pith. sign in

arxiv: 2606.11901 · v1 · pith:LVNYPIRPnew · submitted 2026-06-10 · 💻 cs.RO · cs.AI

DuoBench: A Reproducible Benchmark for Bimanual Manipulation in Simulation and the Real World

Tobias J\"ulg , Seongjin Bien , Simon Hilber , Yannik Blei , Pierre Krack , Maximilian Li , Sven Parusel , Rudolf Lioutikov
show 2 more authors
Florian Walter Wolfram Burgard
This is my paper

Pith reviewed 2026-06-27 09:40 UTC · model grok-4.3

classification 💻 cs.RO cs.AI
keywords bimanual manipulationrobot benchmarkingdual-arm coordinationimitation learningsimulation to real transfervision language actionfailure analysispolicy evaluation
0
0 comments X

The pith

DuoBench reveals that current bimanual policies struggle with early interactions, parallel arm execution, and simulation-to-reality transfer.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper introduces DuoBench as a benchmarking framework consisting of eleven tasks that fall into four coordination categories. These tasks are realized both in simulation and on physical hardware through standardized recipes and printable parts. A stage-based evaluation method breaks performance into semantic phases to identify specific failure points rather than relying on overall success rates. Testing of imitation-learning and vision-language-action policies on the benchmark shows persistent difficulties in the early phases of object contact, simultaneous arm movements, and crossing from simulation to real settings. The framework supplies teleoperated datasets to support further development of dual-arm policies.

Core claim

DuoBench comprises eleven tasks spanning four coordination categories, implemented in simulation and partially reproduced in the real world through reproducible task recipes with 3D-printable assets. It includes a stage-based evaluation scheme for fine-grained semantic failure analysis and provides human-teleoperated datasets for all tasks. Benchmarking of dual-arm imitation-learning and vision-language-action policies demonstrates that current methods remain challenged by bimanual manipulation, particularly in early interaction stages, parallel arm execution, and transfer between simulation and real-world settings.

What carries the argument

The stage-based evaluation scheme that decomposes each task into ordered phases to enable semantic diagnosis of coordination failures.

If this is right

  • Dual-arm policies require targeted improvements for the initial phases of object contact.
  • Simultaneous control of both arms must be addressed as a distinct capability.
  • Methods that reduce the performance gap between simulation and physical execution become necessary.
  • Reproducible task definitions allow consistent comparison of new algorithms across different laboratories.
  • Human teleoperation datasets can serve as direct supervision for learning coordinated behaviors.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The stage-based scheme could be applied to single-arm or multi-robot benchmarks to expose analogous phase-specific weaknesses.
  • Failure patterns identified here may point toward policy architectures that explicitly model inter-arm dependencies.
  • Adding tasks with higher degrees of object complexity or environmental variation would test whether the observed challenges scale.
  • Combining the provided datasets with existing single-arm collections could create mixed training regimes for dual-arm learning.

Load-bearing premise

The eleven tasks and four coordination categories together capture the coordination challenges and failure modes that existing benchmarks miss.

What would settle it

A single policy that reaches high success rates in every stage of all eleven tasks in both simulation and the real world without additional coordination-specific training would indicate that the reported challenges are not general.

Figures

Figures reproduced from arXiv: 2606.11901 by Florian Walter, Maximilian Li, Pierre Krack, Rudolf Lioutikov, Seongjin Bien, Simon Hilber, Sven Parusel, Tobias J\"ulg, Wolfram Burgard, Yannik Blei.

Figure 1
Figure 1. Figure 1: Overview of DuoBench: four bimanual task categories with eleven tasks, and four repli [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: FR3 Duo setup. Left shows the real-world setup with printed assets. Right shows the simulated MuJoCo scene. FR3 Duo is a novel dual-arm arrangement of FR3 robotic arms defined by the manufacturer Franka Robotics. The mounting configuration is chosen such that both robots’ ISO cubes over￾lap, ensuring strong dual-arm manipulability. Both arms have a two-finger Robotiq 2F-85 gripper attached. The setup uses … view at source ↗
Figure 3
Figure 3. Figure 3: Fraction of rollouts in simulation that failed in a given [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Average task progress over normalized time across all rollouts in simulation. [PITH_FULL_IMAGE:figures/full_fig_p013_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Fraction of real-world rollouts that ended in a given stage. Green means fraction of [PITH_FULL_IMAGE:figures/full_fig_p013_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Visual ablation examples: original (top-left), object and texture ablation (top-middle), [PITH_FULL_IMAGE:figures/full_fig_p014_6.png] view at source ↗
read the original abstract

