Pith. sign in

REVIEW 3 major objections 5 cited by

A closed loop of seeing milestones, planning to the next waypoint, and rewinding on stall recovers robotic manipulation failures without extra training data or models.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review · grok-4.5

2026-07-15 00:21 UTC pith:7W2VXCD7

load-bearing objection We cannot review SPR: the supplied full text is a different paper (MIMO-AFDM channel estimation), so the robotics claims stay abstract-only and unauditable. the 3 major comments →

arxiv 2603.09292 v2 pith:7W2VXCD7 submitted 2026-03-10 cs.RO cs.CV

See, Plan, Rewind: Progress-Aware Vision-Language-Action Models for Robust Robotic Manipulation

Tingjun Dai , Mingfei Han , Tingwen Du , Zhiheng Liu , Zihao Zhang , Zhihui Li , Salman Khan , Jun Yu
show 1 more author
Xiaojun Chang
This is my paper
classification cs.RO cs.CV
keywords vision-language-actionrobotic manipulationprogress awarenessspatial subgoalsclosed-loop recoveryLIBEROout-of-distribution robustness
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

Robust robot manipulation needs more than open-loop action prediction: the system must know how far it has come, what intermediate spatial milestone should come next, and when progress has stalled so it can recover. This paper introduces SPR (See, Plan, Rewind), a vision-language-action framework that turns a language instruction into an explicit sequence of spatial subgoals and runs a continuous cycle: observe the current scene and the upcoming milestone, plan a trajectory toward the next 2D waypoint, and rewind to a recoverable earlier state when progress against that sequence stalls. The claim is that this progress-aware closed loop yields reliable error correction without collecting more training data or bolting on separate recovery models. On the LIBERO benchmark the approach improves over a MolmoAct baseline by five percentage points; on the harder LIBERO-Plus suite with unseen instructions and initial states it shows the smallest performance drop among compared systems, indicating stronger out-of-distribution robustness.

Core claim

Measuring task progress through explicit, actionable spatial milestones lets a vision-language-action model ground its current status, anticipate verifiable intermediate states, and detect and recover from failures when progress stalls—implemented as the continuous See-Plan-Rewind cycle that dynamically grounds language into spatial subgoals and rewinds without additional training data or auxiliary models—producing a 5% gain over MolmoAct on LIBERO and the smallest drop on LIBERO-Plus versus OpenVLA-OFT and UniVLA.

What carries the argument

The See-Plan-Rewind (SPR) cycle: continuous closed-loop operation that Sees the current state and next milestone, Plans a trajectory to the next 2D waypoint, and Rewinds to a recoverable state when progress against the expected spatial-subgoal sequence stalls.

Load-bearing premise

Monitoring progress against an expected sequence of spatial milestones is enough, by itself, to detect failures and choose a recoverable rewind state without any extra training data or auxiliary models.

What would settle it

On LIBERO or LIBERO-Plus tasks, disable or corrupt milestone progress monitoring (or force incorrect rewind targets) while keeping the same base policy; if success rate and out-of-distribution drop no longer improve over MolmoAct / OpenVLA-OFT / UniVLA, the closed-loop recovery claim fails.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Progress-aware VLAs can recover from stalled manipulation without collecting failure-recovery datasets or training separate recovery heads.
  • Grounding instructions as ordered spatial subgoals plus 2D waypoints gives a verifiable intermediate representation for status and failure detection.
  • The same cycle should transfer to harder instruction and initial-state shifts with less performance collapse than open-loop baselines.
  • Systems that only plan open-loop trajectories will underperform when progress stalls mid-episode relative to an SPR-style monitor-and-rewind loop.

Where Pith is reading between the lines

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

  • If milestone detection is the load-bearing signal, noisy real-world perception of intermediate states may shrink the claimed robustness gains unless the See step is hardened.
  • The same progress sequence could be used as a training-time reward or curriculum signal, not only as an inference-time recovery mechanism.
  • Rewind-to-recoverable-state may compose with classical motion-planning replaners: SPR decides when and to which milestone, a planner fills the short trajectory.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 0 minor

Summary. The submission is labeled as arXiv:2603.09292, a cs.RO paper introducing See, Plan, Rewind (SPR), a progress-aware vision-language-action framework that grounds language instructions into spatial subgoals and recovers from failures by monitoring progress against an expected milestone sequence, without extra training data or auxiliary models. The abstract reports a 5% absolute gain over MolmoAct on LIBERO and state-of-the-art out-of-distribution robustness (smallest performance drop) on LIBERO-Plus versus OpenVLA-OFT and UniVLA. The full manuscript body, however, is an unrelated communications paper: “Tensor Train Decomposition-based Channel Estimation for MIMO-AFDM Systems with Fractional Delay and Doppler” (tensor-train / Vandermonde factor recovery, embedded T-AF pilots, and a derived Ziv–Zakai bound). No SPR architecture, milestone representation, progress monitor, rewind policy, failure taxonomy, LIBERO results, or ablations appear in the provided full text.

