Robotics · arXiv:2610.12007 v1 · submitted 8 Oct 2026 · 71 references
REACT: Rolling Denoising and Dual Decoupling for Reactive Robot Control with VLA Models
The paper says
“Flow-based vision-language-action (VLA) models generate action chunks for temporally coherent robot motion, but chunked control creates a fundamental closed-loop trade-off: long chunks provide smooth execution, whereas frequent replanning improves reactivity at the cost of action discontinuities.”
more of the abstract
We introduce REACT, a rolling-denoising framework that makes flow-based VLAs more reactive while preserving long-horizon context. Instead of regenerating entire action chunks from scratch, REACT maintains a persistent action buffer with staggered flow timesteps. At each control step, the full horizon is denoised using the latest observation, the cleanest action block is executed, partially refined future blocks are shifted forward, and fresh noise is appended to the tail. As a result, each executed action block is refined across multiple recent observations before deployment. To support real-time control, we further introduce dual decoupling, which separates sensing, VLM encoding, DiT denoising, and action execution, enabling high-frequency observation updates and action streaming under practical compute constraints. Across the RoboTwin 2.0 simulation benchmark and real-world tasks spanning bimanual manipulation and dynamic control on multiple robot platforms, REACT improves task success and reduces reaction latency while producing smoother trajectories than frequent-replanning and asynchronous baselines.
from the abstract
Pith says
Machine review · deepseek-v4-flash · 10 Oct 2026 · what was read was not recorded · not checked by a person
It can be wrong. Challenge this review or reply to it.
Holds up, with conditions. Strong experimental paper with a potentially load-bearing training-inference gap that needs to be addressed. This line is wrong
1 major objection · 5 minor
How often objections from reviews like this one are right: not yet measured
Not checked: figures, code, what the 71 references say.
The paper says, at §3.3 (Eq. 5) and Appendix C.5
The review did not quote this passage.
Pith says: Objection 1 of 1 · major
The paper asserts that each executed action block is refined across multiple recent observations, but the training procedure (Eq. 5) supervises all K denoising stages of each block using a single observation (the observation at the start of the demonstration chunk).
more of objection 1
During inference, each denoising stage of the same block is conditioned on a different, updated observation. Appendix C.5 argues per-step time-position alignment and a rolling-buffer invariant, but it does not address the observation-conditional mismatch: the velocity field v_θ(x, τ, o) depends on observation o, and the distribution over o for a given block at a given τ differs between training and inference. This mismatch undermines the claim that the model has learned to maintain action coherence across changing observations. The authors should either provide empirical evidence that this mismatch does not degrade performance (e.g., by comparing performance when the inference observation is held fixed vs. allowed to vary, or by measuring the effect of the observation change on the denoising predictions) or revise the claim to acknowledge that the multi-observation refinement is a heuristic that works in practice despite the training gap.
5 minor objections
The paper says, at §3.1, Eq. (1)
Pith says: minor objection 1 of 5
The formula for τ_j appears to have a formatting error: it reads "τj = ⌊j/S⌋+ 1K" but should likely be "τ_j = (⌊j/S⌋ + 1) / K". Please correct the typesetting.
The paper says, at §4.2, Table 2
Pith says: minor objection 2 of 5
The overall average success for REACT w/o DD on real-world tasks is listed as 65.7% in the text but the table shows ARX X5 average 68.3 and Franka average 63.0, which gives an average of 65.65%—consistent. However, the rounding to 65.7 vs. the table's 65.7 is fine; please double-check for consistency across all rows.
show 3 more
The paper says, at §6 (Limitations)
Pith says: minor objection 3 of 5
The authors acknowledge that they did not train independent policy seeds, so statistical comparisons reflect evaluation-rollout noise only. This is a transparent limitation but means that the reported success rates lack confidence intervals that account for training variability. Adding a note about the expected effect size would strengthen the paper.
The paper says, at §C.5
Pith says: minor objection 4 of 5
The notation in the block-trajectory alignment paragraph is slightly confusing: "the shifted block that was previously at flow time (k+2)/K" should be (k+2)/K? Actually after the shift, block k was previously block k+1 at time (k+2)/K? Check indexing: if blocks are numbered 0..K-1, after shift the block at new index k was at old index k+1 which had time (k+1+1)/K = (k+2)/K; that is correct. But the sentence could be clarified by explicitly stating the indexing.
The paper says, at Abstract
Pith says: minor objection 5 of 5
The phrase "output throughput(action produced)" is missing a space before the parentheses. Minor formatting.