Bimanual robot systems substantially expand manipulation capabilities, but coordinating two arms introduces additional control complexity and failure modes that are not well captured by existing benchmarks. We introduce DuoBench, an extensible benchmarking framework for bimanual manipulation policies on the FR3 Duo platform. DuoBench comprises eleven tasks spanning four coordination categories, implemented in simulation and partially reproduced in the real world through reproducible task recipes with 3D-printable assets. In addition, we propose a stage-based evaluation scheme that supports fine-grained semantic failure analysis beyond binary success and provide human-teleoperated datasets for all benchmark tasks. We benchmark several dual-arm imitation-learning and vision-language-action policies in simulation and on real hardware. Our results show that current policies remain challenged by bimanual manipulation, particularly in early interaction stages, parallel arm execution, and transfer between simulation and real-world settings. DuoBench provides a reproducible testbed for diagnosing these failure modes and studying future methods for dual-arm policy learning. Code, datasets, and videos are available at https://duobench.github.io/

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript introduces DuoBench, an extensible benchmarking framework for bimanual manipulation policies on the FR3 Duo platform. It comprises eleven tasks spanning four coordination categories, implemented in simulation with partial real-world reproduction via reproducible recipes and 3D-printable assets. A stage-based evaluation scheme is proposed for fine-grained semantic failure analysis beyond binary success, human-teleoperated datasets are provided for all tasks, and several dual-arm imitation-learning and vision-language-action policies are benchmarked. The central claim is that current policies remain challenged by bimanual manipulation (particularly early interaction stages, parallel arm execution, and sim-to-real transfer) and that DuoBench supplies a reproducible testbed for diagnosing these issues.

Significance. If the tasks and stage-based scheme prove effective at exposing coordination and transfer failure modes not captured by prior benchmarks, and if the reproducibility elements (public code, datasets, and assets) function as described, the work could provide a useful standardized testbed for dual-arm policy research. The inclusion of human datasets and partial real-world validation are positive elements that support imitation learning and sim-to-real studies.

major comments (2)
  1. [Abstract] Abstract: the claim that 'our results show that current policies remain challenged' is asserted without reference to specific quantitative outcomes (e.g., success rates per task or stage, number of trials, or statistical measures), which is load-bearing for the diagnostic value of the benchmark.
  2. [Task and evaluation design] Task and evaluation design (Abstract and § on benchmark tasks): the assertion that the eleven tasks and stage-based scheme capture coordination challenges and failure modes missed by existing benchmarks is central to the contribution but lacks explicit comparative analysis or evidence demonstrating unique diagnostic power relative to prior work.
minor comments (1)
  1. [Reproducibility] Reproducibility section: confirm that all linked assets (code, datasets, 3D models) include complete setup instructions matching the claimed partial real-world reproduction to avoid ambiguity for users.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback. We address the two major comments below and will incorporate revisions to strengthen the manuscript.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the claim that 'our results show that current policies remain challenged' is asserted without reference to specific quantitative outcomes (e.g., success rates per task or stage, number of trials, or statistical measures), which is load-bearing for the diagnostic value of the benchmark.

    Authors: We agree that the abstract would benefit from explicit quantitative support for the central claim. The full manuscript contains the relevant experimental results (success rates by task, stage, and policy type, along with trial counts), but these are not summarized in the abstract. In the revised version we will add concise references to key outcomes, such as aggregate success rates for the evaluated imitation-learning and VLA policies and the number of trials per task, while preserving the abstract's length and focus. revision: yes

  2. Referee: [Task and evaluation design] Task and evaluation design (Abstract and § on benchmark tasks): the assertion that the eleven tasks and stage-based scheme capture coordination challenges and failure modes missed by existing benchmarks is central to the contribution but lacks explicit comparative analysis or evidence demonstrating unique diagnostic power relative to prior work.