Significance. If the abstract’s claims were supported by a matching manuscript, a data-free closed-loop recovery mechanism for VLAs that improves both in-distribution success and OOD robustness would be of clear interest to the robotics community. As submitted, those claims cannot be assessed: the body develops MIMO-AFDM channel estimation (TT-SVD factor retrieval, ESPRIT-style parameter extraction, ZZB analysis) and has no bearing on progress-aware manipulation. Significance of the advertised SPR contribution is therefore indeterminate; the materials do not constitute a reviewable robotics paper.

major comments (3)
  1. Title, paper_id (2603.09292, cs.RO), and abstract describe SPR for robotic manipulation; the full manuscript text is a complete, self-contained MIMO-AFDM channel-estimation paper (Sections I–VI, Appendices A–B, Figs. 1–11, Table I). There is no SPR method, no See–Plan–Rewind cycle, no spatial subgoal grounding, and no LIBERO/LIBERO-Plus evaluation. The central claims of the abstract are therefore unsupported by the manuscript body and cannot be verified.
  2. Abstract claim that “monitoring progress against the expected sequence” enables robust error correction “without requiring additional training data or auxiliary models” is load-bearing for the advertised contribution. The provided full text contains no definition of milestones, progress signal, stall detector, recoverable rewind state, or failure taxonomy, so this sufficiency claim is uninspectable.
  3. Reported +5% over MolmoAct on LIBERO and “smallest performance drop” / SOTA OOD robustness on LIBERO-Plus versus OpenVLA-OFT and UniVLA have no corresponding tables, ablations, statistical tests, or experimental protocol in the manuscript. These quantitative claims rest solely on the abstract and cannot be audited.

Circularity Check

0 steps flagged

No circular derivation found: SPR claims are empirical and unauditable beyond the abstract; the supplied full text is an unrelated AFDM paper whose math is self-contained.

full rationale

The target paper (SPR, 2603.09292) is represented only by its abstract: a closed-loop See–Plan–Rewind cycle that grounds language into spatial milestones and rewinds on stalled progress without extra training data or auxiliary models, with reported gains on LIBERO / LIBERO-Plus. That abstract contains no equations, no fitted constants, no uniqueness theorems, and no self-citation chain that forces the result by construction—only ordinary empirical benchmark claims. The CACHEABLE full manuscript is a mismatched, unrelated MIMO-AFDM channel-estimation paper (tensor-train / ZZB, arXiv 2603.09293). Walking that AFDM derivation chain (Theorem 1 input–output model → T-AF pilot tensor → Vandermonde TT-SVD / ESPRIT parameter recovery → closed-form ZZB vs CRB → simulations) shows standard signal-processing constructions from an explicit physical model; none of the load-bearing steps reduce to their own inputs by definition, fitted-then-predicted quantities, or load-bearing self-citation uniqueness. Because circularity requires a quotable reduction of a claimed prediction to its inputs, and neither the SPR abstract nor the supplied AFDM text exhibits such a reduction, the circularity score is 0. (Unauditability of SPR’s method due to the text mismatch is a correctness/completeness issue, not circularity.)

