Robotics · arXiv:2610.38199 v1 · submitted 28 Oct 2026 · 30 references
Embodied at Sea: Imitation Learning for Multi-Task Underwater Manipulation
The paper says
“Behavior Cloning has demonstrated promising generalization capabilities for robotic manipulation in terrestrial environments. However, marine visual distortions and dynamic currents create a severe domain shift that remains a key bottleneck in adapting generalist architectures to underwater field manipulation.”
more of the abstract
To address this, 1616 episodes of underwater manipulation data were collected in calm and turbulent waters and were used to train 14 generalist Vision-Language-Action (VLA) models (DynamicVLA) with varying proprioception observation configurations. Across three tasks on 1260 gradable on-hardware rollouts in a pool, 10 models outperformed the baseline, with the highest-performing model achieving a 93.3% overall success rate, including partial successes. Results are also introduced from a case study evaluating the performance of a single-task Action Chunking Transformer (ACT) and a generalist VLA (SmolVLA) policy for two tasks in marine archaeology and conservation: grasping and mooring. Experiments are performed on an ROV on and around the WWII wreck Coralita X127 and at Manoel Island Marina, Malta. Across 748 collected episodes and 508 gradable on-hardware rollouts, the ACT policy substantially outperformed SmolVLA (47.9% vs. 7.9% pooled success rate), earning a 66.7% success rate on the most successful task at the marine archaeological site. To enable further contributions to embodied intelligence in marine field robotics, the BLUEMBODIED KIT used to collect data, train models, and deploy policies on the ROV has been open sourced.
from the abstract
Pith says
Machine review · deepseek-v4-flash · 31 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. Genuine field deployment with open data and tools, internally consistent numbers, and a credible first-multi-task-VLA claim — but the 93.3% headline rests in part on wrong-color grabs counted as partials. This line is wrong
3 major objections · 5 minor
How often objections from reviews like this one are right: not yet measured
Not checked: figures, code, what the 30 references say.
The paper says, at Abstract and Table I/II captions
The review did not quote this passage.
Pith says: Objection 1 of 3 · major
All headline success rates, including 93.3% and 47.9%, are computed over 'gradable' rollouts, but the manuscript never defines 'gradable', states the number of excluded rollouts, or gives the exclusion criteria. If ungradable rollouts are disproportionately failures (e.g., hardware loss, camera occlusion, telemetry dropout), every reported rate is inflated. Please report the full numbers of rollouts collected vs. excluded per model, task, and site, and provide the written gradability criteria used during labeling.
The paper says, at Table I caption and row #3
The review did not quote this passage.
Pith says: Objection 2 of 3 · major
For the grab task, a partial success is explicitly a grab of the wrong-color rod. The ATTITUDE model's grab-with-partials rate of 96.7% is composed of 10 correct-color and 19 wrong-color grabs out of 30, so the strict correct-color success rate is only 33.3%.
more of objection 2
Because Contribution 1 defines a VLA partly by correct language-command execution, counting wrong-color grabs as partial successes materially inflates the 93.3% headline and the 10-of-14 comparison. Report strict task success as the primary metric, present wrong-color grabs separately, and quantify how the 'beats baseline' count changes when partials are excluded.
The paper says, at Section IV.A and Table II
The review did not quote this passage.
Pith says: Objection 3 of 3 · major
Rollout outcomes are hand-labeled as success/partial/failure by the same team that built the pipeline, and no partial definition is given for moor or remove-ring, nor any inter-annotator agreement or blind-labeling procedure. Since the paper's central empirical claims rest entirely on these labels, please specify the labeling rubric for every task, state who labeled the rollouts, and report agreement measures if more than one labeler was used.
5 minor objections
The paper says, at Table I
Pith says: minor objection 1 of 5
The table header 'succ. part. % % w/ part.' is ambiguous, and for the Moor and Remove Ring rows it is not clear which numbers are strict success, partial success, or success-with-partials. Redraw the table with explicit per-task columns for strict success, partial, and combined rates.