    Authors: The manuscript introduces novel tasks spanning four coordination categories and a stage-based evaluation that enables finer-grained failure analysis than binary success metrics common in prior benchmarks. Our benchmarking results highlight specific issues (early-stage interaction, parallel execution, sim-to-real gaps) that arise under these tasks. However, we acknowledge that an explicit side-by-side comparison of diagnostic power versus representative prior benchmarks is not currently present. We will add a short comparative paragraph in the benchmark tasks section that directly contrasts the failure modes surfaced by DuoBench with those reported in existing single-arm or less-coordinated benchmarks, using concrete examples from our data. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

This is an empirical benchmark paper with no mathematical derivations, equations, fitted parameters, or predictions. The eleven tasks, four coordination categories, stage-based evaluation, and policy benchmarks are defined directly from task requirements and external policy implementations; reproducibility rests on linked code, datasets, and 3D assets rather than any self-referential construction. No self-citation load-bearing steps, ansatzes, or renamings appear. The central claims rest on external comparisons to prior benchmarks and observed policy failures, making the work self-contained.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

This is an empirical benchmark paper with no mathematical derivations, fitted parameters, or physical postulates; the contribution consists of defined tasks, categories, and evaluation protocols rather than any derived quantities.

pith-pipeline@v0.9.1-grok · 5749 in / 1116 out tokens · 24933 ms · 2026-06-27T09:40:10.198495+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

45 extracted references · 2 canonical work pages

  1. [1]

    Todorov, T

    E. Todorov, T. Erez, and Y . Tassa. Mujoco: A physics engine for model-based control. In Proc. of the IEEE/RSJ Int. Conf. on Intelligent Robots and Systems (IROS), 2012

    2012

  2. [2]

    Isaac Sim

    NVIDIA. Isaac Sim. URLhttps://github.com/isaac-sim/IsaacSim

  3. [3]

    Xiang, Y

    F. Xiang, Y . Qin, K. Mo, Y . Xia, H. Zhu, F. Liu, M. Liu, H. Jiang, Y . Yuan, H. Wang, L. Yi, A. X. Chang, L. J. Guibas, and H. Su. SAPIEN: A simulated part-based interactive environ- ment. InProc. of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), 2020

    2020

  4. [4]

    James, Z

    S. James, Z. Ma, D. R. Arrojo, and A. J. Davison. RLBench: The robot learning benchmark & learning environment.IEEE Robotics and Automation Letters, 5(2), 2020

    2020

  5. [5]

    T. Yu, D. Quillen, Z. He, R. Julian, K. Hausman, C. Finn, and S. Levine. Meta-World: A benchmark and evaluation for multi-task and meta reinforcement learning. InProc. of the Conf. on Robot Learning (CoRL), 2020

    2020

  6. [6]

    T. Mu, Z. Ling, F. Xiang, D. C. Yang, X. Li, S. Tao, Z. Huang, Z. Jia, and H. Su. ManiSkill: Generalizable Manipulation Skill Benchmark with Large-Scale Demonstrations. InNeural Information Processing Systems Datasets and Benchmarks Track (Round 2), 2021

    2021

  7. [7]

    J. Gu, F. Xiang, X. Li, Z. Ling, X. Liu, T. Mu, Y . Tang, S. Tao, X. Wei, Y . Yao, X. Yuan, P. Xie, Z. Huang, R. Chen, and H. Su. ManiSkill2: A unified benchmark for generalizable manipulation skills. InProc. of the Int. Conf. on Learning Representations (ICLR), 2023

    2023

  8. [8]

    Stone, F

    T. Stone, F. Xiang, A. Shukla, Y . Qin, X. Hinrichsen, X. Yuan, C. Bao, X. Lin, Y . Liu, T.-K. Chan, Y . Gao, X. Li, T. Mu, N. Xiao, A. Gurha, V . N., Y . W. Choi, Y .-R. Chen, Z. Huang, R. Calandra, R. Chen, S. Luo, and H. Su. Demonstrating gpu parallelized robot simulation and rendering for generalizable embodied ai with ManiSkill3. InProc. of Robotics:...

    2025

  9. [9]

    B. Liu, Y . Zhu, C. Gao, Y . Feng, Q. Liu, Y . Zhu, and P. Stone. Libero: Benchmarking knowl- edge transfer for lifelong robot learning. InAdvances in Neural Information Processing Sys- tems, 2023

    2023

  10. [10]

    Y . Zhu, J. Wong, A. Mandlekar, R. Mart´ın-Mart´ın, A. Joshi, K. Lin, A. Maddukuri, S. Nasiri- any, and Y . Zhu. robosuite: A modular simulation framework and benchmark for robot learn- ing.https://arxiv.org/abs/2009.12293, 2025

    Pith/arXiv arXiv 2009

  11. [11]

    S. Fei, S. Wang, J. Shi, Z. Dai, J. Cai, P. Qian, L. Ji, X. He, S. Zhang, Z. Fei, J. Fu, J. Gong, and X. Qiu. LIBERO-Plus: In-depth Robustness Analysis of Vision-Language-Action Models. https://arxiv.org/abs/2510.13626, 2025

