Pith. sign in

REVIEW 5 major objections 6 minor 1 cited by

Bootstrapping Imitation Learning for Long-horizon Manipulation via Hierarchical Data Collection Space

T0 review · 5 major / 6 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Segmenting a robot task into atomic subtasks and collecting each one from varied start poses around the target—the HD-Space scheme—lets imitation learning policies train from smaller datasets while improving success rates on long-horizon…

desk verdict HD-Space is a promising data-collection idea, but the experiments currently conflate adding short clips with the hierarchical mechanism. read the letter →

arxiv 2505.17389 v1 pith:NCSZZPGF submitted 2025-05-23 cs.RO cs.AI

Jinrong Yang , Kexun Chen , Zhuoling Li , Shengkai Wu , Yong Zhao , Liangliang Ren , Wenqiu Luo , Chaohui Shang
show 4 more authors
Meiyu Zhi Linfeng Gao Mingshan Sun Hui Cheng
This is my paper
classification cs.ROcs.AI MSC 68T4068T05
keywords imitationlearningroboticmanipulationdatacollectionspacelong-horizonhierarchicaltasksegmentationbehaviorcloningactionchunkingtransformer
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

The paper proposes HD-Space, a way of collecting human demonstrations for imitation learning that splits a long-horizon manipulation task into a sequence of atomic subtasks and records each subtask separately, starting from several hand-chosen poses around the target. The claim is that these atomic start poses expose the policy to the states that actually accumulate errors during execution, so the trained model becomes more robust without needing more demonstrations. If this is right, the cost bottleneck of imitation learning is partly a data-collection problem, not just a model-size or data-volume problem: smaller, smarter datasets can train policies that complete longer task sequences. The paper supports the claim with experiments on two simulated and five real-world long-horizon tasks, comparing HD-Space data against the naive full-trajectory collection method at matched episode counts.

What carries the argument

The central object is the atomic data collection space: for each hand-selected atomic subtask, the robot's end-effector is placed at a set of start poses evenly distributed around the target point, and human demonstration is recorded from those poses until the atomic action ends. The overlapping boundaries between consecutive atomic spaces are what let the individual chunks chain into a full long-horizon execution. This design does the work of turning a single narrow trajectory distribution into a broader, proactive coverage of the states where compounding errors typically happen.

What would settle it

Run a controlled comparison on a single long-horizon task where the object's initial pose is moved to positions that fall outside the region spanned by the evenly placed start poses, keeping all other collection details identical; the claim that HD-Space covers error-prone states predicts a large success-rate drop for these out-of-reach configurations, while the naive full-trajectory method should degrade similarly. A result showing equal performance would falsify the coverage mechanism.

Watch

Extended reading notes

Core claim

The central discovery is that where you press 'record' in human demonstration matters as much as how many times you record. HD-Space segments a fine manipulation task into semantically meaningful atomic tasks (e.g., open box, pick up cup, put in box, close box). For each atomic task, the human demonstrator starts the robot end-effector at several evenly placed poses around the target—rather than at one fixed start pose—and records until the atomic action completes. Consecutive atomic spaces overlap at their boundaries, which the paper argues ensures that long-horizon sequences chain smoothly. This yields a dataset that covers more of the state/action space that is error-prone during execution, so the trained policy 'sees as many states around the target point as possible' and suppresses erroneous predicted trajectories. Across their evaluations, HD-Space data improves success rates, increases the mean number of completed sequential subtasks, and reduces the number of frames recorded per episode, making the method both more robust and cheaper than naive collection.

Load-bearing premise

The method relies on a human correctly choosing the atomic task boundaries and the spread of start poses around each target; if those choices miss states that matter, the collected data will not cover the genuinely error-prone region, and the supposed advantage over full-trajectory collection collapses.