The paper says, at Section IV.C
Pith says: minor objection 2 of 5
The phrase 'incredible recovery trajectories' is subjective; replace it with a quantitative description, such as the fraction of rollouts that recovered after an initial miss or the measured deviation during recovery maneuvers.
show 3 more
The paper says, at Contribution 1
Pith says: minor objection 3 of 5
The 'world's first' claim is stronger than the evidence supplied; the paper should say 'to our knowledge, the first reported' and briefly describe the scope of the literature search used to support the novelty claim.
The paper says, at Eq. (1)
Pith says: minor objection 4 of 5
The disturbance channel in Eq. (1) depends on the thruster-to-acceleration mapping Phi, but the paper does not state how Phi was estimated. Since the raw-delta models and delta-token models consume this channel, please describe the identification procedure or state the exact offline fit used.
The paper says, at Abstract
Pith says: minor objection 5 of 5
The abstract says 10 models 'outperformed the baseline' without specifying whether the comparison uses strict success or success-with-partials; state the comparison basis explicitly.
What would settle it
Using the open-sourced BLUEMBODIEDKIT rollout logs, re-score every pool and marine rollout under three denominators: gradable-only success as published; all rollouts with ungradable trials counted as failures; and correct-color-only success for grab tasks with wrong-color grabs counted as failures. Report the resulting ATTITUDE overall rate, the ACT 47.9% pooled marine rate, and the number of models beating baseline under each denominator. If any headline rate shifts by more than 5 percentage points, or if the 10-of-14 comparison changes, the reported rates are protocol-dependent.
from the skeptic's reading
How the review went
4 readings by deepseek-v4-flash · 31 Oct 2026
- Reader holds up, with conditions
- Skeptic agrees
- Editor wrote the verdict above
- Referee major revision: 3 major objections, 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 one thing to know: this is a real field-robotics paper with a real deployment — a text-commanded, multi-task VLA actually running on a BlueROV2 at a WWII wreck and a marina, day and night, plus an open-sourced data/training/deployment kit. The internal arithmetic is consistent (1,260 = 90 × 14; 508 = 432 ACT + 76 SmolVLA; the pooled rates in Table II recompute exactly). The central claim, that observation-space and action-space adaptation matters at least as much as the backbone when moving BC underwater, is directionally supported. I would send this to review, not desk-reject it.
What is genuinely new: by the authors' own survey, prior underwater BC is single-task (AquaBot, UMI-Underwater) and prior underwater VLAs were not evaluated on hardware (USim/u0, UnderwaterVLA). So "first multi-task VLA on a real ROV, to our knowledge" is credible. The 14-configuration proprioception ablation is systematic, the depth-setpoint action split (letting a PID own the vertical axis) is a sensible adaptation, and the recovery-trajectory data collection is a practical touch. The authors also name their own confounds — SmolVLA's gap is likely latency and data volume (b131), not an architectural ceiling, and the pool lacks currents (b117).
The soft spots, in order of size. First, the headline. Table I defines a grab partial as grabbing a rod of the wrong color. For the best model, ATTITUDE, strict correct-color grab success is 33.3%; counted with wrong-color grabs, it is 96.7%. That is 19 of 30 rollouts grabbing the wrong color — roughly a fifth of the 93.3% overall. Since a VLA is defined here partly by executing the text command, a wrong-color grab is a command-execution failure earning partial credit. The paper is transparent — the strict column and the Fig. 8 confusion matrix are there — so this is a framing problem, not a hidden result, but the abstract's precision outruns the protocol.
Second, every rate is computed over 'gradable' rollouts, with exclusion counts and criteria unreported, and outcomes are hand-labeled by the same team that built the pipeline (Section IV.A). Partial is undefined for moor and remove. A revision can fix this by reporting exclusions, an inter-rater check, and per-condition confidence intervals.