    Pith/arXiv arXiv 2025

  12. [12]

    X. Zhou, Y . Xu, G. Tie, Y . Chen, G. Zhang, D. Chu, P. Zhou, and L. Sun. LIBERO-PRO: To- wards Robust and Fair Evaluation of Vision-Language-Action Models Beyond Memorization. https://arxiv.org/abs/2510.03827, 2025. 9

    Pith/arXiv arXiv 2025

  13. [13]

    G. Wang, C. Zhang, Q. Liu, J. Zhang, J. Cai, J. Liu, and X. Liu. LIBERO-X: Robustness Litmus for Vision-Language-Action Models.https://arxiv.org/abs/2602.06556, 2026

    arXiv 2026

  14. [14]

    Nasiriany, A

    S. Nasiriany, A. Maddukuri, L. Zhang, A. Parikh, A. Lo, A. Joshi, A. Mandlekar, and Y . Zhu. RoboCasa: Large-scale simulation of household tasks for generalist robots. InProc. of Robotics: Science and Systems (RSS), 2024

    2024

  15. [15]

    Nasiriany, S

    S. Nasiriany, S. Nasiriany, A. Maddukuri, and Y . Zhu. RoboCasa365: A large-scale simulation framework for training and benchmarking generalist robots. InProc. of the Int. Conf. on Learning Representations (ICLR), 2026

    2026

  16. [16]

    Y . Lee, E. S. Hu, and J. J. Lim. IKEA Furniture Assembly Environment for Long-Horizon Complex Manipulation Tasks. InProc. of the IEEE Int. Conf. on Robotics & Automation (ICRA), 2021

    2021

  17. [17]

    O. Mees, L. Hermann, E. Rosete-Beas, and W. Burgard. CALVIN: A benchmark for language- conditioned policy learning for long-horizon robot manipulation tasks.IEEE Robotics and Automation Letters, 7(3), 2022

    2022

  18. [18]

    Zhang, Z

    S. Zhang, Z. Xu, P. Liu, X. Yu, Y . Li, Q. Gao, Z. Fei, Z. Yin, Z. Wu, Y .-G. Jiang, and X. Qiu. VLABench: A Large-Scale Benchmark for Language-Conditioned Robotics Manipu- lation with Long-Horizon Reasoning Tasks. InProc. of Int. Conf. on Computer Vision (ICCV), 2025

    2025

  19. [19]

    Jiang, A

    Y . Jiang, A. Gupta, Z. Zhang, G. Wang, Y . Dou, Y . Chen, L. Fei-Fei, A. Anandkumar, Y . Zhu, and L. Fan. VIMA: Robot Manipulation with Multimodal Prompts. InProc. of the Int. Conf. on Machine Learning (ICML), 2023

    2023

  20. [20]

    Kumar, R

    V . Kumar, R. Shah, G. Zhou, V . Moens, V . Caggiano, A. Gupta, and A. Rajeswaran. RoboHive: A Unified Framework for Robot Learning. InAdvances in Neural Information Processing Systems, 2023

    2023

  21. [21]

    H. Geng, F. Wang, S. Wei, Y . Li, B. Wang, B. An, H. Lou, C. T. Cheng, P. Li, H. Chen, Y . Liang, Y . Qian, J. Mao, W. Wan, Y . Geng, M. Zhang, J. Lyu, S. Zhao, J. Zhang, C. Xu, J. Zhang, C. Zhao, H. Lu, Y . Ding, R. Gong, Y . Wang, Y . Kuang, R. Wu, B. Jia, H. Dong, S. Huang, Y . Wang, J. Malik, and P. Abbeel. RoboVerse: A unified platform, benchmark and...

    2025

  22. [22]

    Zhang, N

    M. Grotz, M. Shridhar, Y .-W. Chao, T. Asfour, and D. Fox. Twin: Two-handed intelligent benchmark for bimanual manipulation. InProc. of the IEEE Int. Conf. on Robotics & Automa- tion (ICRA), 2025. doi:10.1109/ICRA55743.2025.11128527

    work page doi:10.1109/icra55743.2025.11128527 2025

  23. [23]

    T. Chen, Z. Chen, B. Chen, Z. Cai, Y . Liu, Z. Li, Q. Liang, X. Lin, Y . Ge, Z. Gu, W. Deng, Y . Guo, T. Nian, X. Xie, Q. Chen, K. Su, T. Xu, G. Liu, M. Hu, H. ang Gao, K. Wang, Z. Liang, Y . Qin, X. Yang, P. Luo, and Y . Mu. Robotwin 2.0: A scalable data generator and benchmark with strong domain randomization for robust bimanual robotic manipulation. ht...