Editorial extensions

If this is right

  • At matched training episode counts, HD-Space data improves success rate over naive full-trajectory data by 8 percentage points on the teacup-into-box task, by 18–32 points on mobile bowl grasping, and by 20–44 points on spoon grasping.
  • On mobile grasping tasks, 25 naive plus 25 HD-Space episodes outperform 100 naive episodes, indicating the method extracts more useful signal per episode.
  • HD-Space reduces the number of frames recorded per demonstration (e.g., from 600 to 400 for the teacup task and from 1300 to 120 for grabbing electronic pens), which lowers human demonstration cost directly.
  • The average number of completed sequential subtasks increases when HD-Space data is added, so the trained policy extends to longer-horizon execution.
  • Because the scheme is a data-collection change rather than a network change, the gains are expected to combine with stronger policy architectures such as those using richer visual features.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the coverage mechanism is the real driver, HD-Space is effectively a state-space prior; the natural next step is to derive the sampling radius and density of start poses from the task geometry or from the current policy's failure modes rather than hand-tuning them.
  • A testable extension is to measure the distribution of policy confidence (or action variance) at the overlapping boundary states; the paper's chaining argument predicts these boundaries are the most failure-prone points and should receive proportionally more demonstrations.
  • HD-Space could act as a cheap bootstrap for human-in-the-loop and simulation-based pipelines, since a stronger baseline shortens the error-waiting phase; the paper notes orthogonality to these approaches but does not test the combination.
  • The method's benefit seems to scale with task difficulty: the largest gains appear in tasks requiring precise angle, height, and speed matching, suggesting a quantitative relation could be established between the size of the error-prone region and the number of start poses needed.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. This paper proposes HD-Space, a data collection scheme for imitation learning in long-horizon manipulation. The method decomposes a full task into atomic subtasks, collects human demonstrations that start from varied end-effector poses around each subtask target, and then trains a single policy on a mixture of naive full-task demonstrations and these atomic segments. The authors evaluate the scheme with ACT-style policies on real-world tasks (teacup-in-box, mobile conveyor bowl/spoon grasping, unordered electronic-pen grasping), reporting higher success rates and more completed subtasks under equal episode counts, as well as fewer frames per episode. Two simulation tasks are used only to validate a stronger backbone (ACT+DINOv2+MAE), not the HD-Space mechanism itself.

Significance. If validated, HD-Space would be a simple, practical data-collection protocol that reduces demonstration cost and improves robustness for long-horizon imitation learning. The paper's strengths include real-robot experiments across several task families, a concrete hyperparameter appendix, and a public project page with videos. However, the current evidence does not isolate the proposed mechanism: all comparisons mix full-task demonstrations with atomic segments, episode units are not comparable because HD-Space segments are much shorter, and the claimed composition mechanism is never directly measured. The abstract's claim of evaluation across two simulated and five real-world tasks also overstates what is presented.

major comments (5)
  1. [Section 4.2, Figure 5 and Table 2] All HD-Space conditions contain 25 naive full-task episodes (N25/H25, N25/H50, N25/H75), so the comparison is between full demonstrations only and full demonstrations plus atomic segments. This does not isolate the hierarchical-atomic-space mechanism; the observed gains could come from adding short, varied clips to a base of full-task demonstrations. The central claim that HD-Space alone enables longer-horizon tasks with fewer demonstrations requires a pure-HD condition (e.g., H50, H75, or H100 with no naive episodes) or an appropriate control matched for data content.
  2. [Section 4.2, Table 3 vs Figure 5] Episode counts are not comparable across methods: for the electronic-pen task, a naive episode has 1300 frames while an HD-Space episode has 120 frames. Equal episode counts therefore imply substantially fewer total frames for the HD-Space conditions, and 'fewer episodes' conflates episode length with data volume. The success-rate comparisons should be reported against matched total frame counts or otherwise analyzed per frame.
  3. [Section 3, 'overlapping boundaries'] The claim that overlapping atomic-space boundaries 'ensure the smooth execution of extended manipulation sequences' is asserted but not verified. No rollout-level statistic shows that the terminal states of atomic segments fall inside the start-pose distribution of the following segment, nor that errors do not accumulate at segment transitions. The paper should provide boundary-overlap coverage statistics or an end-to-end rollout analysis of how atomic chunks compose.
  4. [Section 4.1, Table 1, and Abstract] The simulation experiments evaluate only the model backbone (ACT vs ACT+DINOv2+MAE) and do not evaluate HD-Space itself. The abstract's statement that the method was evaluated 'across two simulated and five real-world long-horizon manipulation tasks' is therefore misleading. In addition, Ref. [50] (ADC) is described as the closest prior method, but no comparison to ADC is run, and the paper gives no justification for omitting it as a baseline.
  5. [Figures 5 and Table 2] Success rates and completed-subtask counts are reported without error bars, standard deviations, or trial counts. Given that some reported differences are as small as 8 percentage points, the results need variance information and, ideally, statistical tests across seeds or evaluation trials before the claimed improvements can be assessed.
minor comments (6)
  1. [Abstract and Section 4.1] The abstract says five real-world tasks, while Section 4.1 says three; clarify the count, for example by treating the mobile conveyor task with its speed conditions as separate task variants.
  2. [Section 3, first paragraph] The phrase 'segments tdivides' appears to be a typo; it should read 'segments' or 'divides'.
  3. [Section 3, Eq. (1)] The notation for the action sequence horizon N and the subscript on the expectation is implicit; define these explicitly.
  4. [Section 3, atomic segmentation and sampling] The protocol for choosing atomic task boundaries and for 'evenly placing' start poses around the target is underspecified; report the sampling radius, density, and any validation that the segmentation captures decision-relevant structure.
  5. [Table 2] The heading 'Max lenght' should be 'Max length'.
  6. [Section 4.1, conveyor task] The claim that Ref. [62] supports the conveyor-belt speed challenge is unclear, since [62] appears to be a streaming-perception paper rather than a manipulation source; please clarify or replace the reference.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: HD-Space's gains are empirical rollout results; the two self-citations are minor and non-load-bearing.