Axiom & Free-Parameter Ledger

0 free parameters · 3 axioms · 1 invented entities

Abstract-only review of a robotics VLA paper. Load-bearing premises are domain assumptions about progress measurement and recovery, not free parameters or invented physical entities. No fitted constants appear in the abstract.

axioms (3)
  • domain assumption Task progress can be measured via an explicit sequence of actionable spatial milestones derived from language instructions.
    Central premise of the abstract; if milestones are not reliably groundable from vision+language, See/Plan/Rewind collapses.
  • ad hoc to paper Stalled progress against the expected milestone sequence is a sufficient signal to detect failure and select a recoverable rewind state without extra training data or auxiliary models.
    Explicit abstract claim of data-free closed-loop recovery; this is the distinctive operational assumption.
  • domain assumption LIBERO and LIBERO-Plus success rates are adequate proxies for real-world manipulation robustness under distribution shift.
    All reported gains are on these benchmarks; standard but unproven transfer assumption.
invented entities (1)
  • SPR (See, Plan, Rewind) progress-aware VLA cycle no independent evidence
    purpose: Name the closed-loop controller that grounds language into spatial subgoals, plans 2D waypoints, and rewinds on stall.
    Framework name introduced by the paper; independent evidence would require released code and external replications, which are not present in the abstract.

pith-pipeline@v1.1.0-grok45 · 25931 in / 2504 out tokens · 23080 ms · 2026-07-15T00:21:45.460529+00:00 · methodology

0 comments
read the original abstract

Measurement of task progress through explicit, actionable milestones is critical for robust robotic manipulation. This progress awareness enables a model to ground its current task status, anticipate verifiable intermediate states, and detect and recover from failures when progress stalls. To embody this capability, we introduce \textbf{S}ee, \textbf{P}lan, \textbf{R}ewind (SPR), a progress-aware vision-language-action framework that dynamically grounds language instructions into a sequence of spatial subgoals. SPR operates through a continuous core cycle, Seeing the current state and upcoming milestone, Planning a trajectory towards the next 2D waypoint, and Rewinding to a recoverable state upon failure by monitoring progress against the expected sequence. This closed-loop approach enables robust error correction without requiring additional training data or auxiliary models. Extensive experiments demonstrate the framework's effectiveness, generalization and robustness: SPR outperforms the MolmoAct baseline by 5\% on the LIBERO benchmark. On the challenging LIBERO-Plus benchmark with unseen instructions and initial states, SPR achieves state-of-the-art robustness with the smallest performance drop, surpassing OpenVLA-OFT and UniVLA, demonstrating superior out-of-distribution robustness.

discussion (0)

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

Forward citations

Cited by 5 Pith papers

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

  1. Back to the Familiar Future: Failure Recovery for VLA Policies via Pre-Imagined Milestone Selection

    cs.RO 2026-06 unverdicted novelty 7.0

    B2FF pre-generates a milestone bank of familiar future states from the clean initial observation and uses a recoverability-aware selector to guide VLA policies back from deviations, raising average success rate from 5...

  2. RoboProcessBench: Benchmarking Process-Aware Understanding in Vision-Language Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    RoboProcessBench is a new benchmark decomposing process-aware understanding into static monitoring and dynamic reasoning across 12 question families, with evaluations showing VLM limitations but post-training gains on...

  3. RoboProcessBench: Benchmarking Process-Aware Understanding in Vision-Language Robotic Manipulation

    cs.RO 2026-06 conditional novelty 6.0

    A 12-family, ~58k-question benchmark reveals that VLMs are weak at judging robotic manipulation progress and temporal order, and that fine-tuning on it improves local state, motion, and primitive-aware cues.

  4. Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

    cs.RO 2026-05 unverdicted novelty 6.0

    Hide-and-Seek uses contrastive objectives on trajectories to localize failure signals in VLA models from trajectory-level supervision alone.

  5. Action with Visual Primitives

    cs.RO 2026-05 unverdicted novelty 6.0

    AVP architecture has VLM emit visual-primitive tokens to condition flow-matching action expert, yielding 27.61% higher success rate than pi_0.5 on real-robot pick-and-place tasks.

Reference graph

Works this paper leans on

39 extracted references · 1 linked inside Pith · cited by 4 Pith papers

  1. [1]

    Performance degradation of ofdm systems due to doppler spreading,

    T. Wang, J. Proakis, E. Masry, and J. Zeidler, “Performance degradation of ofdm systems due to doppler spreading,”IEEE Trans. Wireless Commun., vol. 5, no. 6, pp. 1422–1432, Jun. 2006

    2006

  2. [2]

    Chunk-based resource allocation in OFDMA systems - part i: chunk allocation,

    H. Zhu and J. Wang, “Chunk-based resource allocation in OFDMA systems - part i: chunk allocation,”IEEE Trans. Commun., vol. 57, no. 9, pp. 2734–2744, Sep. 2009

    2009

  3. [3]