What would settle it
Train a variant, REACT-multiobs, where during training each block's K denoising stages are conditioned on K different observations from the demonstration, temporally aligned with the inference schedule. For a demo chunk at time t, condition block k (which will be executed after k blocks) on observation o_{t + k*S} rather than always on o_t. Compare REACT-multiobs to standard REACT on three RoboTwin tasks with dynamic perturbations (press_stapler, turn_switch, beat_block_hammer) using 5 training seeds each. If REACT-multiobs yields significantly higher success (p<0.05 via paired t-test) or lower jerk, the observation mismatch is harmful; if no significant difference, the current training is sufficient and the alignment assumption holds.
from the skeptic's reading
How the review went
4 readings by deepseek-v4-flash · 10 Oct 2026
- Reader holds up, with conditions
- Skeptic partly agrees
- Editor wrote the verdict above
- Referee major revision: 1 major objection, 5 minor
Not run on this paper: the simulated authors' defence.
Pith did not check this review's objections against the paper.
The review in full
The review in brief, machine-written
The paper does something genuinely useful: it identifies the coherence–freshness trade-off in chunked VLA control and builds a rolling-denoising system that lets a long-horizon buffer react to new observations without sacrificing smoothness. The experiments are thorough—seven simulation tasks, six real-world tasks, multiple baselines, ablations, statistical tests. The gains over frequent-replanning and async baselines are large and consistent. On dynamic tasks like Pour Rice and Reaction Game, REACT roughly matches human teleoperation latency while preserving trajectory quality. That is a real step forward for deploying VLAs in settings that cannot tolerate open-loop chattering.
Credit where it is due: the paper ships real robot demos, the ablation over (K, S) is honest about the constraints of the pretrained backbone, and the limitations section is unusually transparent—no independent training seeds, single backbone, sensitivity to the rolling schedule.
Now the soft spot that matters most, and it is a load-bearing one. The stress-test note hits a genuine gap. Staircase training (Eq. 5) supervises all K denoising stages of a block using a single observation from the start of the demonstration chunk. During inference, each denoising stage of the same block conditions on a different, updated observation. Appendix C.5 argues time-position alignment but says nothing about the observation-conditional mismatch. The claim that 'each executed action block is refined across multiple recent observations' is the central justification for why REACT improves both success and smoothness. If the training never exposes the model to blocks that are denoised under changing observations, then the multi-observation refinement argument is unsupported. The empirical results are impressive, so the model apparently generalizes across this gap—but the paper does not explain why, and the argument as written is incomplete. This needs a plausible theoretical argument or a training modification (e.g., sampling observation crops from the demonstration chunk's time window) before the paper's main claim holds water.
The other weaknesses—single backbone, no training seeds—are acknowledged and are standard reviewer concerns, not dealbreakers.
Who gets value from this paper: anyone working on real-time VLA deployment, especially in dynamic manipulation or contact-rich tasks. The method is practical and the evaluation is rigorous.
Recommendation: send to peer review. The paper is clearly important and the experimental evidence is strong, but the training-inference inconsistency in the rolling denoising argument must be resolved before acceptance. A revision that either provides a formal justification for the observed generalization or modifies training to include observation changes would lift this from promising to solid.
The strongest objection
The central claim that REACT improves the success-smoothness-reactivity trade-off relies on rolling denoising where each action block is refined across K consecutive observations (Section 3.1). However, staircase training (Equation 5) supervises all K denoising stages of each block using a single observation from the demonstration chunk (the observation at the start of the chunk). During inference, each denoising stage of the same block is conditioned on a different, updated observation. Appendix C.5 argues time-position alignment but does not address this observation-conditional mismatch. The velocity field v_θ(x, τ, o) depends on observation o; the training distribution over o for a given block at a given τ differs from the inference distribution because training pairs a fixed o with multiple τ for the same block, while inference pairs a different o with each τ. This creates a gap in the claimed 'many-observations-to-one-block refinement' logic: the model was never trained to handle a block undergoing denoising steps with varying observations. While the empirical results are promising, the theoretical foundation for multi-observation refinement is incomplete, making the claim that 'each executed action block is refined across multiple recent observations' unsupported by the training procedure. The load-bearing assumption that the model learns to maintain action coherence across changing observations is not justified.
The reader's weakest_assumption identified train-inference alignment and sensitivity to K,S as concerns. My attack focuses on a specific sub-aspect of that alignment: the observation-conditional mismatch during denoising. The reader did not explicitly separate the time-position alignment from the observation conditioning, but implicitly covered it under 'alignment is argued in Appendix C.5 but not formally proven.' Appendix C.5 proves time-position alignment but omits the observation mismatch, which is a distinct issue. Therefore I partially agree. The concrete test directly probes whether the mismatch matters in practice. If the test shows no harm, the central claim holds with the current training; if it shows harm, the paper must revise the training to match inference or adjust the claim. Since the paper already conditions acceptance on addressing limitations, and this test is feasible, the verdict should remain CONDITIONAL (UNCHANGED from the reader's verdict).