Minor: n=30 per task per model leaves "10 of 14 beat baseline" directional rather than decisive; the site-agnostic mooring conclusion rests on n=21 and n=19; and the Phi mapping in Eq. (1) is never estimated, so the raw-delta disturbance channel is under-specified. The stress-test arithmetic holds; its word "dominated" is too strong, since wrong-color grabs are about a fifth of the overall headline rather than most of it.
Who gets value: anyone in underwater manipulation, VLA adaptation, or evaluation methodology for field robotics. A reading group would get a genuine hour debating what "success" should mean for language-conditioned policies. Recommendation: conditionally accept — the capability claim is probably real, but the protocol needs revision before the headline numbers can be trusted.
The strongest objection
Contribution 1, the paper's strongest claim, is that a single text-commanded multi-task VLA works on real underwater hardware (abstract; Contribution 1). The load-bearing evidence is the pool ablation's headline rate, 93.3% including partials for the ATTITUDE model (Table I; b110). That rate is computed only over rollouts the authors classed as 'gradable', with exclusion counts and criteria unreported (abstract b7; Table I caption), and every rollout is hand-labeled success/partial/failure by the same team that built the pipeline (Section IV.A). The partial-success definition compounds the problem: for the grab task, a partial is explicitly 'grabbing a rod of the wrong color' (Table I caption), while for moor and remove ring no partial definition is given. For ATTITUDE, grab success with partials is 96.7%, but correct-color success is only 33.3%; 19 of 30 rollouts grabbed the wrong-color rod and still counted toward the headline. Because Contribution 1 defines a VLA partly by correct language-command execution, counting wrong-color grabs as partial successes means the 93.3% headline and the '10 of 14 beat baseline' result are dominated by outcomes that fail the language-command component of the claim. This is not an internal contradiction, but it means the headline precision is not yet supported by the reported evaluation protocol.
The reader's weakest assumption identified the same load-bearing concern: the reported success rates measure the evaluation protocol as much as, or more than, policy competence, because ungradable rollouts are excluded without counts, partial successes are loosely and inconsistently defined, and labeling is done by the same team. I agree with that diagnosis and do not find a separate, stronger objection. I also give credit where the paper has independent support: the arithmetic is internally consistent (1,260 pool rollouts match 14 models times 90; 508 marine rollouts match Table II), the authors acknowledge the SmolVLA gap is likely latency and data quantity rather than architecture (b131), and they acknowledge the pool lacks realistic currents (b117). The main outstanding issue is specifically that the headline numbers, especially 93.3% and the 10-of-14 comparison, include wrong-color grabs as partial successes and exclude an unreported set of ungradable rollouts. That concern does not overturn the central direction of the paper, since many models do beat the visual-only baseline and the marine ACT results are reported per condition, but it does mean the precision of the headline claims is unsupported until the evaluation protocol is reported transparently. The proposed re-scoring check would settle whether the concern actually moves the numbers; if it does not, the protocol criticism is cosmetic, but if it does, the central claim needs to be restated with the more conservative rates.
The referee's report
If the headline numbers withstand audit, this is a substantial field-robotics contribution: the dataset sizes are large for underwater manipulation, the telemetry-ablation study is systematic, the depth-controller action split is a sensible embodiment-specific adaptation, and the open-sourced pipeline is a genuine community asset. The authors are also honest about the main confounds, stating that the SmolVLA gap probably reflects latency and data quantity (Section IV.C) and that the pool lacks realistic currents (Section IV.B). The arithmetic is internally consistent (14 models x 90 rollouts = 1,260; the Table II row counts sum to 508), which is a good sign. The main weakness is not the experimental design but the evaluation protocol: the reported rates depend on unstated 'gradable' exclusions and on a partial-credit definition that counts wrong-color grabs, so the precision of the headline claims is currently not supported by the paper's own reporting.
