Pith.

Robotics · arXiv:2611.07964 v1 · submitted 5 Nov 2026 · 36 references

Time-Indexed Skill Success Models for Latency-Aware Skill Selection and Timing in Dynamic Manipulation

Mohsen Omidi, Maria Chiara Leva, Brian Vaughan

The paper says

“Robots that manipulate moving objects act on observations that are outdated when the chosen skill starts, and the best skill can change within a few hundred milliseconds.”

more of the abstract

Skill competence models, however, predict success for a launch at the moment of observation. We propose time-indexed skill success models, which predict whole-task success when a skill is launched after a given delay. Trained on rollouts branched at a grid of launch delays, they let a receding-horizon planner select both the skill and its launch time under fixed or random latency. In a MuJoCo study in which a UR5e captures a sliding puck with three skills, the planner exceeded a static competence model by 2.8 to 6.2 percentage points at 100 to 300 ms of latency, including when that model was evaluated on point or sampled forecasts of the launch state. Across five complete runs (three training seeds and two distribution shifts), all 80 paired 95% intervals against four baselines excluded zero. On delayed launches, the time-indexed model was calibrated (expected calibration error 0.009 against 0.239), and pausing the world during latency overstated success by up to 23 points. On a real UR5e that captures a rolling ball after 0.2 s of injected latency, a static model trained in a hardware-matched simulator predicted 99% success on a prespecified subset of 35 delayed launches that it rated at least 0.8. Of these, 63% succeeded (95% interval 46 to 77%), and the time-indexed model predicted 56%. In this single-skill, predictable regime, a matched physics-based forecast was also well calibrated.

from the abstract

Pith says

Machine review · deepseek-v4-flash · 6 Nov 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. Delay-indexed success models are a genuine increment with careful simulation evidence; the planner gains rest on MuJoCo alone, no artifacts are released, and the paper deserves review with those as conditions. This line is wrong

3 major objections · 2 minor

How often objections from reviews like this one are right: not yet measured

Are you an author? What you can do.

Not checked: figures, code, what the 36 references say.

The paper says, at Section IV-C and Algorithm 1

The review did not quote this passage.

Pith says: Objection 1 of 3 · major

Equation (2) and the surrounding text say that the planner selects a skill k* and an extra wait delta*, with the skill starting at t + ell + delta*. In Algorithm 1, however, a nonzero delta* is never executed: the robot only holds the ready pose and re-decides after max(Delta_t, ell), and a launch is triggered only when delta* = 0, the wait exceeds delta_max, or the object is lost.

more of objection 1

The implemented policy is therefore a one-step wait-and-replan rule ('launch now if the best candidate delay is zero, otherwise wait one control period and replan'), not a planner that commits to the launch time returned by Eq. (2). Please state this explicitly and define the candidate wait as part of an inner optimization whose first action is executed; otherwise the reader cannot tell which policy generated Table I.

The paper says, at Section IV-C and Table S1

The review did not quote this passage.

Pith says: Objection 2 of 3 · major

The training grid T stops at 1.3 s, while the planner can query G_k for launch delays up to ell + delta = 1.5 s under the fixed 300 ms latency, and up to about 1.7 s under the capped heavy-tailed latency (or at least 1.45 s with the largest mid-quantile used by the -dist planners).

more of objection 2

Section VIII says the models 'interpolate between nine grid delays', but the optimization in Eq. (2) can evaluate the model outside its training support. Please restrict candidate waits so that ell + delta stays inside T, extend the grid, or state and justify the extrapolation; the calibration and exact-label guarantees are established on the grid, not beyond it.

The paper says, at Sections VII and VIII

The review did not quote this passage.