The referee's report
If the empirical results hold, REACT offers a practical improvement to the way chunked VLA policies are deployed in dynamic environments. It identifies the coherence–freshness trade-off and provides a system-level solution that requires no changes to model weights. The dual-decoupling scheduler is a straightforward engineering contribution that could be adopted by other methods. The staircase training procedure also appears to accelerate convergence. However, the theoretical justification for the central claim—that each executed action block is refined across multiple recent observations—is incomplete, which tempers the significance of the contribution.
Pith's reading of the paper (a machine's summary, not the authors')
Flow-based vision-language-action models generate long action chunks for smooth motion, but this creates a closed-loop problem: the robot cannot react to new observations until the entire chunk finishes. REACT shows that the trade-off between coherence and reactivity can be overcome by maintaining a single rolling action buffer. Instead of regenerating chunks from scratch, the buffer is denoised one Euler step per control cycle using the latest observation; the cleanest block is executed, partially refined blocks shift forward, and fresh noise is appended. A dual-decoupling scheduler separates sensing, VLM encoding, DiT denoising, and execution, allowing high-frequency observation updates without waiting for any one stage. On RoboTwin 2.0 and real bimanual manipulation tasks, REACT outperforms both frequent-replanning and asynchronous baselines in task success and trajectory smoothness.
What carries the argument. The rolling action buffer with staircase flow timestamps is the central mechanism. Each buffer block k is assigned a deterministic flow time τ_k = (k+1)/K, and one Euler update per control step advances every block by one denoising stage. After the update, the front block (now clean) is executed; the remaining blocks are shifted left, and a fresh Gaussian-noise block is appended. This ensures that every executed block has been denoised through K stages, each conditioned on a different observation. Dual decoupling is the complementary scheduler: a pool of M asynchronous VLM workers encode selected frames into a latest-ready cache, so the DiT denoising step can always consume the freshest embedding without waiting for VLM inference.
Core claim. The paper claims that for flow-based VLA policies, the coherence–freshness conflict is not fundamental. By adopting a rolling denoising schedule with a persistent action buffer and a staircase flow-time assignment, each action block is refined across multiple consecutive observations before execution. Combined with a dual-decoupled inference pipeline that asynchronously feeds VLM embeddings to the diffusion transformer, the system achieves camera-rate action streaming while preserving the long-horizon action mode that makes chunked control smooth. The empirical evidence shows that this design raises the Pareto frontier of success, smoothness, and reactivity on both simulated and real-world bimanual manipulation tasks, including dynamic tasks such as targeted pouring and a visual reaction game.
The premise it rests on. The matched staircase training (Equation 5) that supervises K deterministic denoising stages per example is assumed to align with the rolling inference schedule so that the model learns to maintain action coherence across block boundaries, but this alignment is not formally proven and the performance is sensitive to the choice of K and S.
If the paper is right (Pith's inferences, not the authors')
VLA policies can use long action horizons (e.g., H=50) in reactive settings without the success degradation and jerk that come from frequent replanning or asynchronous stitched chunks.
Training efficiency improves because the staircase objective supervises all denoising stages in one forward pass, reaching target closed-loop success in half the optimization steps.
The rolling denoising pattern can be applied to any dual-system flow-based VLA that separates a perception-language backbone from an action expert, provided the training schedule is matched.
Real-time deployment becomes feasible on a single consumer GPU (RTX 4090) without sacrificing reactivity, since dual decoupling hides VLM latency behind DiT updates.
The core idea of refining a persistent action buffer across observations is not limited to flow matching; it could be adapted to diffusion-based policies that use iterative denoising, potentially benefiting a wider class of visuomotor architectures.
Dual decoupling shifts the bottleneck from model inference to camera-frame throughput, meaning that as VLA models become faster (via distillation or quantization), the rolling schedule will become the dominant determinant of reactivity.
Because each executed block is shaped by K observations, the method implicitly smooths the action distribution over time, which may reduce the need for explicit temporal ensembling or action smoothing in downstream controllers.
The paper's ablation (Figure 11) suggests that for a fixed compute budget, increasing K (number of rolling refinements) consistently improves success, implying that multi-observation refinement is more valuable than very short block sizes for most tasks.
Check this review
Pith stored no record of this review's passes and took no fingerprint of it. Reviews made by code from 13 Nov 2026 on carry both.
How often reviews like this one are right
Reviews like this one: written by deepseek-v4-flash, where what was read was not recorded.