    Chunk-based resource allocation in OFDMA systemspart ii: Joint chunk, power and bit allocation,

    ——, “Chunk-based resource allocation in OFDMA systemspart ii: Joint chunk, power and bit allocation,”IEEE Trans. Commun., vol. 60, no. 2, pp. 499–509, Feb. 2012

    2012

  4. [4]

    Orthogonal time frequency space modu- lation,

    R. Hadani, S. Rakib, M. Tsatsanis, A. Monk, A. J. Goldsmith, A. F. Molisch, and R. Calderbank, “Orthogonal time frequency space modu- lation,” in2017 IEEE Wireless Communications and Networking Con- ference (WCNC), 2017, pp. 1–6

    2017

  5. [5]

    OTFS: A new generation of modulation addressing the challenges of 5G,

    R. Hadani and A. Monk, “OTFS: A new generation of modulation addressing the challenges of 5G,” 2018. [Online]. Available: https: //arxiv.org/abs/1802.02623

    Pith/arXiv arXiv 2018

  6. [6]

    New delay doppler communication paradigm in 6g era: A survey of orthogonal time frequency space (OTFS),

    W. Yuan, S. Li, Z. Wei, Y . Cui, J. Jiang, H. Zhang, and P. Fan, “New delay doppler communication paradigm in 6g era: A survey of orthogonal time frequency space (OTFS),”China Commun., vol. 20, no. 6, pp. 1–25, Jun. 2023

    2023

  7. [7]

    A survey on orthogonal time frequency space modulation,

    M. Aldababsa, S. zyurt, G. K. Kurt, and O. Kucur, “A survey on orthogonal time frequency space modulation,”IEEE Open J. Commun. Soc., vol. 5, pp. 4483–4518, Jul. 2024

    2024

  8. [8]

    Affine frequency division multiplexing for next generation wireless communications,

    A. Bemani, N. Ksairi, and M. Kountouris, “Affine frequency division multiplexing for next generation wireless communications,”IEEE Trans. Wireless Commun., vol. 22, no. 11, pp. 8214–8229, Nov. 2023

    2023

  9. [9]

    AFDM: A full diversity next generation waveform for high mobility communications,

    ——, “AFDM: A full diversity next generation waveform for high mobility communications,” in2021 IEEE International Conference on Communications Workshops (ICC Workshops), 2021, pp. 1–6

    2021

  10. [10]

    From OFDM to AFDM: Enabling adaptive integrated sensing and communication in high-mobility scenarios,

    H. Yin, Y . Tang, J. Xiong, F. Liu, Y . Ni, Q. Luo, R. Bomfin, M. Chafii, M. Kountouris, and C. Masouros, “From OFDM to AFDM: Enabling adaptive integrated sensing and communication in high-mobility scenarios,” 2025. [Online]. Available: https://arxiv.org/abs/2510.27192

    arXiv 2025

  11. [11]

    AFDM-enabled integrated sensing and communication: Theoretical framework and pilot design,

    F. Zhang, Z. Wang, T. Mao, T. Jiao, Y . Zhuo, M. Wen, W. Xiang, S. Chen, and G. K. Karagiannidis, “AFDM-enabled integrated sensing and communication: Theoretical framework and pilot design,”IEEE J. Sel. Areas Commun., vol. 44, pp. 310–324, Sep. 2025

    2025

  12. [12]

    Diagonally reconstructed channel estimation for MIMO-AFDM with inter-doppler interference in doubly selective channels,

    H. Yin, X. Wei, Y . Tang, and K. Yang, “Diagonally reconstructed channel estimation for MIMO-AFDM with inter-doppler interference in doubly selective channels,”IEEE Trans. Wireless Commun., vol. 23, no. 10, pp. 14 066–14 079, Oct. 2024

    2024

  13. [13]

    Pilot aided channel estimation for AFDM in doubly dispersive channels,

    H. Yin and Y . Tang, “Pilot aided channel estimation for AFDM in doubly dispersive channels,” in2022 IEEE/CIC International Conference on Communications in China (ICCC), 2022, pp. 308–313

    2022

  14. [14]

    Sparse channel estimation and data detection for AFDM with superimposed pilot scheme,

    Z. Cao, M. Wen, and Y . Huang, “Sparse channel estimation and data detection for AFDM with superimposed pilot scheme,” in2024 IEEE 24th International Conference on Communication Technology (ICCT), 2024, pp. 1540–1544