Pith says: Objection 3 of 3 · major

The central quantitative claim is the planner advantage in Table I and Tables S4/S6, but this comparison is entirely within MuJoCo with analytic skills and a synthetic tracker. The hardware experiment is explicitly scoped to calibration and does not exercise the planner, as the authors acknowledge in Section VII and Section VIII, and I agree that a hardware planner comparison would require an impractical number of trials.

more of objection 3

What remains unresolved is reproducibility: no code, data, trained models, or simulator configuration is released, so the headline comparison cannot be independently checked from the text. Given the modest 3 to 6 point effect sizes, I ask for a release artifact or at minimum a detailed protocol with exact simulator settings and model checkpoints so that the central claim is verifiable.

2 minor objections

The paper says, at Algorithm 1, step 10

Pith says: minor objection 1 of 2

The instruction to 'decide again after max(Delta_t, ell)' is ambiguous in the random-latency setting, where the planner knows only the distribution of ell but not the realized value. Please specify whether this schedule uses the realized latency, an upper bound, or a quantile, and how it is implemented for the -dist variants.

The paper says, at Section VI-B and Table I

Pith says: minor objection 2 of 2

The text says TI-flat waited for at least one decision period in 28 to 36% of episodes. Because the loop in Algorithm 1 waits max(Delta_t, ell) rather than Delta_t, it would help to state explicitly how a 'decision period' is counted in this statistic, especially under random latency.

What would settle it

Run the paper's planner comparison on the real UR5e for the slow-launch band at ℓ = 200 ms, using the authors' own power estimate of roughly 380 trials per planner (two-sided α = 0.05, 80% power, Section VIII), with the TI-flat versus B1 and TI-flat versus B1s paired differences pre-registered before data collection. If the 95% interval for either difference includes zero, the central quantitative claim is not confirmed outside simulation; if both intervals exclude zero, the concern is settled in the paper's favor.

from the skeptic's reading

How the review went

4 readings by deepseek-v4-flash · 6 Nov 2026

  1. Reader holds up, with conditions confidence moderate
  2. Skeptic partly agrees 8 min 3 s · 26,049 tokens read, 28,043 written
  3. Editor wrote the verdict above 43 s · 26,938 tokens read, 6,727 written
  4. Referee major revision: 3 major objections, 2 minor 1 min 55 s · 26,395 tokens read, 17,586 written

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

Two things you should know. The core idea is real: a skill success model that takes launch delay τ as an input, Gk(z, τ), trained with a branching rollout recipe, driving a planner that chooses both skill and wait time. The cited competence and delay-compensation literature does not contain this delay-indexed formulation or the branched labeling procedure. And the paper is honest in ways that matter — it says plainly where its evidence stops, and its evidence stops short of the headline claim in one respect.

What it does well. Proposition 1 is one line from the tower property, but it earns its place: it shows a static model on a point forecast is biased, equation (4) puts the error on the forecast covariance, and the pair predicts exactly the regime where a matched physics forecast suffices — which the hardware test then shows. The branched rollout scheme is a reusable recipe: M+1 rollouts per episode yield (M+1)(M+2)/2 exact (state, delay, outcome) pairs because the robot idles at the ready pose. The evaluation protocol is careful: paired episodes, bootstrap intervals, five complete runs across seeds and shifted dynamics, and a frozen-clock control showing up to 23 points of overstatement. The hardware result is the most striking single finding: a static model trained in a matched simulator predicted 99% on 35 delayed launches, 63% succeeded, and the time-indexed model predicted 56%. Brier over all 49 launches fell from 0.260 to 0.031.

Soft spots, in proportion. The headline planner gains (2.8–6.2 points over the best baselines at 100–300 ms) come from MuJoCo alone, and no code or data are released to check them. The stress-test note is fair: the one real-robot comparison that runs a forecast against the time-indexed model slightly favors the forecast (Brier difference −0.011, interval excluding zero). The authors explain this via Proposition 1 — small forecast spread relative to a 16 mm tolerance — but it still caps the claim. They flag all of this in Section VIII, including that the hardware test was never meant to exercise the planner. The ready-pose idling assumption is load-bearing for the exactness of the branched pairs; they flag that pre-positioning would break it. The gains are modest in absolute terms, and they say so. Citation pattern looks clean.