How often objections from reviews like this one are right: not yet measured
Reviews by people
This review was written by a machine. No person has reviewed this paper on Pith yet. A scientist listed on Pith can review it for a fee, and their review is published here whatever it concludes.
The paper
Abstract
Flow-based vision-language-action (VLA) models generate action chunks for temporally coherent robot motion, but chunked control creates a fundamental closed-loop trade-off: long chunks provide smooth execution, whereas frequent replanning improves reactivity at the cost of action discontinuities. We introduce REACT, a rolling-denoising framework that makes flow-based VLAs more reactive while preserving long-horizon context. Instead of regenerating entire action chunks from scratch, REACT maintains a persistent action buffer with staggered flow timesteps. At each control step, the full horizon is denoised using the latest observation, the cleanest action block is executed, partially refined future blocks are shifted forward, and fresh noise is appended to the tail. As a result, each executed action block is refined across multiple recent observations before deployment. To support real-time control, we further introduce dual decoupling, which separates sensing, VLM encoding, DiT denoising, and action execution, enabling high-frequency observation updates and action streaming under practical compute constraints. Across the RoboTwin 2.0 simulation benchmark and real-world tasks spanning bimanual manipulation and dynamic control on multiple robot platforms, REACT improves task success and reduces reaction latency while producing smoother trajectories than frequent-replanning and asynchronous baselines.
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. No sentence in this record was written by a model.
abstract compiled · 2026-10-09
The authors' main claim, as the classifier picked it out 0.99
“Across the RoboTwin 2.0 simulation benchmark and real-world tasks spanning bimanual manipulation and dynamic control on multiple robot platforms, REACT improves task success and reduces reaction latency while producing smoother trajectories than frequent-replanning and asynchronous baselines.”
How the main claim is stated: asserted 1.00
What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.18 · human participants 0.04 · controlled comparison 0.95 · randomized assignment 0.02 · states quantitative result 0.05 · compares with prior results 0.96 · mentions code or data availability 0.03
The abstract, sentence by sentence
This table labels the abstract only (arXiv listing abstract v1). The number is the probability the classifier gave the label.
| Sentence | Role | The authors' own work | States the main conclusion | Strength |
|---|---|---|---|---|
| Flow-based vision-language-action (VLA) models generate action chunks for temporally coherent robot motion, but chunked control creates a fundamental closed-loop trade-off: long chunks provide smooth execution, whereas frequent replanning improves reactivity at the cost of action discontinuities. | background 0.98 | 0.08 | 0.05 | asserted 1.00 |
| We introduce REACT, a rolling-denoising framework that makes flow-based VLAs more reactive while preserving long-horizon context. | contribution 1.00 | 0.86 | 0.21 | asserted 1.00 |
| Instead of regenerating entire action chunks from scratch, REACT maintains a persistent action buffer with staggered flow timesteps. | method 0.97 | 0.78 | 0.04 | asserted 1.00 |
| At each control step, the full horizon is denoised using the latest observation, the cleanest action block is executed, partially refined future blocks are shifted forward, and fresh noise is appended to the tail. | method 1.00 | 0.83 | 0.06 | asserted 0.99 |
| As a result, each executed action block is refined across multiple recent observations before deployment. | method 0.45 | 0.79 | 0.09 | asserted 1.00 |
| To support real-time control, we further introduce dual decoupling, which separates sensing, VLM encoding, DiT denoising, and action execution, enabling high-frequency observation updates and action streaming under practical compute constraints. | method 0.72 | 0.91 | 0.07 | asserted 1.00 |
| Across the RoboTwin 2.0 simulation benchmark and real-world tasks spanning bimanual manipulation and dynamic control on multiple robot platforms, REACT improves task success and reduces reaction latency while producing smoother trajectories than frequent-replanning and asynchronous baselines. | result 1.00 | 0.96 | 0.85 | asserted 1.00 |
What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.68
What was examined, and what was not
Sources read
- arXiv listing abstract v1
Not examined
- full text
- figures and tables
- references and citing sentences
- relations to other papers
Levels run
- P abstract
Not run
- S sections
- T body sentences
- C citations
- R cross-paper relations
How the labels were made
Compiled by classifier-1.13.0 via TypeSafe API, question bank abstract-p1-20260923, 2026-10-09: 1 request, 5,765 input tokens.
Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. This record as JSON
Figures (14)
Figures are the authors', reproduced from the paper.