    2024

  15. [15]

    An AFDM off-grid channel estimation based on sparse bayesian learning,

    F. Yang, S. Luo, L. Wu, D. Song, R. Lin, and S. Xie, “An AFDM off-grid channel estimation based on sparse bayesian learning,” in2024 IEEE 24th International Conference on Communication Technology (ICCT), 2024, pp. 1565–1569

    2024

  16. [16]

    Low-complexity channel estimation for internet of vehicles AFDM communications with sparse bayesian learning,

    X. Li, H. Wang, Y . Ge, X. Shen, M. Wen, S. Zhang, and Y . L. Guan, “Low-complexity channel estimation for internet of vehicles AFDM communications with sparse bayesian learning,”IEEE Internet Things J., vol. 13, no. 5, pp. 9795–9810, Mar. 2026

    2026

  17. [17]

    Optimum receiver design for wireless broad-band systems using OFDM. i,

    M. Speth, S. Fechtel, G. Fock, and H. Meyr, “Optimum receiver design for wireless broad-band systems using OFDM. i,”IEEE Trans. Commun., vol. 47, no. 11, pp. 1668–1677, Nov. 1999

    1999

  18. [18]

    Integrated sensing and communications with affine frequency division multiplexing,

    A. Bemani, N. Ksairi, and M. Kountouris, “Integrated sensing and communications with affine frequency division multiplexing,”IEEE Wireless Commun. Lett., vol. 13, no. 5, pp. 1255–1259, May 2024

    2024

  19. [19]

    Low complexity AFDM- based isac utilizing fractional delay-doppler feature extraction,

    S. Zhu, L. Mei, Z. Li, Z. Du, and Y . Song, “Low complexity AFDM- based isac utilizing fractional delay-doppler feature extraction,”IEEE Wireless Commun. Lett., vol. 15, pp. 335–339, Oct. 2026

    2026

  20. [20]

    Rectified als algorithm for multidimensional harmonic retrieval,

    R. Boyer and P. Comon, “Rectified als algorithm for multidimensional harmonic retrieval,” in2016 IEEE Sensor Array and Multichannel Signal Processing Workshop (SAM), 2016, pp. 1–5

    2016

  21. [21]

    Time-varying phase noise estimation, channel estimation, and data detection in RIS-assisted mIMO systems via tensor analysis,

    J. Du, Y . Cheng, L. Jin, and F. Gao, “Time-varying phase noise estimation, channel estimation, and data detection in RIS-assisted mIMO systems via tensor analysis,”IEEE Trans. Signal Process., vol. 71, pp. 3426–3441, Sep. 2023

    2023

  22. [22]

    Tensor-based channel estimation for massive MIMO-OFDM systems,

    D. C. Ara ´ujo, A. L. F. de Almeida, J. P. C. L. Da Costa, and R. T. de Sousa, “Tensor-based channel estimation for massive MIMO-OFDM systems,”IEEE Access, vol. 7, pp. 42 133–42 147, Mar. 2019

    2019

  23. [23]

    Local convergence of the alternating least squares algorithm for canonical tensor approximation,

    A. Uschmajew, “Local convergence of the alternating least squares algorithm for canonical tensor approximation,”SIAM J. Matrix Anal. Appl., vol. 33, no. 2, pp. 639–652, 2012. [Online]. Available: https://doi.org/10.1137/110843587

    doi:10.1137/110843587 2012

  24. [24]

    Tensor-train decomposition,

    I. V . Oseledets, “Tensor-train decomposition,”SIAM J. Sci. Comput., vol. 33, no. 5, pp. 2295–2317, 2011. [Online]. Available: https: //doi.org/10.1137/090752286

    doi:10.1137/090752286 2011

  25. [25]

    Multidimensional harmonic retrieval based on vandermonde tensor train,

    Y . Zniyed, R. Boyer, A. L. de Almeida, and G. Favier, “Multidimensional harmonic retrieval based on vandermonde tensor train,”Signal Process., vol. 163, pp. 75–86, 2019

    2019

  26. [26]

    An esprit-based supervised channel estimation method using tensor train decomposition for mmwave 3-d MIMO-OFDM systems,

    X. Gong, W. Chen, L. Sun, J. Chen, and B. Ai, “An esprit-based supervised channel estimation method using tensor train decomposition for mmwave 3-d MIMO-OFDM systems,”IEEE Trans. Signal Process., vol. 71, pp. 555–570, 2023