full rationale

The paper's central claim is that collecting demonstrations in hierarchically segmented atomic state/action spaces improves imitation-learning policy performance, especially for long-horizon manipulation. This is an empirical claim evaluated by real-robot and simulator rollouts, not a derivation that reduces to its inputs. The training loss in Eq. (1) is standard behavior cloning, and the HD-Space method is a data-collection protocol: segment the task into atomic subtasks and sample start poses around each target. No parameter is fitted to a subset of data and then reported as a prediction; no uniqueness theorem is imported; and no known result is merely renamed. The paper acknowledges in its Limitation section that HD-Space is currently a single-task data-collection scheme, which is a scope limitation rather than a circular step. The two self-citations (Refs. [58] and [62]) are not load-bearing for the central claim: Ref. [58] is cited only to support the choice of DINOv2-small as a visual backbone, and Table 1 independently demonstrates that backbone improvement on external ACT benchmarks before any HD-Space data is used; Ref. [62] merely motivates conveyor-belt speed challenges. The assertion that overlapping boundaries "ensure the smooth execution of extended manipulation sequences" is an untested hypothesis, and the lack of a pure-HD control and boundary-overlap statistics is a correctness or experimental-design concern, but it is not circularity because no equation or construction makes the claimed success equivalent to the segmentation definitions. The observed gains could in principle be explained by adding short, uniform clips to full-task demonstrations, but that is an alternative empirical explanation, not a reduction by construction.

Assumptions & free parameters 3 free parameters · 6 assumptions · 0 invented entities

No free parameters are fit to data in a mathematical model. However, the method's effectiveness depends on unquantified design choices (subtask boundaries, start-pose coverage, episode ratios), and the central robustness hypothesis is assumed. The policy network uses standard hyperparameters borrowed from ACT, so no extra invented entities appear.

free parameters (3)
  • Atomic task decomposition (number and boundaries of subtasks)
    Hand-defined per task in Section 3 (teacup task split into open box, pick cup, put in box, close box). No automated criterion is given, and the method's benefit depends on this choice.
  • Starting-pose sampling distribution for each atomic space
    Section 3 says the end effector is 'evenly placed around the target point', but the radius, number, and density of start poses are unspecified. This distribution defines the data coverage.
  • Episode ratios N25/H25, N25/H50, N25/H75 = 25/25, 25/50, 25/75
    Used as ablations in Fig. 5 and Table 2; not fitted to data, but the reported improvements are conditional on these ratios.
assumptions (6)
  • standard math Supervised behavior cloning loss (Eq. 1) is the appropriate objective for learning manipulation policies.
    Standard IL formulation, not derived in the paper.
  • domain assumption The ACT transformer with DINOv2 features and MAE augmentation is a sufficient policy class.
    Used as the evaluation model (Section 3, Data Usage); no proof that results transfer to other IL architectures.
  • domain assumption Expert teleoperated demonstrations are successful and accurate enough to train on.
    The pipeline assumes demonstrations are high quality (Section 3), but no filtering or quality metric is provided.
  • ad hoc to paper Covering states around atomic target points improves robustness and reduces compounding errors.
    This is the central hypothesis (Section 3, Practical Considerations), assumed rather than derived or tested in isolation.
  • ad hoc to paper Overlapping boundaries between atomic spaces guarantee smooth chaining of subtasks.
    Section 3 asserts overlapping boundaries 'ensure the smooth execution of extended manipulation sequences', but no mechanism or rollout analysis is provided.
  • ad hoc to paper The atomic action segmentation criterion (locating a pose then executing a fine operation) captures decision-relevant task structure.
    Hand-chosen segmentation; not validated against alternative segmentations.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bootstrapping Imitation Learning for Long-horizon Manipulation via Hierarchical Data Collection Space." pith.science (2026). https://pith.science/paper/NCSZZPGF