I agree with the reader's CONDITIONAL verdict. The rigor is real; the significance is capped by simulation-only planner evidence and missing artifacts. Who it is for: anyone working on skill selection, competence models, or calibration under latency. It deserves a serious referee. My recommendation: send it, and make artifact release plus some out-of-simulation check of the planner the explicit conditions for acceptance.

The strongest objection

The central claim is the receding-horizon planner's 2.8 to 6.2 point advantage over static and forecast baselines at 100 to 300 ms of latency (Table I, S4, S6). For this gain to carry the paper's real-robot motivation, the MuJoCo evaluation with analytic skills and a synthetic tracker must be representative. The hardware experiment (Section VII) is explicitly scoped to calibration: one skill, one delay, 49 launches on one day, and the authors state it is not meant to show planner outperformance (b174, b230). It therefore does not exercise the central quantitative claim. The authors acknowledge this limitation, but what remains unresolved is that no code, data, or independent implementation is released, so the headline comparison cannot be checked from the text alone, and no term of the planner comparison has out-of-simulation validation. The fact that the one real-robot comparison actually run (hardware-matched forecast versus time-indexed, Section VII-C) slightly favors the forecast is consistent with the conditions in Proposition 1 and Equation (4), but it also underscores that the relative gain over forecast-based baselines is task-dependent; the only task in which the planner gain is demonstrated is synthetic. This is a scope and reproducibility risk, not an internal inconsistency.

Read in good faith: the paper defines G_k(z, τ) cleanly, branched rollouts provide exact labels under the stated ready-pose idling condition, the planner comparison is paired over the same 2,000 episodes with episode-bootstrap confidence intervals, and replication over five runs is reported transparently. Proposition 1 and the forecast-error expansion in Equation (4) correctly identify when forecast baselines should suffice, and the hardware result in which a matched physics forecast slightly beats the time-indexed model is consistent with that theory rather than a contradiction. The most load-bearing risk is not a mathematical slip but the gap between the central quantitative claim (planner gains) and the evidence: the hardware test deliberately does not exercise the planner, and no artifact is released for independent checking. Because the authors state this limitation explicitly, I do not treat it as a hidden flaw; I also do not see an internal inconsistency that would justify rejection. The reader's CONDITIONAL verdict is appropriate; this concern strengthens the reason for the condition rather than changing it, so I would keep the verdict unchanged.

The referee's report

The core idea is simple, clearly formulated, and a natural extension of competence-model planning: instead of asking whether a skill succeeds if launched immediately, ask whether it succeeds if launched after the latency. If the results hold, this gives an actionable way to make skill selection robust to perception and computation delay, and the branched-rollout construction is a clean way to generate exact labels under the stated idling assumption. The evaluation is unusually careful: paired episodes, bootstrap intervals, five complete runs, a frozen-clock control, and a hardware protocol with a pre-specified subset and an a priori success criterion. The Proposition 1/Eq. (4) analysis provides a useful diagnostic for when forecast-based correction is sufficient, and the hardware result is reported honestly, including the regime where a matched forecast is competitive. The main residual risk is that the headline planner gain has no out-of-simulation validation and the implementation is not released, which limits independent checking of the central quantitative claim.