References (71)
- [1] 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. 9 on PithPith's reviewarXiv
- [2] B. Zitkovich, T. Yu, S. Xu, P. Xu, T. Xiao, F. Xia, J. Wu, P. Wohlhart, S. Welker, A. Wahid, et al. Rt-2: Vision-language-action models transfer web knowledge to robotic control. InConference on Robot Learning, pages 2165–2183. PMLR, 2023 on Pith
- [3] 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 on PithPith's reviewarXiv
- [4] K. Black, N. Brown, D. Driess, A. Esmail, M. Equi, C. Finn, N. Fusai, L. Groom, K. Hausman, B. Ichter, et al. π0: A vision-language-action flow model for general robot control.arXiv preprint arXiv:2410.24164, 2024 on PithPith's reviewarXiv
- [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 on PithPith's reviewarXiv
- [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, 44(10-11):1684–1704, 2025 on Pith
- [7] P. Intelligence, K. Black, N. Brown, J. Darpinian, K. Dhabalia, D. Driess, A. Esmail, M. Equi, C. Finn, N. Fusai, et al. π0.5: a vision-language-action model with open-world generalization. arXiv preprint arXiv:2504.16054, 2025 on PithPith's reviewarXiv
- [8] K. Black, M. Galliker, and S. Levine. Real-time execution of action chunking flow policies. Advances in Neural Information Processing Systems, 38:33383–33407, 2026 on Pith
- [9] K. Black, A. Z. Ren, M. Equi, and S. Levine. Training-time action conditioning for efficient real-time chunking.arXiv preprint arXiv:2512.05964, 2025 on PitharXiv
- [10] S. Xu, Y . Wang, C. Xia, D. Zhu, T. Huang, and C. Xu. Vla-cache: Efficient vision-language- action manipulation via adaptive token caching.Advances in Neural Information Processing Systems, 38:164448–164473, 2026 on Pith
- [11] X. Pei, Y . Chen, S. Xu, Y . Wang, Y . Shi, and C. Xu. Action-aware dynamic pruning for efficient vision-language-action manipulation.arXiv preprint arXiv:2509.22093, 2025 on PitharXiv
- [12] Y . Yang, Y . Wang, Z. Wen, L. Zhongwei, C. Zou, Z. Zhang, C. Wen, and L. Zhang. Efficientvla: Training-free acceleration and compression for vision-language-action models.Advances in Neural Information Processing Systems, 38:40891–40914, 2026 on Pith
- [13] Y . Ma, Y . Zhou, Y . Yang, T. Wang, and H. Fan. Running vlas at real-time speed.arXiv preprint arXiv:2510.26742, 2025 on PitharXiv
- [14] Y . Lu, Z. Liu, X. Fan, Z. Yang, J. Hou, J. Li, K. Ding, and H. Zhao. Faster: Rethinking real-time flow vlas.arXiv preprint arXiv:2603.19199, 2026 on PithPith's reviewarXiv
- [15] L. Lai, A. Z. Huang, and S. J. Gershman. Action chunking as conditional policy compression. Cognition, 264:106201, 2025 on Pith
- [16] A. George and A. B. Farimani. One act play: Single demonstration behavior cloning with action chunking transformers.arXiv preprint arXiv:2309.10175, 2023 on PithPith's reviewarXiv
- [17] 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 on PithPith's reviewarXiv
- [18] Y . Lipman, R. T. Chen, H. Ben-Hamu, M. Nickel, and M. Le. Flow matching for generative modeling.arXiv preprint arXiv:2210.02747, 2022. 10 on PithPith's reviewarXiv
- [19] C. He, X. Liu, G. M. S. Camps, J. Bruno, G. A. Sartoretti, and M. Schwager. Demystifying robot diffusion policies: Action memorization and a simple lookup table alternative. InThe Fourteenth International Conference on Learning Representations, 2026 on Pith
- [20] C. He, G. S. Camps, X. Liu, M. Schwager, and G. Sartoretti. Latent theory of mind: A decen- tralized diffusion architecture for cooperative manipulation.arXiv preprint arXiv:2505.09144, 2025 on PithPith's reviewarXiv
show 51 more
- [21] 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. In2024 IEEE International Conference on Robotics and Automati... on Pith