@misc{pith2026250517389,
  author       = {Pith},
  title        = {Pith review of: Bootstrapping Imitation Learning for Long-horizon Manipulation via Hierarchical Data Collection Space},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/NCSZZPGF}},
  note         = {Machine review of arXiv:2505.17389}
}
read the original abstract

Imitation learning (IL) with human demonstrations is a promising method for robotic manipulation tasks. While minimal demonstrations enable robotic action execution, achieving high success rates and generalization requires high cost, e.g., continuously adding data or incrementally conducting human-in-loop processes with complex hardware/software systems. In this paper, we rethink the state/action space of the data collection pipeline as well as the underlying factors responsible for the prediction of non-robust actions. To this end, we introduce a Hierarchical Data Collection Space (HD-Space) for robotic imitation learning, a simple data collection scheme, endowing the model to train with proactive and high-quality data. Specifically, We segment the fine manipulation task into multiple key atomic tasks from a high-level perspective and design atomic state/action spaces for human demonstrations, aiming to generate robust IL data. We conduct empirical evaluations across two simulated and five real-world long-horizon manipulation tasks and demonstrate that IL policy training with HD-Space-based data can achieve significantly enhanced policy performance. HD-Space allows the use of a small amount of demonstration data to train a more powerful policy, particularly for long-horizon manipulation tasks. We aim for HD-Space to offer insights into optimizing data quality and guiding data scaling. project page: https://hd-space-robotics.github.io.

Figures

Figures reproduced from arXiv: 2505.17389 by the authors.

Figure 1
Figure 1. Conceptual comparison of human demonstration data spaces for imitation learning. For simplification, we use a 2D manipulation state/action space for elaboration. (a) The naive data col￾lection method records the entire manipulation trajectory, recording a single distribution of trajecto￾ries in the second to fourth stages. Therefore, the model is prone to incorrectly predicted trajectories in the subtasks of stages … view at source ↗
Figure 2
Figure 2. Practice and comparison of naive and HD-Space data collection methods in the “put the teacup into the box” task. It first segments the task into multiple atomic subtasks and designs corresponding atomic state/action spaces for exploring a more robust data collection process. HD-Space: Atomic tasks Segmentation. Given a long-horizon manipulation task, it initiates by pinpointing the nodes of key actions, taking into … view at source ↗
Figure 3
Figure 3. Robust space considerations. Blue and purple dotted lines represent the trajectories covered by naive and HD-Space data collection processes. Orange lines represent prediction actions. Space#4: Place the end of the robotic arm in the space around the box, then move it to the box rope, close the gripper, and close the box. Practical Considerations. As shown in [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: Long-horizon tasks including one desktop task and two types of factory-related tasks. Effectiveness of the Stronger Model. The improvement results are summarized in Tab. 1. Our IL model significantly improves performance compared to the baseline ACT model on the Transf…
Figure 5
Figure 5. Figure 5 [PITH_FULL_IMAGE:figures/full_fig_p007_5.png]
Figure 6
Figure 6. Figure 6: Qualitative analysis of the failed cases and the successful ones solved by HD-Space. 4.3 Qualitative Analysis. We monitored multiple erroneous operation behaviors on the Grabbing spoon in mobile bowls task. As shown in [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: Real-robot setups including the Cobot mobile ALOHA robot and PiPER robot with teleoperation system. Appendix B: Detailed Real-robot Setup As depicted in [PITH_FULL_IMAGE:figures/full_fig_p013_7.png]

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Closing the Lab-to-Store Gap: A Data-Efficient Post-Training and Experience-Driven Learning VLA Framework for Retail Humanoids

    cs.RO 2026-07 conditional novelty 6.0 of 10

    On a real humanoid restocking chips, a data-efficient post-training pipeline raised success from 0% to 32%, one experience-driven refinement pass to 42%, and a second pass dropped it to 22%.