Pith's reading of the paper (a machine's summary, not the authors')

The paper claims that a robot manipulating a moving object should score each skill by its whole-task success when launched after a delay, not by the success of an immediate launch from the observed state, because 100-300 ms of perception-to-action latency can change which skill is best. It defines the time-indexed success model $G_k(z, \tau)$, the probability that the task succeeds if skill $k$ starts $\tau$ seconds after observing state $z$, and shows that evaluating a static competence model on a forecast of the launch state is exact only in expectation: a point forecast errs roughly in proportion to the forecast spread times the local curvature of the success function. The enabling mechanism is branched rollouts, which give exact training labels because the robot idles at a ready pose, so a rollout launched from each grid-time snapshot is a valid sample for every earlier decision. A receding-horizon planner using the time-indexed model beat static, forecast-based, and rule-based planners by 2.8-6.2 percentage points at 100-300 ms injected latency in simulation, with all 80 paired 95% intervals above zero across five runs, and was calibrated on delayed launches where the static model was not (ECE 0.009 versus 0.239). On a real robot catching a rolling ball, the static model predicted 99% success on its 35 confident delayed launches, of which only 63% succeeded, while the time-indexed model predicted 56%.

What carries the argument. The central object is the time-indexed skill success model $G_k(z, \tau)$ — the probability that the whole task succeeds if skill $k$ is launched $\tau$ seconds after the decision state $z$ was observed — together with the identity $G_k(z, \tau) = \mathbb{E}[G_k(z_{t+\tau}, 0) \mid z_t = z]$, which holds because the robot idles until launch and the decision state is a sufficient statistic. This identity turns the delay problem into a function-approximation problem: the delay enters as an input to a classifier, and the model needs no forecast model at run time. The training mechanism that supplies the exact labels is branching: the simulator runs once to each grid delay, a snapshot is stored, every skill is launched from every snapshot, and each rollout's outcome labels all earlier decisions, producing $(M+1)(M+2)/2$ exact pairs per episode (45 for the nine-delay grid). Around this sits a receding-horizon planner that every 0.1 s chooses the skill and extra wait maximizing the expectation of $G_k$ under the latency distribution, using eight equal-probability quantiles when only the distribution is known.

Core claim. The central claim is that latency-aware skill selection should use the whole-task success probability as a function of the decision state and the launch delay, $G_k(z, \tau) = \Pr(\text{success} \mid z_t = z,\ k \text{ starts at } t+\tau)$, and that this function is learnable exactly from branched rollouts. Proposition 1 states that when the robot idles until launch and $z_t$ is a sufficient statistic, $G_k(z, \tau) = \mathbb{E}[G_k(z_{t+\tau}, 0) \mid z_t = z]$, so forecast-based correction is exact only on average: to second order the point-forecast error is $\tfrac{1}{2}\mathrm{tr}(\nabla_z^2 G_k(\hat z_\tau, 0)\, \Sigma_\tau)$, large where success changes sharply, for example at the edge of a capture tolerance. The training mechanism is branching: with the robot held at its ready pose the world evolves identically whether or not a skill will launch, so a rollout branched at grid delay $\tau_j$ yields an exact labelled pair $(z_i, \tau_j - \tau_i, y)$ for every earlier decision, giving $(M+1)(M+2)/2$ pairs per episode with no off-policy correction. The resulting TI-flat planner was the best planner at every tested latency, exceeding the static model by 2.9-5.3 points, the sampled-forecast model by 2.8-5.7, the point-forecast model by 4.0-6.2, and the rule-based planner by 3.9-7.2 at 100-300 ms, with every paired 95% interval above zero in all five runs, and a factored variant differing from the static baseline only by the delay input gained 1.3-4.9 points. The claim also includes a prediction about calibration: on delayed launches the time-indexed model was calibrated (ECE 0.009) where the static model was severely overconfident (ECE 0.239), and the hardware test reproduced the gap, with the static model's 99% mean prediction on its confident delayed subset corresponding to a 63% observed success rate.

The premise it rests on. The load-bearing premise is that the robot does nothing during the latency and the chosen wait, so a rollout branched from a later snapshot is an exact training label (and the identity $G_k(z, \tau) = \mathbb{E}[G_k(z_{t+\tau}, 0) \mid z_t = z]$ holds) for every earlier decision; if the robot pre-positions, moves during the delay, or runs a learned policy before launch, that exactness breaks and the method must be reworked, as the authors acknowledge.