Pith's reading of the paper (a machine's summary, not the authors')
The question the authors set out to answer is whether the generalist vision-language-action policies that now work on land survive the move to a floating robot in the sea, where currents shove the vehicle, the camera view shifts between every action, and light and turbidity distort what the model sees. Their answer is that the model is not the main lever: what you feed it to observe and how you let it act are. They collect 1,616 teleoperated episodes in a pool, train fourteen variants of one policy architecture that differ only in their sensor inputs, and test them on real hardware, where ten variants beat a camera-only baseline and the best reaches 93.3 percent when partial successes are counted. A separate field study carries two policies to the wreck of the Coralita X127 and to a Maltese marina, where a small single-task policy substantially outperforms a much larger generalist one. The claim worth caring about is that no underwater-pretrained backbone is needed: a policy trained on terrestrial data, given the right attitude readings and a depth set-point handled by a feedback controller, can latch onto a shipwreck. If that holds, marine archaeology and conservation groups get a practical path to autonomous manipulation built from off-the-shelf parts and an open-sourced toolkit.
What carries the argument. Three pieces of adaptation carry the argument. The first is a modular observation space cut from the ROV's 19-channel MAVLink telemetry row: an ATTITUDE block (roll $\phi$, pitch $\theta$, yaw $\psi$), an IMU block that adds body-frame accelerations, EKF-fused angular rates, vertical velocity and three vibration channels, and a thruster-PWM block of the six commanded motor outputs. The second is a disturbance channel: the vehicle's expected acceleration is modelled as $a(t) = G_{\text{orient}}(t) + \Phi u(t) + \Delta(t)$, and the residual $\Delta(t)$ between predicted and measured acceleration is treated as the environmental disturbance; because a single frame carries no turbulence history, a window of $\Delta$ frames (tested from 1.2 to 6.0 seconds) is compressed through the FAST tokenizer so the policy sees a compact signature of the sloshing. The third is an action split: a six-dimensional action $a_t = [v_x, v_y, \omega_z, z_{\text{ref}}, c_{\text{tilt}}, e_{\text{ref}}]$ in which vertical motion is a depth set-point handed to a closed-loop PID controller reading the on-board pressure sensor, rather than a raw vertical command. That last choice is what the authors credit for the vehicle holding station on a sloping shipwreck deck and for the recovery manoeuvres that appear when depth is in the loop.
Core claim. The authors claim, to their knowledge, the first successful deployment of a multi-task vision-language-action model underwater on real hardware, where a single model executes several tasks and is told which one to execute by a text command. The evidence for the multi-task claim comes from the pool: fourteen DynamicVLA variants, differing only in which of a 19-channel telemetry stream they observe and how disturbance is represented, are graded over rollouts on three tasks (grab a rod of a commanded colour, moor on a post, remove a ring). Ten of the fourteen beat the images-only baseline once partial successes are counted, and the best configuration, plain attitude (roll, pitch, yaw) trained on both calm and turbulent episodes, reaches 93.3 percent overall at a 95.2 millisecond forward-pass latency. Crucially, more sensing was not better: adding IMU and thruster-PWM channels helped time-to-completion but not success, and the authors conclude that a compact observation space balances prediction speed against task success. The field evidence is a different comparison: a single-task Action Chunking Transformer versus a generalist SmolVLA policy, trained on site, over 508 gradable rollouts, where ACT succeeds on 47.9 percent pooled against 7.9 percent, and reaches 66.7 percent on mooring a rope at the shipwreck. The unifying claim the authors draw from both is that effective underwater deployment depends on how the action and observation spaces are adapted to the physical environment at least as much as on the choice of model, and that a backbone trained only on terrestrial data suffices.
The premise it rests on. The load-bearing premise is that the reported success rates measure how well the policies actually manipulate, which requires that scoring only the rollouts the authors judged gradable and counting a wrong-colour grab as a partial success did not systematically favour the better-performing configurations.