Reference graph

Works this paper leans on

61 extracted references · 25 canonical work pages · cited by 1 Pith paper

  1. [50]

    Huang, Y

    S. Huang, Y . Liao, S. Feng, S. Jiang, S. Liu, H. Li, M. Yao, and G. Ren. Adversarial data col- lection: Human-collaborative perturbations for efficient and robust robotic imitation learning. arXiv preprint arXiv:2503.11646, 2025

    arXiv 2025

  2. [1]

    D. A. Pomerleau. Alvinn: An autonomous land vehicle in a neural network. Advances in neural information processing systems, 1, 1988

    1988

  3. [2]

    Hussein, M

    A. Hussein, M. M. Gaber, E. Elyan, and C. Jayne. Imitation learning: A survey of learning methods. ACM Computing Surveys (CSUR), 50(2):1–35, 2017

    work page 2017

  4. [3]

    E. Jang, A. Irpan, M. Khansari, D. Kappler, F. Ebert, C. Lynch, S. Levine, and C. Finn. Bc-z: Zero-shot task generalization with robotic imitation learning. In Conference on Robot Learn- ing, pages 991–1002. PMLR, 2022

    work page 2022

  5. [4]

    Mandlekar, D

    A. Mandlekar, D. Xu, J. Wong, S. Nasiriany, C. Wang, R. Kulkarni, L. Fei-Fei, S. Savarese, Y . Zhu, and R. Mart´ın-Mart´ın. What matters in learning from offline human demonstrations for robot manipulation. Conference on Robot Learning (CoRL) , 2021

    work page 2021

  6. [5]

    T. Z. Zhao, V . Kumar, S. Levine, and C. Finn. Learning fine-grained bimanual manipulation with low-cost hardware. arXiv preprint arXiv:2304.13705, 2023

    arXiv 2023

  7. [6]

    C. Chi, Z. Xu, S. Feng, E. Cousineau, Y . Du, B. Burchfiel, R. Tedrake, and S. Song. Diffusion policy: Visuomotor policy learning via action diffusion. The International Journal of Robotics Research, page 02783649241273668, 2023

    2023

  8. [7]

    Y . Ze, G. Zhang, K. Zhang, C. Hu, M. Wang, and H. Xu. 3d diffusion policy: Generalizable visuomotor policy learning via simple 3d representations. arXiv preprint arXiv:2403.03954 , 2024

    arXiv 2024

Show all 61 references
  1. [8]

    Goyal, J

    A. Goyal, J. Xu, Y . Guo, V . Blukis, Y .-W. Chao, and D. Fox. Rvt: Robotic view transformer for 3d object manipulation. In Conference on Robot Learning, pages 694–710. PMLR, 2023

    2023

  2. [9]

    C. Chi, Z. Xu, C. Pan, E. Cousineau, B. Burchfiel, S. Feng, R. Tedrake, and S. Song. Universal manipulation interface: In-the-wild robot teaching without in-the-wild robots. arXiv preprint arXiv:2402.10329, 2024

    2024 arXiv

  3. [10]

    Z. Fu, T. Z. Zhao, and C. Finn. Mobile aloha: Learning bimanual mobile manipulation with low-cost whole-body teleoperation. arXiv preprint arXiv:2401.02117, 2024

    2024 arXiv

  4. [11]

    Z. Fu, Q. Zhao, Q. Wu, G. Wetzstein, and C. Finn. Humanplus: Humanoid shadowing and imitation from humans. arXiv preprint arXiv:2406.10454, 2024

    2024 arXiv

  5. [12]

    Cheng, J

    X. Cheng, J. Li, S. Yang, G. Yang, and X. Wang. Open-television: Teleoperation with immer- sive active visual feedback. arXiv preprint arXiv:2407.01512, 2024

    2024 arXiv

  6. [14]

    T. P. Lillicrap, J. J. Hunt, A. Pritzel, N. Heess, T. Erez, Y . Tassa, D. Silver, and D. Wierstra. Continuous control with deep reinforcement learning. arXiv preprint arXiv:1509.02971, 2015