If the paper is right (Pith's inferences, not the authors')

Any static competence model can be upgraded to a time-indexed one without new online machinery: the same observation is fed to the network with $\tau$ appended, and the training data come from branched rollouts taken while the robot idles.

Forecast-based delay compensation is sufficient only when the forecast spread is small relative to the skill's tolerance; the second-order error formula identifies where point forecasting breaks, and the experiments show sampled forecasts recover only part of the loss.

Evaluation protocol matters: freezing the world during latency overstated adaptive planners by 6 to 23 points and hid most of the reported differences, so dynamic-manipulation benchmarks should keep the clock running during inference.

Knowing only the distribution of the per-decision latency captured most of the attainable benefit, since telling the planner the realized latency of each decision added at most about one point.

A single temperature rescale cannot repair static models across delays, because recalibrating for delayed launches degraded zero-delay calibration; the overconfidence is structural, not a scaling artefact.

The identity (3) suggests a quantitative criterion for when time-indexing matters most: the gain over forecasting should grow with the ratio of forecast spread to the task's success-tolerance curvature, which is why the hardware demo (millimetre spread versus a 16 mm tolerance, one skill) showed near-parity while the three-skill simulation showed the largest gaps.

The branched-rollout trick transfers to any task where the robot can commit to idling while the world evolves, but as soon as the delay is used productively — pre-positioning, or a learned policy acting before launch — exact labels disappear and the method would need importance weights or a pre-launch policy model; the authors flag the break but not the repair.

A cheap diagnostic follows from the hardware result: systems that output competence scores for dynamic tasks should be audited by binning predicted success against observed outcomes conditioned on inference delay, since even a model that is calibrated at zero delay can be badly overconfident on delayed launches.

The same delay-indexing could be folded into the confidence heads of vision-language-action policies, making competence estimates a function of expected start time rather than a post hoc forecast correction.

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

Every value above, in the review's public record (JSON)

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.

Request a human review

How Pith checks papers · The full audit of this paper

The paper

Abstract

Robots that manipulate moving objects act on observations that are outdated when the chosen skill starts, and the best skill can change within a few hundred milliseconds. Skill competence models, however, predict success for a launch at the moment of observation. We propose time-indexed skill success models, which predict whole-task success when a skill is launched after a given delay. Trained on rollouts branched at a grid of launch delays, they let a receding-horizon planner select both the skill and its launch time under fixed or random latency. In a MuJoCo study in which a UR5e captures a sliding puck with three skills, the planner exceeded a static competence model by 2.8 to 6.2 percentage points at 100 to 300 ms of latency, including when that model was evaluated on point or sampled forecasts of the launch state. Across five complete runs (three training seeds and two distribution shifts), all 80 paired 95% intervals against four baselines excluded zero. On delayed launches, the time-indexed model was calibrated (expected calibration error 0.009 against 0.239), and pausing the world during latency overstated success by up to 23 points. On a real UR5e that captures a rolling ball after 0.2 s of injected latency, a static model trained in a hardware-matched simulator predicted 99% success on a prespecified subset of 35 delayed launches that it rated at least 0.8. Of these, 63% succeeded (95% interval 46 to 77%), and the time-indexed model predicted 56%. In this single-skill, predictable regime, a matched physics-based forecast was also well calibrated.

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 · 2026-11-06

The authors' main claim, as the classifier picked it out 0.84

“In a MuJoCo study in which a UR5e captures a sliding puck with three skills, the planner exceeded a static competence model by 2.8 to 6.2 percentage points at 100 to 300 ms of latency, including when that model was evaluated on point or sampled forecasts of the launch state.”

How the main claim is stated: asserted 1.00

What the abstract itself states, as probabilities: claims first 0.07 · states limitation 0.56 · human participants 0.02 · controlled comparison 0.96 · randomized assignment 0.10 · states quantitative result 0.99 · compares with prior results 0.98 · 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.