    2023

  27. [27]

    Low- rank tensor decomposition-aided channel estimation for millimeter wave mimo-ofdm systems,

    Z. Zhou, J. Fang, L. Yang, H. Li, Z. Chen, and R. S. Blum, “Low- rank tensor decomposition-aided channel estimation for millimeter wave mimo-ofdm systems,”IEEE J. Selected Areas Commun., vol. 35, no. 7, pp. 1524–1538, Jul. 2017

    2017

  28. [28]

    Newtonized orthogonal matching pursuit for high-resolution target detection in sparse ofdm isac systems,

    S. N. H. Shah, S. Semper, A. U. Khan, C. Schneider, and J. Robert, “Newtonized orthogonal matching pursuit for high-resolution target detection in sparse ofdm isac systems,”IEEE Trans. V eh. Technol., vol. 74, no. 10, pp. 16 137–16 151, Oct. 2025

    2025

  29. [29]

    Some lower bounds on signal parameter estima- tion,

    J. Ziv and M. Zakai, “Some lower bounds on signal parameter estima- tion,”IEEE Trans. Inf. Theory, vol. 15, no. 3, pp. 386–391, May 1969

    1969

  30. [30]

    S. M. Kay,Fundamentals of statistical signal processing: estimation theory. Prentice-Hall, Inc., 1993

    1993

  31. [31]

    H. L. Van Trees,Detection, estimation, and modulation theory, part I: detection, estimation, and linear modulation theory. John Wiley & Sons, 2004

    2004

  32. [32]

    Extended ziv- zakai lower bound for vector parameter estimation,

    K. Bell, Y . Steinberg, Y . Ephraim, and H. Van Trees, “Extended ziv- zakai lower bound for vector parameter estimation,”IEEE Trans. Inf. Theory, vol. 43, no. 2, pp. 624–637, Mar. 1997

    1997

  33. [33]

    Ziv-zakai bound for 2d-DOAs estimation,

    Z. Zhang, Z. Shi, C. Shao, J. Chen, M. S. Greco, and F. Gini, “Ziv-zakai bound for 2d-DOAs estimation,”IEEE Trans. Signal Process., vol. 72, pp. 2483–2497, Mar. 2024

    2024

  34. [34]

    H. L. Van Trees,Detection, estimation, and modulation theory, part III: radar-sonar signal processing and Gaussian signals in noise. John Wiley & Sons, 2001

    2001

  35. [35]

    Ziv-zakai bound for DOAs estimation,

    Z. Zhang, Z. Shi, and Y . Gu, “Ziv-zakai bound for DOAs estimation,” IEEE Trans. Signal Process., vol. 71, pp. 136–149, Dec. 2023

    2023

  36. [36]

    Embedded pilot-aided channel estimation for OTFS in delaydoppler channels,

    P. Raviteja, K. T. Phan, and Y . Hong, “Embedded pilot-aided channel estimation for OTFS in delaydoppler channels,”IEEE Trans. V eh. Tech., vol. 68, no. 5, pp. 4906–4917, May 2019

    2019

  37. [37]

    On the diversity of uncoded OTFS modulation in doubly-dispersive channels,

    G. D. Surabhi, R. M. Augustine, and A. Chockalingam, “On the diversity of uncoded OTFS modulation in doubly-dispersive channels,”IEEE Trans. Wireless Commun., vol. 18, no. 6, pp. 3049–3063, Apr. 2019

    2019

  38. [38]

    Interference can- cellation and iterative detection for orthogonal time frequency space modulation,

    P. Raviteja, K. T. Phan, Y . Hong, and E. Viterbo, “Interference can- cellation and iterative detection for orthogonal time frequency space modulation,”IEEE Trans. Wireless Commun., vol. 17, no. 10, pp. 6501– 6515, Oct. 2018

    2018

  39. [39]

    Ambiguity function analysis of AFDM signals for integrated sensing and communications,

    H. Yin, Y . Tang, Y . Ni, Z. Wang, G. Chen, J. Xiong, K. Yang, M. Kountouris, Y . Liang Guan, and Y . Zeng, “Ambiguity function analysis of AFDM signals for integrated sensing and communications,” IEEE J. Sel. Areas Commun., vol. 44, pp. 196–211, Sep. 2026

    2026