    2015 arXiv

  7. [15]

    H. Tan, X. Xu, C. Ying, X. Mao, S. Liu, X. Zhang, H. Su, and J. Zhu. Manibox: Enhanc- ing spatial grasping generalization via scalable simulation data generation. arXiv preprint arXiv:2411.01850, 2024

    2024

  8. [16]

    Mandlekar, D

    A. Mandlekar, D. Xu, R. Mart ´ın-Mart´ın, Y . Zhu, L. Fei-Fei, and S. Savarese. Human-in-the- loop imitation learning using remote teleoperation. arXiv preprint arXiv:2012.06733, 2020

    2012 arXiv

  9. [17]

    J. Luo, C. Xu, J. Wu, and S. Levine. Precise and dexterous robotic manipulation via human- in-the-loop reinforcement learning. arXiv preprint arXiv:2410.21845, 2024. 9

    2024 arXiv

  10. [18]

    Hiranaka, M

    A. Hiranaka, M. Hwang, S. Lee, C. Wang, L. Fei-Fei, J. Wu, and R. Zhang. Primitive skill- based robot learning from human evaluative feedback. In 2023 IEEE/RSJ International Con- ference on Intelligent Robots and Systems (IROS) , pages 7817–7824. IEEE, 2023

    2023

  11. [19]

    X. Li, J. Li, Z. Zhang, R. Zhang, F. Jia, T. Wang, H. Fan, K.-K. Tseng, and R. Wang. Robogsim: A real2sim2real robotic gaussian splatting simulator. arXiv preprint arXiv:2411.11839, 2024

    2024 arXiv

  12. [20]

    Katara, Z

    P. Katara, Z. Xian, and K. Fragkiadaki. Gen2sim: Scaling up robot learning in simulation with generative models. In 2024 IEEE International Conference on Robotics and Automation (ICRA), pages 6672–6679. IEEE, 2024

    2024

  13. [21]

    K. Wang, W. R. Johnson, S. Lu, X. Huang, J. Booth, R. Kramer-Bottiglio, M. Aanjaneya, and K. Bekris. Real2sim2real transfer for control of cable-driven robots via a differentiable physics engine. In 2023 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS...

    2023

  14. [22]

    Y . Wu, L. Pan, W. Wu, G. Wang, Y . Miao, F. Xu, and H. Wang. Rl-gsbridge: 3d gaus- sian splatting based real2sim2real method for robotic manipulation learning. arXiv preprint arXiv:2409.20291, 2024

    2024 arXiv

  15. [23]

    X. Han, M. Liu, Y . Chen, J. Yu, X. Lyu, Y . Tian, B. Wang, W. Zhang, and J. Pang. Re3sim: Generating high-fidelity simulation data via 3d-photorealistic real-to-sim for robotic manipu- lation. arXiv preprint arXiv:2502.08645, 2025

    2025 arXiv

  16. [24]

    L. Wang, R. Guo, Q. Vuong, Y . Qin, H. Su, and H. Christensen. A real2sim2real method for robust object grasping with neural surface reconstruction. In 2023 IEEE 19th International Conference on Automation Science and Engineering (CASE) , pages 1–8. IEEE, 2023

    2023

  17. [25]

    Zhang, C

    X. Zhang, C. Wang, L. Sun, Z. Wu, X. Zhu, and M. Tomizuka. Efficient sim-to-real transfer of contact-rich manipulation skills with online admittance residual learning. In Conference on Robot Learning (CoRL), pages 1621–1639. PMLR, 2023

    2023

  18. [26]

    S. Schaal. Is imitation learning the route to humanoid robots? Trends in cognitive sciences, 3 (6):233–242, 1999

    1999

  19. [27]

    B. D. Argall, S. Chernova, M. Veloso, and B. Browning. A survey of robot learning from demonstration. Robotics and autonomous systems , 57(5):469–483, 2009

    2009

  20. [28]

    L. X. Shi, A. Sharma, T. Z. Zhao, and C. Finn. Waypoint-based imitation learning for robotic manipulation. Conference on Robot Learning (CoRL) , 2023

    2023

  21. [29]

    James and A

    S. James and A. J. Davison. Q-attention: Enabling efficient learning for vision-based robotic manipulation. IEEE Robotics and Automation Letters , 7(2):1612–1619, 2022

