Pith. sign in

REVIEW 2 major objections 4 minor 39 references

Tail-Calibrated Soft-Output GRAND for Finite-Memory Noise-Effect Posteriors

T0 review · 2 major / 4 minor · reviewed 2026-08-15 · deepseek-v4-flash

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

Pith's one-line read This paper shows that the missing-list probability in GRAND decoding can be estimated as the ambient posterior tail mass times a random-codebook occupancy factor, and that under exact enumeration the first listed codeword is…

desk verdict Solid, honest extension of SOGRAND to finite-memory posteriors, with clean bounds for a random-codebook estimator; the main gap is that the headline RLC experiments use that estimator as an approximation with no bias bound. read the letter →

arxiv 2608.04068 v1 pith:OKXNR44E submitted 2026-08-04 cs.IT eess.SPmath.IT

Behrooz Razeghi This is my paper · ORCID
classification cs.ITeess.SPmath.IT MSC 94B3594A15
keywords guessingrandomadditivenoisedecodingsoft-outputchannelswithmemoryposteriortailprobabilitycodingAPPlog-likelihoodratiofinite-staterecursionmissing-list
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper introduces a decoding rule that gives soft outputs—reliability estimates and log-likelihood ratios—for GRAND-style noise-guessing decoders when the noise process has memory. The key move is to separate the ambient posterior over all possible noise effects, modeled by a finite-memory energy, from the codebook-restricted denominator that enters the a posteriori probability (APP). After q queries, the unqueried part of that denominator is estimated as the ambient posterior tail mass times the occupancy probability of the remaining random codebook, p_q=(M-|L_q|)/(2^n-q). Under exact enumeration the first listed codeword is ML, and the same tail decomposition yields blockwise and bitwise APP estimates. This matters because it extends soft-output GRAND from memoryless channels to correlated channels without interleaving and replaces a fixed query limit with a principled, tail-calibrated stopping rule.

What carries the argument

The load-bearing object is the finite-memory posterior energy E_{r^n}(z^n) = sum_i alpha_i(z_i;r^n) + sum_i beta_i(z_{i-1},z_i;r^n), normalized by a partition function Z(r^n) computed by a finite-state forward recursion; a forward-backward recursion gives full-space bit masses, and subtracting queried-prefix contributions yields ambient tail bit masses. The random-codebook correction p_q=(M-|L_q|)/(2^n-q) turns the ambient tail mass T_q into the estimator bU_q=p_qT_q for the unqueried denominator contribution, and the same tail decomposition yields bitwise APP LLRs. The burst-atom decomposition of binary patterns into separated runs provides an additive ordering cost, so candidate noise effects can be enumerated in nondecreasing energy by a finite-state search.

What would settle it

Take a small random linear code and a fixed finite-memory posterior, enumerate all unqueried noise effects to compute the exact U_q for many transcripts, and compare the average with p_qT_q; a systematic deviation beyond the variance bound of Theorem 5.2 would show the random-codebook identity does not transfer to the linear-code setting.

Watch

Extended reading notes

Core claim

With exact nondecreasing-energy enumeration and no abandonment, the first codeword returned by the membership-query sequence is maximum-likelihood for the likelihood model that defines the posterior energy. Separately, under a fixed-size random-codebook ensemble, after q queries the conditional expectation of the unqueried codebook-restricted denominator contribution U_q equals p_q T_q, where T_q is the ambient posterior mass of the unqueried noise-effect tail and p_q is the fraction of remaining codewords among remaining candidate words; variance and concentration bounds follow from the effective support size of that tail. The decoder therefore stops when the plug-in missing-list estimate bU_q/(S_q+bU_q) falls below a tolerance, and it uses the same posterior-tail decomposition to form blockwise APP estimates and bitwise APP log-likelihood ratios under a fixed 0-over-1 convention. In the tested Gauss-Markov and binary Markov random-linear-code settings, the matched finite-memory ordering gives lower BLER point estimates and fewer average membership queries than the memoryless and approximate-independence orderings tested, at the price of more enumeration work in the software implementation.

Load-bearing premise

The load-bearing premise is that unqueried candidate words behave like a uniformly random subset; the paper's main experiments use random linear codes, where that premise is known to fail and p_qT_q is used only as an approximation.

Editorial extensions

If this is right

  • With exact energy-ordered enumeration and no abandonment, decoding is ML for the model that defines the posterior energy, so correlated-channel ordering need not sacrifice finite-block optimality.
  • A decoder that stops when bU_q/(S_q+bU_q) is at most eta keeps the true missing-list probability within the perturbation bound (1+epsilon)eta/(1+epsilon*eta) whenever the estimator's relative error is at most epsilon.
  • The same posterior-tail decomposition supplies blockwise APP estimates and bitwise APP LLRs, enabling soft-output exchange in iterative decoding of product and GLDPC codes.
  • If the abandoned ambient posterior tail mass is at most epsilon_n, the abandoned decoder's block-error probability exceeds full GRAND's by at most epsilon_n, so rates below the noise entropy rate remain achievable.
  • In the reported Gauss-Markov and binary Markov experiments, the matched finite-memory ordering lowers BLER point estimates and average membership-query counts relative to the tested ORBGRAND-AI and block-product orderings.