    Pith/arXiv arXiv 2025

  24. [24]

    X. Peng, C. Gao, L. Jin, A. Li, and S. Liu. Bicoord: A bimanual manipulation benchmark to- wards long-horizon spatial-temporal coordination.https://arxiv.org/abs/2604.05831, 2026

    Pith/arXiv arXiv 2026

  25. [25]

    Srivastava, C

    S. Srivastava, C. Li, M. Lingelbach, R. Mart ´ın-Mart´ın, F. Xia, K. E. Vainio, Z. Lian, C. Gok- men, S. Buch, K. Liu, S. Savarese, H. Gweon, J. Wu, and L. Fei-Fei. BEHA VIOR: Benchmark for Everyday Household Activities in Virtual, Interactive, and Ecological Environments. In Proc. of the Conf. on Robot Learning (CoRL), 2022. 10

    2022

  26. [26]

    Y . Chen, Y . Geng, F. Zhong, J. Ji, J. Jiang, Z. Lu, H. Dong, and Y . Yang. Bi-DexHands: To- wards human-level bimanual dexterous manipulation.IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(5), 2024

    2024

  27. [27]

    X. Wu, Z. Liang, Y . Ma, M. Hu, Z. Qin, and X. Li. ST-BiBench: Benchmarking Multi-Stream Multimodal Coordination in Bimanual Embodied Tasks for MLLMs.https://arxiv.org/ abs/2602.08392, 2026

    Pith/arXiv arXiv 2026

  28. [28]

    Sferrazza, D.-M

    C. Sferrazza, D.-M. Huang, X. Lin, Y . Lee, and P. Abbeel. HumanoidBench: Simulated hu- manoid benchmark for whole-body locomotion and manipulation. InProc. of Robotics: Sci- ence and Systems (RSS), 2024

    2024

  29. [29]

    Chernyadev, N

    N. Chernyadev, N. Backshall, X. Ma, Y . Lu, Y . Seo, and S. James. BiGym: A Demo-Driven Mobile Bi-Manual Manipulation Benchmark. InProc. of the Conf. on Robot Learning (CoRL), 2025

    2025

  30. [30]

    J. Luo, C. Xu, F. Liu, L. Tan, Z. Lin, J. Wu, P. Abbeel, and S. Levine. FMB: A functional manipulation benchmark for generalizable robotic learning.Int. Journal of Robotics Research (IJRR), 44(4), 2025

    2025

  31. [31]

    M. Heo, Y . Lee, D. Lee, and J. J. Lim. Furniturebench: Reproducible real-world benchmark for long-horizon complex manipulation.Int. Journal of Robotics Research (IJRR), 2023

    2023

  32. [32]

    K. Wu, C. Hou, J. Liu, Z. Che, X. Ju, Z. Yang, M. Li, Y . Zhao, Z. Xu, G. Yang, S. Fan, X. Wang, F. Liao, Z. Zhao, G. Li, Z. Jin, L. Wang, J. Mao, N. Liu, P. Ren, Q. Zhang, Y . Lyu, M. Liu, H. Jingyang, Y . Luo, Z. Gao, C. Li, C. Gu, Y . Fu, D. Wu, X. Wang, S. Chen, Z. Wang, P. An, S. Qian, S. Zhang, and J. Tang. RoboMIND: Benchmark on multi-embodiment in...

    2025

  33. [33]

    Atreya, K

    P. Atreya, K. Pertsch, T. Lee, M. J. Kim, A. Jain, A. Kuramshin, C. Neary, E. S. Hu, K. Arora, K. Ellis, L. Macesanu, M. Leonard, M. Cho, O. Aslan, S. Dass, T. Wang, X. Yuan, A. Gupta, D. Jayaraman, G. Berseth, K. Daniilidis, R. Mart ´ın-Mart´ın, Y . Lee, P. Liang, C. Finn, and S. Levine. RoboArena: Distributed Real-World Evaluation of Generalist Robot Po...