    2022

  22. [30]

    J. Pari, N. M. Shafiullah, S. P. Arunachalam, and L. Pinto. The surprising effectiveness of representation learning for visual imitation. arXiv preprint arXiv:2112.01511, 2021

    2021 arXiv

  23. [31]

    N. M. Shafiullah, Z. Cui, A. A. Altanzaya, and L. Pinto. Behavior transformers: Cloning k modes with one stone. Advances in neural information processing systems , 35:22955–22968, 2022

    2022

  24. [32]

    Y . Zhu, A. Joshi, P. Stone, and Y . Zhu. Viola: Imitation learning for vision-based manipulation with object proposal priors. In Conference on Robot Learning , pages 1199–1210. PMLR, 2023

    2023

  25. [33]

    A. Lee, I. Chuang, L.-Y . Chen, and I. Soltani. Interact: Inter-dependency aware action chunk- ing with hierarchical attention transformers for bimanual manipulation. Conference on Robot Learning (CoRL), 2024. 10

    2024

  26. [34]

    Z. Chen, X. Yuan, T. Mu, and H. Su. Responsive noise-relaying diffusion policy: Responsive and efficient visuomotor control. arXiv preprint arXiv:2502.12724, 2025

    2025 arXiv

  27. [35]

    Pearce, T

    T. Pearce, T. Rashid, A. Kanervisto, D. Bignell, M. Sun, R. Georgescu, S. V . Macua, S. Z. Tan, I. Momennejad, K. Hofmann, et al. Imitating human behaviour with diffusion models. arXiv preprint arXiv:2301.10677, 2023

    2023 arXiv

  28. [36]

    Bharadhwaj, J

    H. Bharadhwaj, J. Vakil, M. Sharma, A. Gupta, S. Tulsiani, and V . Kumar. Roboagent: Gener- alization and efficiency in robot manipulation via semantic augmentations and action chunking. In 2024 IEEE International Conference on Robotics and Automation (ICRA) , pages 4788–

    2024

  29. [37]

    Haldar, Z

    S. Haldar, Z. Peng, and L. Pinto. Baku: An efficient transformer for multi-task policy learning. arXiv preprint arXiv:2406.07539, 2024

    2024 arXiv

  30. [38]

    Brohan, N

    A. Brohan, N. Brown, J. Carbajal, Y . Chebotar, J. Dabis, C. Finn, K. Gopalakrishnan, K. Haus- man, A. Herzog, J. Hsu, et al. Rt-1: Robotics transformer for real-world control at scale. arXiv preprint arXiv:2212.06817, 2022

    2022 arXiv

  31. [39]

    M. J. Kim, K. Pertsch, S. Karamcheti, T. Xiao, A. Balakrishna, S. Nair, R. Rafailov, E. Foster, G. Lam, P. Sanketi, et al. Openvla: An open-source vision-language-action model. arXiv preprint arXiv:2406.09246, 2024

    2024 arXiv

  32. [40]

    Black, N

    K. Black, N. Brown, D. Driess, A. Esmail, M. Equi, C. Finn, N. Fusai, L. Groom, K. Hausman, B. Ichter, et al. pi0: A vision-language-action flow model for general robot control. arXiv preprint arXiv:2410.24164, 2024

    2024 arXiv

  33. [41]

    O’Neill, A

    A. O’Neill, A. Rehman, A. Maddukuri, A. Gupta, A. Padalkar, A. Lee, A. Pooley, A. Gupta, A. Mandlekar, A. Jain, et al. Open x-embodiment: Robotic learning datasets and rt-x models: Open x-embodiment collaboration 0. In 2024 IEEE International Conference on Robotics and Automat...