If the paper is right (Pith's inferences, not the authors')
A single text-commanded policy can be trained on underwater teleoperation data and then run on a tethered, thruster-driven vehicle with commodity hardware, without any underwater-pretrained vision backbone.
Reading a few attitude channels from the autopilot is worth more than adding many sensors: the smallest proprioceptive observation space tested gave the best success rate while larger IMU and thruster-PWM spaces mainly bought faster task completion.
Splitting vertical control out of the learned policy and into a depth set-point tracked by a PID controller lets a policy operate at varying depths on a sloping wreck rather than only at a fixed station.
Small single-task policies remain strong competition for generalist models in the field: the Action Chunking Transformer beat the larger generalist policy on every task family, which the authors attribute to more frequent replanning and more task-specific data rather than to architecture.
Success degrades at night and on harder target geometry, so light budget and target shape, not just model choice, set the practical operating envelope for field deployments.
The pool ranking and the field ranking disagree about which policy class wins, and the authors never run a multi-task model at the shipwreck; a reader should treat the 93.3 percent pool figure and the 47.9 percent field figure as measuring different things, and a single experiment running the best pool configuration on the wreck would be the natural bridge.
Because the authors report success on hand-labelled rollouts over the subset they class as gradable, the headline rates are only comparable across models if the gradable filter and the partial-success rule were applied identically to every configuration, which the paper does not show; a blind regrade of a sample of the rollouts would settle this cheaply.
If the argument that observation and action spaces dominate model choice generalises, the next useful ablation is the reverse of the one run here: fix the best observation space and vary the backbone, which would separate what the spacecraft telemetry did from what the architecture did.
The disturbance-residual channel is the paper's most transferable idea, since the same expected-versus-measured acceleration arithmetic works on any vehicle with an autopilot, and it would be interesting to see whether it helps in a strong current where the tokenised window carries more signal than it did in a pool.
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
Behavior Cloning has demonstrated promising generalization capabilities for robotic manipulation in terrestrial environments. However, marine visual distortions and dynamic currents create a severe domain shift that remains a key bottleneck in adapting generalist architectures to underwater field manipulation. To address this, 1616 episodes of underwater manipulation data were collected in calm and turbulent waters and were used to train 14 generalist Vision-Language-Action (VLA) models (DynamicVLA) with varying proprioception observation configurations. Across three tasks on 1260 gradable on-hardware rollouts in a pool, 10 models outperformed the baseline, with the highest-performing model achieving a 93.3% overall success rate, including partial successes. Results are also introduced from a case study evaluating the performance of a single-task Action Chunking Transformer (ACT) and a generalist VLA (SmolVLA) policy for two tasks in marine archaeology and conservation: grasping and mooring. Experiments are performed on an ROV on and around the WWII wreck Coralita X127 and at Manoel Island Marina, Malta. Across 748 collected episodes and 508 gradable on-hardware rollouts, the ACT policy substantially outperformed SmolVLA (47.9% vs. 7.9% pooled success rate), earning a 66.7% success rate on the most successful task at the marine archaeological site. To enable further contributions to embodied intelligence in marine field robotics, the BLUEMBODIED KIT used to collect data, train models, and deploy policies on the ROV has been open sourced.
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 · 2026-10-30
The authors' main claim, as the classifier picked it out 0.73
“Across three tasks on 1260 gradable on-hardware rollouts in a pool, 10 models outperformed the baseline, with the highest-performing model achieving a 93.3% overall success rate, including partial successes.”