    2025

  34. [34]

    Khazatsky, K

    A. Khazatsky, K. Pertsch, S. Nair, A. Balakrishna, S. Dasari, S. Karamcheti, S. Nasiriany, M. K. Srirama, L. Y . Chen, K. Ellis, P. D. Fagan, J. Hejna, M. Itkina, M. Lepert, Y . J. Ma, P. T. Miller, J. Wu, S. Belkhale, S. Dass, H. Ha, A. Jain, A. Lee, Y . Lee, M. Memmel, S. Park, I. Radosavovic, K. Wang, A. Zhan, K. Black, C. Chi, K. B. Hatch, S. Lin, J. ...

    2024

  35. [35]

    J ¨ulg, P

    T. J ¨ulg, P. Krack, S. Bien, Y . Blei, K. Gamal, K. Nakahara, J. Hechtl, R. Calandra, W. Burgard, and F. Walter. Robot Control Stack: A Lean Ecosystem for Robot Learning at Scale.https: //arxiv.org/abs/2509.14932, 2025

    arXiv 2025

  36. [36]

    R. S. Sutton, A. G. Barto, et al.Reinforcement learning: An introduction, volume 1. MIT press Cambridge, 1998

    1998

  37. [37]

    Franka documentation portal.https://www.franka.de/ documents, 2026

    Franka Robotics GmbH. Franka documentation portal.https://www.franka.de/ documents, 2026. 11

    2026

  38. [38]

    Zakka, Y

    K. Zakka, Y . Tassa, and MuJoCo Menagerie Contributors. MuJoCo Menagerie: A col- lection of high-quality simulation models for MuJoCo, 2022. URLhttp://github.com/ google-deepmind/mujoco_menagerie

    2022

  39. [39]

    Krebs and T

    F. Krebs and T. Asfour. A bimanual manipulation taxonomy.IEEE Robotics and Automation Letters, 7(4), 2022. doi:10.1109/LRA.2022.3196158

    work page doi:10.1109/lra.2022.3196158 2022

  40. [40]

    Towers, A

    M. Towers, A. Kwiatkowski, J. Terry, J. U. Balis, G. D. Cola, T. Deleu, M. Goul ˜ao, A. Kallinteris, M. Krimmel, A. KG, R. Perez-Vicente, A. Pierr´e, S. Schulhoff, J. J. Tai, H. Tan, and O. G. Younis. Gymnasium: A standard interface for reinforcement learning environments. InAdvances in Neural Information Processing Systems, 2025

    2025

  41. [41]

    Jiang, Q

    X. Jiang, Q. Yuan, E. U. Dincer, H. Zhou, G. Li, X. Li, X. Jia, T. Schnizer, N. Schreiber, W. Liao, J. Haag, K. Li, G. Neumann, and R. Lioutikov. IRIS: An immersive robot interaction system. InProc. of the Conf. on Robot Learning (CoRL), volume 305, 2025

    2025

  42. [42]

    J ¨ulg, K

    T. J ¨ulg, K. Gamal, N. Nilavadi, P. Krack, S. Bien, M. Krawez, F. Walter, and W. Burgard. VLAgents: A Policy Server for Efficient VLA Inference.https://arxiv.org/abs/2601. 11250, 2026

    2026

  43. [43]

    T. Zhao, V . Kumar, S. Levine, and C. Finn. Learning fine-grained bimanual manipulation with low-cost hardware. InProc. of Robotics: Science and Systems (RSS), 2023

    2023

  44. [44]

    Intelligence, K

    P. Intelligence, K. Black, N. Brown, J. Darpinian, K. Dhabalia, D. Driess, A. Esmail, M. Equi, C. Finn, N. Fusai, M. Y . Galliker, D. Ghosh, L. Groom, K. Hausman, B. Ichter, S. Jakubczak, T. Jones, L. Ke, D. LeBlanc, S. Levine, A. Li-Bell, M. Mothukuri, S. Nair, K. Pertsch, A. Z. Ren, L. X. Shi, L. Smith, J. T. Springenberg, K. Stachowicz, J. Tanner, Q. V...

    2025

  45. [45]

    A”,π0.5 as “π

    J. Zheng, J. Li, Z. Wang, D. Liu, X. Kang, Y . Feng, Y . Zheng, J. Zou, Y . Chen, J. Zeng, Y .-Q. Zhang, J. Pang, J. Liu, T. Wang, and X. Zhan. X-VLA: Soft-prompted transformer as scalable cross-embodiment vision-language-action model.https://arxiv.org/abs/2510.10274, 2025. 12 A Further metrics 0 1 3 T ask Progress Hinge-Chest 0 1 2 3 Spring-Door 0 1 6 Po...

    Pith/arXiv arXiv 2025