    2024

  34. [42]

    Khazatsky, K

    A. Khazatsky, K. Pertsch, S. Nair, A. Balakrishna, S. Dasari, S. Karamcheti, S. Nasiriany, M. K. Srirama, L. Y . Chen, K. Ellis, et al. Droid: A large-scale in-the-wild robot manipulation dataset. arXiv preprint arXiv:2403.12945, 2024

    2024 arXiv

  35. [43]

    K. Wu, C. Hou, J. Liu, Z. Che, X. Ju, Z. Yang, M. Li, Y . Zhao, Z. Xu, G. Yang, et al. Robomind: Benchmark on multi-embodiment intelligence normative data for robot manipulation. arXiv preprint arXiv:2412.13877, 2024

    2024 arXiv

  36. [44]

    Shridhar, L

    M. Shridhar, L. Manuelli, and D. Fox. Perceiver-actor: A multi-task transformer for robotic manipulation. In Conference on Robot Learning, pages 785–799. PMLR, 2023

    2023

  37. [45]

    Hsiao and T

    K. Hsiao and T. Lozano-Perez. Imitation learning of whole-body grasps. In 2006 IEEE/RSJ international conference on intelligent robots and systems , pages 5657–5662. IEEE, 2006

    2006

  38. [46]

    Akgun, M

    B. Akgun, M. Cakmak, K. Jiang, and A. L. Thomaz. Keyframe-based learning from demon- stration: Method and evaluation. International Journal of Social Robotics , 4:343–355, 2012

    2012

  39. [47]

    A. Zeng, P. Florence, J. Tompson, S. Welker, J. Chien, M. Attarian, T. Armstrong, I. Krasin, D. Duong, V . Sindhwani, et al. Transporter networks: Rearranging the visual world for robotic manipulation. In Conference on Robot Learning, pages 726–747. PMLR, 2021

    2021

  40. [48]

    Shridhar, L

    M. Shridhar, L. Manuelli, and D. Fox. Cliport: What and where pathways for robotic manipu- lation. In Conference on robot learning, pages 894–906. PMLR, 2022

    2022

  41. [49]

    Morrison, P

    D. Morrison, P. Corke, and J. Leitner. Closing the loop for robotic grasping: A real-time, generative grasp synthesis approach. arXiv preprint arXiv:1804.05172, 2018. 11

    2018 arXiv

  42. [51]

    Belkhale, Y

    S. Belkhale, Y . Cui, and D. Sadigh. Data quality in imitation learning. Advances in neural information processing systems, 36:80375–80395, 2023

    2023

  43. [52]

    Samadi, K

    A. Samadi, K. Koufos, K. Debattista, and M. Dianati. Good data is all imitation learning needs. arXiv preprint arXiv:2409.17605, 2024

    2024 arXiv

  44. [53]

    F. Lin, Y . Hu, P. Sheng, C. Wen, J. You, and Y . Gao. Data scaling laws in imitation learning for robotic manipulation. arXiv preprint arXiv:2410.18647, 2024

    2024 arXiv

  45. [54]

    Carion, F

    N. Carion, F. Massa, G. Synnaeve, N. Usunier, A. Kirillov, and S. Zagoruyko. End-to-end object detection with transformers. In European conference on computer vision , pages 213–

  46. [55]

    K. Sohn, H. Lee, and X. Yan. Learning structured output representation using deep conditional generative models. Advances in neural information processing systems , 28, 2015

    2015

  47. [56]

    D. P. Kingma, M. Welling, et al. Auto-encoding variational bayes, 2013

    2013

  48. [57]

    Oquab, T

    M. Oquab, T. Darcet, T. Moutakanni, H. V o, M. Szafraniec, V . Khalidov, P. Fernandez, D. Haz- iza, F. Massa, A. El-Nouby, et al. Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193, 2023

    2023 arXiv

  49. [58]

    Z. Li, L. Ren, J. Yang, Y . Zhao, X. Wu, Z. Xu, X. Bai, and H. Zhao. Virt: Vision instructed transformer for robotic manipulation. arXiv preprint arXiv:2410.07169, 2024

    2024 arXiv

  50. [59]

    J. Borucka. City walk: a didactic innovative experiment in architectural education. World Transactions on Engineering and Technology Education, 17:158–163, 2019

    2019

  51. [60]

    K. He, X. Chen, S. Xie, Y . Li, P. Doll ´ar, and R. Girshick. Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 16000–16009, 2022

    2022

  52. [61]

    URL https://www.trossenrobotics.com/viperx-300-robot-arm- 6dof.aspx

    Viperx 300 robot arm 6dof. URL https://www.trossenrobotics.com/viperx-300-robot-arm- 6dof.aspx

  53. [62]

    J. Yang, S. Liu, Z. Li, X. Li, , and J. Sun. Real-time object detection for streaming perception. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5385–5395, 2022. 12 Appendix A: Hyperparameters As shown in Tab. 4, we use the same hyp...

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.