How the main claim is stated: asserted 1.00
What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.45 · human participants 0.06 · controlled comparison 0.94 · randomized assignment 0.02 · states quantitative result 0.99 · compares with prior results 0.97 · mentions code or data availability 0.94
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 |
|---|---|---|---|---|
| Behavior Cloning has demonstrated promising generalization capabilities for robotic manipulation in terrestrial environments. | background 1.00 | 0.06 | 0.02 | asserted 0.98 |
| However, marine visual distortions and dynamic currents create a severe domain shift that remains a key bottleneck in adapting generalist architectures to underwater field manipulation. | gap 0.80 | 0.11 | 0.04 | asserted 1.00 |
| To address this, 1616 episodes of underwater manipulation data were collected in calm and turbulent waters and were used to train 14 generalist Vision-Language-Action (VLA) models (DynamicVLA) with varying proprioception observation configurations. | method 0.97 | 0.96 | 0.05 | asserted 0.99 |
| Across three tasks on 1260 gradable on-hardware rollouts in a pool, 10 models outperformed the baseline, with the highest-performing model achieving a 93.3% overall success rate, including partial successes. | result 1.00 | 0.97 | 0.32 | asserted 1.00 |
| Results are also introduced from a case study evaluating the performance of a single-task Action Chunking Transformer (ACT) and a generalist VLA (SmolVLA) policy for two tasks in marine archaeology and conservation: grasping and mooring. | contribution 0.66 | 0.94 | 0.06 | asserted 0.89 |
| Experiments are performed on an ROV on and around the WWII wreck Coralita X127 and at Manoel Island Marina, Malta. | method 0.95 | 0.92 | 0.02 | asserted 0.98 |
| Across 748 collected episodes and 508 gradable on-hardware rollouts, the ACT policy substantially outperformed SmolVLA (47.9% vs. 7.9% pooled success rate), earning a 66.7% success rate on the most successful task at the marine archaeological site. | result 1.00 | 0.97 | 0.17 | asserted 1.00 |
| To enable further contributions to embodied intelligence in marine field robotics, the BLUEMBODIED KIT used to collect data, train models, and deploy policies on the ROV has been open sourced. | contribution 0.86 | 0.87 | 0.08 | asserted 1.00 |
What kind of paper the classifier reports this is: empirical study 0.74 · evidence from observation or experiment 1.00
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, 2026-10-30: 1 request, 6,615 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 (8)
Figures are the authors', reproduced from the paper.
References (30)
- [1] E. Gouinet al., “Extremely deep bounce dives: planning and physi- ological challenges based on the experiences of a sample of french- speaking technical divers,”Diving Hyperb. Med., vol. 55, no. 2, p. 203, 2025 on Pith
- [2] T. Z. Zhaoet al., “Learning fine-grained bimanual manipulation with low-cost hardware,” arXiv:2304.13705, 2023 on PithPith's reviewarXiv
- [3] R. Liuet al., “Self-improving autonomous underwater manipulation,” arXiv:2410.18969, 2024 on PithPith's reviewarXiv
- [4] H. Liet al., “Umi-underwater: Learning underwater manipulation without underwater teleoperation,” arXiv:2603.27012, 2026 on PitharXiv
- [5] Y . R. Petillotet al., “Underwater robots: From remotely operated ve- hicles to intervention-autonomous underwater vehicles,”IEEE Robot. Autom. Mag., vol. 26, no. 2, pp. 94–101, 2019 on Pith
- [6] O. Khatibet al., “Ocean one: A robotic avatar for oceanic discovery,” IEEE Robot. Autom. Mag., vol. 23, no. 4, pp. 20–29, 2016 on Pith