- [22] A.-C. Cheng, Y . Ji, Z. Yang, Z. Gongye, X. Zou, J. Kautz, E. Bıyık, H. Yin, S. Liu, and X. Wang. Navila: Legged robot vision-language-action model for navigation.arXiv preprint arXiv:2412.04453, 2024 on PithPith's reviewarXiv
- [23] H. Zhen, X. Qiu, P. Chen, J. Yang, X. Yan, Y . Du, Y . Hong, and C. Gan. 3d-vla: A 3d vision-language-action generative world model.arXiv preprint arXiv:2403.09631, 2024 on PithPith's reviewarXiv
- [24] R. Zheng, Y . Liang, S. Huang, J. Gao, H. Daumé III, A. Kolobov, F. Huang, and J. Yang. Tracevla: Visual trace prompting enhances spatial-temporal awareness for generalist robotic policies. InInternational Conference on Learning Representations, volume 2025, pages 54277– 54296, 2025 on Pith
- [25] S. Liu, L. Wu, B. Li, H. Tan, H. Chen, Z. Wang, K. Xu, H. Su, and J. Zhu. Rdt-1b: a diffusion foundation model for bimanual manipulation. InInternational Conference on Learning Representations, volume 2025, pages 29982–30009, 2025 on Pith
- [26] C. He, G. Sun, Y . Bai, J. Lu, J. Zhao, and G. Sartoretti. Falcon: Actively decoupled visuomotor policies for loco-manipulation with foundation-model-based coordination.arXiv preprint arXiv:2512.04381, 2025 on PitharXiv
- [27] 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 on PithPith's reviewarXiv
- [28] H. R. Walke, K. Black, T. Z. Zhao, Q. Vuong, C. Zheng, P. Hansen-Estruch, A. W. He, V . Myers, M. J. Kim, M. Du, et al. Bridgedata v2: A dataset for robot learning at scale. InConference on Robot Learning, pages 1723–1736. PMLR, 2023 on Pith
- [29] H.-S. Fang, H. Fang, Z. Tang, J. Liu, C. Wang, J. Wang, H. Zhu, and C. Lu. Rh20t: A compre- hensive robotic dataset for learning diverse skills in one-shot.arXiv preprint arXiv:2307.00595, 2023 on PithPith's reviewarXiv
- [30] A. Mandlekar, Y . Zhu, A. Garg, J. Booher, M. Spero, A. Tung, J. Gao, J. Emmons, A. Gupta, E. Orbay, et al. Roboturk: A crowdsourcing platform for robotic skill learning through imitation. InConference on Robot Learning, pages 879–893. PMLR, 2018 on Pith
- [31] G. R. Team, S. Abeyruwan, J. Ainslie, J.-B. Alayrac, M. G. Arenas, T. Armstrong, A. Balakr- ishna, R. Baruch, M. Bauza, M. Blokzijl, et al. Gemini robotics: Bringing ai into the physical world.arXiv preprint arXiv:2503.20020, 2025 on PithPith's reviewarXiv
- [32] M. Reuss, H. Zhou, M. Rühle, Ö. E. Ya˘gmurlu, F. Otto, and R. Lioutikov. Flower: Democratizing generalist robot policies with efficient vision-language-action flow policies.arXiv preprint arXiv:2509.04996, 2025 on PithPith's reviewarXiv
- [33] Z. Yu, B. Wang, P. Zeng, H. Zhang, J. Zhang, Z. Wang, L. Gao, J. Song, N. Sebe, and H. T. Shen. A survey on efficient vision-language-action models.arXiv preprint arXiv:2510.24795, 2025. 11 on PithPith's reviewarXiv
- [34] W. Guan, Q. Hu, A. Li, and J. Cheng. Efficient vision-language-action models for embodied manipulation: A systematic survey.arXiv preprint arXiv:2510.17111, 2025 on PithPith's reviewarXiv
- [35] P. Budzianowski, W. Maa, M. Freed, J. Mo, W. Hsiao, A. Xie, T. Młoduchowski, V . Tipnis, and B. Bolte. Edgevla: Efficient vision-language-action models.arXiv preprint arXiv:2507.14049, 2025 on PithPith's reviewarXiv
- [36] J. Wen, Y . Zhu, J. Li, M. Zhu, Z. Tang, K. Wu, Z. Xu, N. Liu, R. Cheng, C. Shen, et al. Tinyvla: Towards fast, data-efficient vision-language-action models for robotic manipulation.IEEE Robotics and Automation Letters, 2025 on Pith
- [37] M. Shukor, D. Aubakirova, F. Capuano, P. Kooijmans, S. Palma, A. Zouitine, M. Aractingi, C. Pascal, M. Russi, A. Marafioti, et al. Smolvla: A vision-language-action model for affordable and efficient robotics.arXiv preprint arXiv:2506.01844, 2025 on PithPith's reviewarXiv
- [38] J. Chen, J. Wang, L. Chen, C. Cai, and J. Lu. Nanovla: Routing decoupled vision-language understanding for nano-sized generalist robotic policies.arXiv preprint arXiv:2510.25122, 2025 on PitharXiv
