/RVT/rvt/models/rvt_agent.py official repository ran licence not identified · pointer only · 27f29abd3e4d0c00 · report
eval_con NVlabs/RVT/rvt/models/rvt_agent.py official repository ran licence not identified · pointer only · c78b038bba8b4c1d · report
eval_con_cls NVlabs/RVT/rvt/models/rvt_agent.py official repository ran licence not identified · pointer only · 40867edc55ab08ae · report
exists NVlabs/RVT/rvt/mvt/attn.py official repository ran · violated contract no licence file found · pointer only · aa5486a3650902d8 · report
normalize_quaternion NVlabs/RVT/rvt/mvt/aug_utils.py official repository ran fingerprinted licence not identified · pointer only · f6a528fb2776e15a · report
perturb_se3 NVlabs/RVT/rvt/mvt/augmentation.py official repository ran licence not identified · pointer only · 09dc9e73dc9e6c80 · report
place_pc_in_cube NVlabs/RVT/rvt/mvt/utils.py official repository ran licence not identified · pointer only · 0c63dffe9fabe5b9 · report
rand_discrete NVlabs/RVT/rvt/mvt/aug_utils.py official repository ran licence not identified · pointer only · ec79cd9a87f20318 · report
rand_dist NVlabs/RVT/rvt/mvt/aug_utils.py official repository ran licence not identified · pointer only · 426bb232645ec17f · report
select_feat_from_hm NVlabs/RVT/rvt/mvt/renderer.py official repository ran licence not identified · pointer only · ffa1b9799883b91f · report
trans_pc NVlabs/RVT/rvt/mvt/utils.py official repository ran licence not identified · pointer only · bdc47824aee91b47 · report

Tasks

Robot ManipulationRobot Manipulation Generalization

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Robot Manipulation RLBench RVT-2 Inference Speed (fps) 20.6 #5 of 18 Archive leaderboard report
Robot Manipulation RLBench RVT-2 Input Image Size 128 #5 of 18 Archive leaderboard report
Robot Manipulation RLBench RVT-2 Succ. Rate (18 tasks, 100 demo/task) 81.4 #5 of 18 Archive leaderboard report
Robot Manipulation RLBench RVT-2 Training Time (V100 x 8 x day) 0.83 #5 of 18 Archive leaderboard report
Robot Manipulation Generalization GEMBench RVT-2 Average Success Rate 44.0 #4 of 6 Archive leaderboard report
Robot Manipulation Generalization GEMBench RVT-2 Average Success Rate (L1) 89.1±0.8 #4 of 6 Archive leaderboard report
Robot Manipulation Generalization GEMBench RVT-2 Average Success Rate (L2) 51.0±2.3 #4 of 6 Archive leaderboard report
Robot Manipulation Generalization GEMBench RVT-2 Average Success Rate (L3) 36.0±2.2 #4 of 6 Archive leaderboard report
Robot Manipulation Generalization GEMBench RVT-2 Average Success Rate (L4) 0.0±0.0 #4 of 6 Archive leaderboard report
Robot Manipulation Generalization The COLOSSEUM RVT2 Average decrease average across all perturbations -19.5 #5 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.

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