- [7] E. Morganet al., “Autonomous underwater manipulation: Current trends in dynamics, control, planning, perception, and future direc- tions,”Curr . Robot. Rep., vol. 3, pp. 187–198, 2022 on Pith
- [8] Y . Jadhav and Y . Moon, “Rov manipulation from observation and exploration using deep reinforcement learning,”J. Adv. Res. Ocean Eng., vol. 3, no. 3, pp. 136–148, 2017 on Pith
- [9] I. Carluchoet al., “A reinforcement learning control approach for underwater manipulation under position and torque constraints,” in Proc. Global OCEANS, 2020 on Pith
- [10] C. Chiet al., “Diffusion policy: Visuomotor policy learning via action diffusion,” arXiv:2303.04137, 2024 on PithPith's reviewarXiv
- [11] Z. Fuet al., “Mobile aloha: Learning bimanual mobile manipulation with low-cost whole-body teleoperation,” arXiv:2401.02117, 2024 on PithPith's reviewarXiv
- [12] A. Brohanet al., “RT-1: Robotics Transformer for Real-World Control at Scale,” inProc. Robot.: Sci. Syst. (RSS), 2023 on Pith
- [13] B. Zitkovichet al., “Rt-2: Vision-language-action models transfer web knowledge to robotic control,” inProc. Conf. Robot Learn. (CoRL), 2023, pp. 2165–2183 on Pith
- [14] A. O’Neillet al., “Open x-embodiment: Robotic learning datasets and rt-x models,” arXiv:2310.08864, 2025 on PithPith's reviewarXiv
- [15] M. J. Kimet al., “Openvla: An open-source vision-language-action model,” inProc. Conf. Robot Learn. (CoRL), 2025, pp. 2679–2713 on Pith
- [16] K. Blacket al., “π 0: A vision-language-action flow model for general robot control,” arXiv:2410.24164, 2026 on PithPith's reviewarXiv
- [17] NVIDIAet al., “Gr00t n1: An open foundation model for generalist humanoid robots,” arXiv:2503.14734, 2025 on PithPith's reviewarXiv
- [18] P. Wuet al., “Gello: A general, low-cost, and intuitive teleoperation framework for robot manipulators,” arXiv:2309.13037, 2024 on PithPith's reviewarXiv
- [19] Z. K. Wenget al., “Levr: A modular vr teleoperation framework for imitation learning in dexterous manipulation,” arXiv:2509.14349, 2025 on PitharXiv
- [20] R. Yanget al., “Egovla: Learning vision-language-action models from egocentric human videos,” arXiv:2507.12440, 2025 on PithPith's reviewarXiv
show 10 more
- [21] M. Shukoret al., “Smolvla: A vision-language-action model for affordable and efficient robotics,” arXiv:2506.01844, 2025 on PithPith's reviewarXiv
- [22] H. Xieet al., “Dynamicvla: A vision-language-action model for dynamic object manipulation,” arXiv:2601.22153, 2026 on PithPith's reviewarXiv
- [23] J. Tuckeret al., “π, but make it fly: Physics-guided transfer of vla models to aerial manipulation,” arXiv:2603.25038, 2026 on PitharXiv
- [24] C. Chiet al., “Universal Manipulation Interface: In-The-Wild Robot Teaching Without In-The-Wild Robots,” inProc. Robot.: Sci. Syst. (RSS), 2024 on Pith
- [25] J. Songet al., “Oceansim: A gpu-accelerated underwater robot per- ception simulation framework,” inProc. IEEE/RSJ Int. Conf. Intell. Robots Syst. (IROS), 2025 on Pith
- [26] J. Guet al., “Usim and u0: A vision-language-action dataset and model for general underwater robots,” arXiv:2510.07869, 2025 on PithPith's reviewarXiv
- [27] Z. Wanget al., “Underwatervla: Dual-brain vision-language-action ar- chitecture for autonomous underwater navigation,” arXiv:2509.22441, 2025 on PitharXiv
- [28] R. Zhaoet al., “Sim2real-vla: Zero-shot generalization of synthesized skills to realistic manipulation,” inProc. Int. Conf. Learn. Represent. (ICLR), 2026 on Pith
- [29] K. Pertschet al., “Fast: Efficient action tokenization for vision- language-action models,” arXiv:2501.09747, 2025 on PithPith's reviewarXiv
- [30] R. Cadeneet al., “Lerobot: An open-source library for end-to-end robot learning,” arXiv:2602.22818, 2026 on PitharXiv
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