SentenceRoleThe authors' own workStates the main conclusionStrength
Robots that manipulate moving objects act on observations that are outdated when the chosen skill starts, and the best skill can change within a few hundred milliseconds. background 0.99 0.11 0.03 asserted 1.00
Skill competence models, however, predict success for a launch at the moment of observation. gap 0.87 0.08 0.03 asserted 1.00
We propose time-indexed skill success models, which predict whole-task success when a skill is launched after a given delay. contribution 0.89 0.80 0.32 asserted 0.99
Trained on rollouts branched at a grid of launch delays, they let a receding-horizon planner select both the skill and its launch time under fixed or random latency. method 0.94 0.88 0.14 asserted 1.00
In a MuJoCo study in which a UR5e captures a sliding puck with three skills, the planner exceeded a static competence model by 2.8 to 6.2 percentage points at 100 to 300 ms of latency, including when that model was evaluated on point or sampled forecasts of the launch state. result 1.00 0.95 0.39 asserted 1.00
Across five complete runs (three training seeds and two distribution shifts), all 80 paired 95% intervals against four baselines excluded zero. result 1.00 0.95 0.15 asserted 1.00
On delayed launches, the time-indexed model was calibrated (expected calibration error 0.009 against 0.239), and pausing the world during latency overstated success by up to 23 points. result 1.00 0.93 0.11 asserted 1.00
On a real UR5e that captures a rolling ball after 0.2 s of injected latency, a static model trained in a hardware-matched simulator predicted 99% success on a prespecified subset of 35 delayed launches that it rated at least 0.8. result 0.86 0.89 0.06 asserted 1.00
Of these, 63% succeeded (95% interval 46 to 77%), and the time-indexed model predicted 56%. result 1.00 0.93 0.09 asserted 1.00
In this single-skill, predictable regime, a matched physics-based forecast was also well calibrated. result 0.60 0.83 0.04 asserted 0.99

What kind of paper the classifier reports this is: method or tool 1.00 · evidence from observation or experiment 0.80

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, 2026-11-06: 1 request, 7,780 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 (4)
Figure 1 of the paper
Figure 1. Task and timing problem. Left: A UR5e must capture a puck that slides towards an open edge and place it in a bin. Every 0.1 s, it chooses one of
Figure 2 of the paper
Figure 2. Paired differences in whole-task success between TI-flat and each
Figure 3 of the paper
Figure 3. InterceptGrasp skill-success predictors on held-out launches. (a)
Figure 4 of the paper
Figure 4. Hardware test. (a) Set-up: a UR5e with an open Hand-E gripper captures a golf ball rolling down an inclined board (blue arrow) at the catch line

Figures are the authors', reproduced from the paper.