- [39] J. Williams, K. D. Gupta, R. George, and M. Sarkar. Lite vla: Efficient vision-language-action control on cpu-bound edge robots.arXiv preprint arXiv:2511.05642, 2025 on PitharXiv
- [40] Z. Liu, Y . Chen, H. Cai, T. Lin, S. Yang, Z. Liu, and B. Zhao. Vla-pruner: Temporal-aware dual-level visual token pruning for efficient vision-language-action inference.arXiv preprint arXiv:2511.16449, 2025 on PithPith's reviewarXiv
- [41] R. Zhang, M. Dong, Y . Zhang, L. Heng, X. Chi, G. Dai, L. Du, D. Wang, Y . Du, and S. Zhang. Mole-vla: Dynamic layer-skipping vision language action model via mixture-of-layers for efficient robot manipulation. InProceedings of the AAAI Conference on Artificial Intelligence, v... on Pith
- [42] Y . Chen and X. Li. Rlrc: Reinforcement learning-based recovery for compressed vision- language-action models.arXiv preprint arXiv:2506.17639, 2025 on PithPith's reviewarXiv
- [43] S. Wang, R. Yu, Z. Yuan, C. Yu, F. Gao, Y . Wang, and D. F. Wong. Spec-vla: speculative decoding for vision-language-action models with relaxed acceptance. InProceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 26916–26928, 2025 on Pith
- [44] J. Jia, G. Li, X. Chen, T. An, Y . Hu, J. Li, X. Guo, and J. Yang. Action-to-action flow matching. arXiv preprint arXiv:2602.07322, 2026 on PithPith's reviewarXiv
- [45] K. Pertsch, K. Stachowicz, B. Ichter, D. Driess, S. Nair, Q. Vuong, O. Mees, C. Finn, and S. Levine. Fast: Efficient action tokenization for vision-language-action models.arXiv preprint arXiv:2501.09747, 2025 on PithPith's reviewarXiv
- [46] Y . Dai, H. Gu, T. Wang, Q. Cheng, Y . Zheng, Z. Qiu, L. Gong, W. Lou, and X. Zhou. Ac- tionflow: A pipelined action acceleration for vision language models on edge.arXiv preprint arXiv:2512.20276, 2025 on PithPith's reviewarXiv
- [47] H. Wang, C. Xiong, R. Wang, and X. Chen. Bitvla: 1-bit vision-language-action models for robotics manipulation.arXiv preprint arXiv:2506.07530, 2025 on PitharXiv
- [48] Y . Zhao, L. Zhao, B. Cheng, G. Yao, X. Wen, and H. Gao. Vla-rail: A real-time asynchronous inference linker for vla models and robots.arXiv preprint arXiv:2512.24673, 2025 on PitharXiv
- [49] Y . Jiang, S. Cheng, Y . Ding, F. Gao, and B. Qi. Asyncvla: Asynchronous flow matching for vision-language-action models.arXiv preprint arXiv:2511.14148, 2025. 12 on PithPith's reviewarXiv
- [50] H. Xie, B. Wen, J. Zheng, Z. Chen, F. Hong, H. Diao, and Z. Liu. Dynamicvla: A vision- language-action model for dynamic object manipulation.arXiv preprint arXiv:2601.22153, 2026 on PithPith's reviewarXiv
- [51] R. Cai, J. Guo, X. He, P. Jin, J. Li, B. Lin, F. Liu, W. Liu, F. Ma, K. Ma, et al. Xiaomi-robotics- 0: An open-sourced vision-language-action model with real-time execution.arXiv preprint arXiv:2602.12684, 2026 on PitharXiv
- [52] Y . Liu, H. Yu, J. Zhao, B. Li, D. Zhang, M. Li, W. Wu, Y . Hu, J. Xie, J. Guo, et al. Learning native continuation for action chunking flow policies.arXiv preprint arXiv:2602.12978, 2026 on PithPith's reviewarXiv
- [53] A. Liao, D.-K. Kim, M. O. Smith, A.-a. Agha-mohammadi, and S. Omidshafiei. Delay-aware diffusion policy: Bridging the observation-execution gap in dynamic tasks.arXiv preprint arXiv:2512.07697, 2025 on PitharXiv
- [54] H. Wang, G. Zhang, Y . Yan, Y . Shang, R. R. Kompella, and G. Liu. Real-time robot execution with masked action chunking.arXiv preprint arXiv:2601.20130, 2026 on PitharXiv
- [55] J. Tang, Y . Sun, Y . Zhao, S. Yang, Y . Lin, Z. Zhang, J. Hou, Y . Lu, Z. Liu, and S. Han. Vlash: Real-time vlas via future-state-aware asynchronous inference.arXiv preprint arXiv:2512.01031, 2025 on PithPith's reviewarXiv