References (36)
  1. [1] K. Black, M. Y . Galliker, and S. Levine, “Real-time execution of action chunking flow policies,” inAdvances in Neural Information Processing Systems, 2025 on Pith
  2. [2] 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
  3. [3] H. Wei, X. Xu, Z. Cheng, H. Yin, A. Ma, B. Yu, J. Zhou, and J. Lu, “F2F-AP: Flow-to-future asynchronous policy for real-time dynamic manipulation,”arXiv preprint arXiv:2604.02408, 2026 on PithPith's reviewarXiv
  4. [4] G. Konidaris, L. P. Kaelbling, and T. Lozano-P ´erez, “From skills to symbols: Learning symbolic representations for abstract high-level planning,”Journal of Artificial Intelligence Research, vol. 61, pp. 215– 289, 2018 on Pith
  5. [5] A. Bagaria, B. Abbatematteo, O. Gottesman, M. Corsaro, S. Rammohan, and G. Konidaris, “Effectively learning initiation sets in hierarchical reinforcement learning,”Advances in Neural Information Processing Systems, vol. 36, pp. 73 441–73 463, 2023 on Pith
  6. [6] M. Ahn, A. Brohan, N. Brown, Y . Chebotar, O. Cortes, B. David, C. Finn, C. Fu, K. Gopalakrishnan, K. Hausmanet al., “Do as I can, not as I say: Grounding language in robotic affordances,”arXiv preprint arXiv:2204.01691, 2022 on PithPith's reviewarXiv
  7. [7] C. Agia, T. Migimatsu, J. Wu, and J. Bohg, “STAP: Sequencing task- agnostic policies,” inProc. IEEE Int. Conf. Robotics and Automation (ICRA), 2023 on Pith
  8. [8] G. Liu, J. De Winter, Y . Durodi ´e, D. Steckelmacher, A. Nowe, and B. Vanderborght, “Optimistic reinforcement learning-based skill inser- tions for task and motion planning,”IEEE Robotics and Automation Letters, vol. 9, no. 6, pp. 5974–5981, 2024 on Pith
  9. [9] O. J. M. Smith, “Closer control of loops with dead time,”Chemical Engineering Progress, vol. 53, no. 5, pp. 217–219, 1957 on Pith
  10. [10] 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,” 2026, arXiv:2512.01031 on PithPith's reviewarXiv
  11. [11] 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
  12. [12] K. V . Katsikopoulos and S. E. Engelbrecht, “Markov decision processes with delays and asynchronous cost collection,”IEEE Transactions on Automatic Control, vol. 48, no. 4, pp. 568–574, 2003 on Pith
  13. [13] S. Ramstedt and C. Pal, “Real-time reinforcement learning,” inAdvances in Neural Information Processing Systems, 2019 on Pith
  14. [14] Y . Bouteiller, S. Ramstedt, G. Beltrame, C. Pal, and J. Binas, “Rein- forcement learning with random delays,” inProc. Int. Conf. Learning Representations (ICLR), 2021 on Pith
  15. [15] M. Li, Y .-X. Wang, and D. Ramanan, “Towards streaming perception,” inProc. European Conf. Computer Vision (ECCV), 2020 on Pith
  16. [16] Y . Chen, W. Zhang, and X. Li, “Reflex: Enabling fast and predictive vision-language-action models for reaction-critical manipulation,”arXiv preprint arXiv:2608.14379, 2026 on PithPith's reviewarXiv
  17. [17] S. S. Shperberg, A. Coles, B. Cserna, E. Karpas, W. Ruml, and S. E. Shimony, “Allocating planning effort when actions expire,” in Proceedings of the AAAI Conference on Artificial Intelligence, vol. 33, no. 01, 2019, pp. 2371–2378 on Pith
  18. [18] A. Coles, E. Karpas, S. E. Shimony, S. Shperberg, and W. Ruml, “Concurrent planning and execution using dispatch-dependent values,” inProc. Int. Joint Conf. Artificial Intelligence (IJCAI), 2025 on Pith
  19. [19] Y . Sung, S. S. Shperberg, Q. Wang, and P. Stone, “Effort allocation for deadline-aware task and motion planning: A metareasoning approach,” 2024, arXiv:2410.05828 on PithPith's reviewarXiv