- [56] D. Ruhe, J. Heek, T. Salimans, and E. Hoogeboom. Rolling diffusion models.arXiv preprint arXiv:2402.09470, 2024 on PithPith's reviewarXiv
- [57] J. Kim, J. Kang, J. Choi, and B. Han. Fifo-diffusion: Generating infinite videos from text without training.Advances in Neural Information Processing Systems, 37:89834–89868, 2024 on Pith
- [58] A. Kodaira, T. Hou, J. Hou, M. Georgopoulos, F. Juefei-Xu, M. Tomizuka, and Y . Zhao. Streamdit: Real-time streaming text-to-video generation.arXiv preprint arXiv:2507.03745, 2025 on PitharXiv
- [59] C. Deng, D. Zhu, K. Li, S. Guang, and H. Fan. Causal diffusion transformers for generative modeling.arXiv preprint arXiv:2412.12095, 2024 on PithPith's reviewarXiv
- [60] Z. Zhang, R. Liu, R. Hanocka, and K. Aberman. Tedi: Temporally-entangled diffusion for long-term motion synthesis. InACM SIGGRAPH 2024 Conference Papers, pages 1–11, 2024 on Pith
- [61] S. H. Høeg, Y . Du, and O. Egeland. Streaming diffusion policy: Fast policy synthesis with variable noise diffusion models.arXiv preprint arXiv:2406.04806, 2024 on PithPith's reviewarXiv
- [62] 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 on PithPith's reviewarXiv
- [63] B. Chen, D. Martí Monsó, Y . Du, M. Simchowitz, R. Tedrake, and V . Sitzmann. Diffusion forcing: Next-token prediction meets full-sequence diffusion.Advances in Neural Information Processing Systems, 37:24081–24125, 2024 on Pith
- [64] H. Fang, J. Duan, D. Clay, S. Wang, S. Liu, W. Huang, X. Fan, W.-C. Tsai, S. Chen, Y . R. Wang, et al. Molmoact2: Action reasoning models for real-world deployment.arXiv preprint arXiv:2605.02881, 2026 on PithPith's reviewarXiv
- [65] T. Chen, Z. Chen, B. Chen, Z. Cai, Y . Liu, Z. Li, Q. Liang, X. Lin, Y . Ge, Z. Gu, et al. Robotwin 2.0: A scalable data generator and benchmark with strong domain randomization for robust bimanual robotic manipulation.arXiv preprint arXiv:2506.18088, 2025 on PithPith's reviewarXiv
- [66] J. Bjorck, F. Castañeda, N. Cherniadev, X. Da, R. Ding, L. Fan, Y . Fang, D. Fox, F. Hu, S. Huang, et al. Gr00t n1: An open foundation model for generalist humanoid robots.arXiv preprint arXiv:2503.14734, 2025. 13 on PithPith's reviewarXiv
- [67] J. Zheng, J. Li, Z. Wang, D. Liu, X. Kang, Y . Feng, Y . Zheng, J. Zou, Y . Chen, J. Zeng, et al. X-vla: Soft-prompted transformer as scalable cross-embodiment vision-language-action model. arXiv preprint arXiv:2510.10274, 2025 on PithPith's reviewarXiv
- [68] J. Pai, L. Achenbach, V . Montesinos, B. Forrai, O. Mees, and E. Nava. mimic-video: Video- action models for generalizable robot control beyond vlas.arXiv preprint arXiv:2512.15692, 2025 on PithPith's reviewarXiv
- [69] L. Li, Q. Zhang, Y . Luo, S. Yang, R. Wang, F. Han, M. Yu, Z. Gao, N. Xue, X. Zhu, et al. Causal world modeling for robot control.arXiv preprint arXiv:2601.21998, 2026 on PithPith's reviewarXiv
- [70] S. Lian, B. Yu, X. Lin, Z. Shen, L. T. Yang, Y . Jin, H. Liu, C. Wu, H. Yuan, C. Huang, et al. Intentvla: Short-horizon intent modeling for aliased robot manipulation.arXiv preprint arXiv:2605.14712, 2026 on PithPith's reviewarXiv
- [71] Y . Zhou, C. Barnes, J. Lu, J. Yang, and H. Li. On the continuity of rotation representations in neural networks. InProceedings of the IEEE/CVF conference on computer vision and pattern recognition, pages 5745–5753, 2019. A Failure Case Analysis We expand on the first analysis... on Pith
Challenges and discussion
Disagree with the review, or know something it missed? Leave a comment, or challenge this review formally: a challenge becomes a permanent entry in this paper's record, and its decision is published here.
Come back by email
A short list of the week's newly reviewed papers. Nothing is sent until you confirm from your inbox, and every email has a link that stops it. No account.
No comments yet.
Sign in to comment