  20. [20] Z. Yang, C. R. Garrett, T. Lozano-P ´erez, L. P. Kaelbling, and D. Fox, “Sequence-based plan feasibility prediction for efficient task and motion planning,” inProc. Robotics: Science and Systems (RSS), 2023 on Pith
show 16 more
  1. [21] T. P. Zollo and R. Zemel, “Confidence calibration in vision-language- action models,” 2025, arXiv:2507.17383 on PithPith's reviewarXiv
  2. [22] I. Akinola, J. Xu, S. Song, and P. K. Allen, “Dynamic grasping with reachability and motion awareness,” inProc. IEEE/RSJ Int. Conf. Intelligent Robots and Systems (IROS), 2021 on Pith
  3. [23] G. Olin, L. Chen, N. Gandotra, M. Likhachev, and H. Choset, “Think fast: Real-time kinodynamic belief-space planning for projectile in- terception,” in2026 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2026, pp. 22 285–22 291 on Pith
  4. [24] L. Chen, A. Garrett, A. Niu, and Z. Kingston, “Becoming a fruit ninja: Real-time probabilistic kinodynamic planning for manipulator projectile interception,”arXiv preprint arXiv:2609.22608, 2026 on PithPith's reviewarXiv
  5. [25] S. Park, W. Lee, S. Jin, Y . Kim, J. Park, and J. C. Ye, “Dynamic manipulation with world-action models via counterfactual planning,” 2026, arXiv:2609.33172 on PithPith's reviewarXiv
  6. [26] P. Liu, J. G ¨unster, N. Funk, S. Gr ¨oger, D. Chen, H. Bou-Ammar, others, and J. Peters, “A retrospective on the robot air hockey challenge: Benchmarking robust, reliable, and safe learning techniques for real- world robotics,” inAdvances in Neural Information Processing Syst... on Pith
  7. [27] Z. Wang, A. Boularias, K. M ¨ulling, B. Sch ¨olkopf, and J. Peters, “Anticipatory action selection for human-robot table tennis,”Artificial Intelligence, vol. 247, pp. 399–414, 2017 on Pith
  8. [28] H. E. Zhang, R. Geng, Y . Li, Y . Niu, Y . Wang, R. Haque, K. Zbiss, G. Luo, H.-p. Wang, H. E. Tseng, and D. Zhao, “Anticipatory robot goalkeeping via monotone optimal stopping,” 2026, arXiv:2609.23976 on PithPith's reviewarXiv
  9. [29] C. Guo, G. Pleiss, Y . Sun, and K. Q. Weinberger, “On calibration of modern neural networks,” inProc. Int. Conf. Machine Learning (ICML), ser. PMLR, vol. 70, 2017 on Pith
  10. [30] K. Zakka, Y . Tassa, and MuJoCo Menagerie Contributors, “MuJoCo Menagerie: A collection of high-quality simulation models for MuJoCo,” https://github.com/google-deepmind/mujoco menagerie, 2022 on Pith
  11. [31] E. Todorov, T. Erez, and Y . Tassa, “MuJoCo: A physics engine for model-based control,” inProc. IEEE/RSJ Int. Conf. Intelligent Robots and Systems (IROS), 2012, pp. 5026–5033 on Pith
  12. [32] Google DeepMind and NVIDIA, “MuJoCo Warp: A GPU-accelerated implementation of MuJoCo,” https://github.com/google-deepmind/ mujoco warp, 2025, version 3.14.0 on Pith
  13. [33] B. Efron and R. J. Tibshirani,An Introduction to the Bootstrap. Chap- man and Hall/CRC, 1994 on Pith
  14. [34] G. W. Brier, “Verification of forecasts expressed in terms of probability,” Monthly Weather Review, vol. 78, no. 1, pp. 1–3, 1950 on Pith
  15. [35] E. B. Wilson, “Probable inference, the law of succession, and statistical inference,”Journal of the American Statistical Association, vol. 22, no. 158, pp. 209–212, 1927. OMIDIet al.: TIME-INDEXED SKILL SUCCESS MODELS FOR LATENCY-AW ARE SKILL SELECTION 9 on Pith
  16. [36] F. Pedregosa, G. Varoquaux, A. Gramfort, V . Michel, B. Thirion, O. Grisel, M. Blondel, P. Prettenhofer, R. Weiss, V . Dubourg, J. Vander- plas, A. Passos, D. Cournapeau, M. Brucher, M. Perrot, and ´E. Duch- esnay, “Scikit-learn: Machine learning in Python,”Journal of Machine ... 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.

No comments yet.

Sign in to comment