Reading between the lines

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

  • Because the conditional-mean identity is proved for uniformly random codebooks, applying p_qT_q to random linear codes is an unproven transfer; a code-aware correction that exploits subspace structure could improve calibration further.
  • The enumeration bottleneck at low SNR suggests the practical value of the method will hinge on faster exact or approximate enumerators for finite-memory energies; if such enumerators appear, the same posterior-tail machinery should scale to longer blocklengths and iterative loops.
  • The bitwise tail-marginal decomposition could also support list-quality metrics beyond missing-list probability, such as per-position reliability for erasure or retransmission decisions.
  • A direct small-n experiment comparing p_qT_q with the true conditional mean under uniform-sampling and random-linear ensembles would quantify when linear-code membership coupling matters.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

2 major / 4 minor

Summary. This paper introduces Tail-Calibrated SOGRAND, a GRAND variant for binary additive channels with finite-memory noise-effect posteriors. The decoder orders candidate noise effects by a finite-memory posterior energy, queries codebook membership, computes the partition function and ambient tail masses by finite-state forward/backward recursions, and estimates the unqueried codebook-restricted denominator by p_q T_q, where p_q=(M-|L_q|)/(2^n-q) is the remaining random-codebook occupancy probability and T_q is the ambient tail mass. This estimate feeds a missing-list stopping rule, blockwise APPs, and bitwise APP LLRs. The paper proves that exact enumeration yields an ML codeword (Theorem 5.1), that p_qT_q is conditionally unbiased with variance and concentration bounds under the fixed-size random-codebook model (Theorem 5.2), a perturbation bound for the plug-in missing-list estimate (Theorem 5.4), and an ambient-tail abandonment error bound (Theorem 5.5). Numerical evaluations compare the exact finite-memory ordering against ORBGRAND, ORBGRAND-AI, and block-product ablations in Gauss-Markov and hard-decision Markov/Gilbert-Elliott noise, and validate missing-list calibration in a fixed-size random-codebook experiment and in an RLC experiment.

Significance. Taken at face value, the random-codebook analysis is the paper's cleanest contribution: Theorem 5.2 gives an exact conditional-mean identity and computable variance and concentration bounds for the missing-list denominator estimator, and the finite-state computation of posterior weights and tail masses is useful for GRAND with correlated noise. The ML property and ambient-tail abandonment bound are elementary but correctly stated. The numerical section reports a consistent ordering in BLER and membership-query counts favoring the matched finite-memory enumerator, and the fixed-size random-codebook calibration (Table I) supports the occupancy model. The main limitation is that the supporting theory for tail calibration is proved for uniformly random M-subset codebooks, while the headline experiments use random linear codes; the paper discloses this in Remark 5.3 but does not quantify the resulting bias. This does not affect the first-hit BLER and membership-query comparisons, but it does leave the tail-stopping and soft-output claims in the demonstrated setting without a theoretical guarantee.

major comments (2)
  1. [Remark 5.3, Sections 4.3-4.5 and 6.2] The advertised missing-list calibration, tail-stopping rule, and bitwise APP estimates are used in the main random-linear-code experiments, but Theorem 5.2's conditional-mean, variance, and concentration guarantees are proved only for the fixed-size uniformly random M-subset codebook. Remark 5.3 concedes that conditional uniformity does not hold for a random linear code and that p_q T_q is used only as an approximation. Because the linear-subspace structure forces every word in span(L_q) to be a codeword with probability one while p_q assigns a much smaller occupancy probability, the bias of bU_q in the demonstrated setting is unquantified. The first-hit BLER and membership-query comparisons do not depend on bU_q, but the tail-stopping and soft-output claims (Sections 4.4-4.5, Fig. 5, Fig. 6, Table VI) do. Please either extend the analysis to RLCs, even under a restricted ensemble or with a worst-case bias bound, or explicitly rescope the theoretical claims so that all RLC tail-calibration statements are labeled as heuristic and are accompanied by a quantified empirical bias characterization over the reported operating points.
  2. [Theorem 5.2, Theorem 5.4, Algorithm 1, Section 6.4.2] Theorems 5.2 and 5.4 are stated for a fixed query index q and a fixed transcript event E_q(ell_q), whereas Algorithm 1 and the calibration experiments evaluate bPmiss at the random stopping time q_eta (Algorithm 1, line 18; Section 6.4.2). The proof of Theorem 5.2 in fact goes through conditional on any event that fixes the intersection C^n cap B_q, so the stopping-time selection is not fatal; however, the manuscript should state this explicitly. As written, a reader cannot tell whether the advertised tail-stopping guarantee is meant to cover the random stopping time or only fixed query indices. Please add a remark clarifying that the unbiasedness and relative-error bound hold conditionally on the transcript, including the event {q_eta = q}, and discuss the nonlinearity of f(u) = u/(S_q + u) when converting the unbiasedness of bU_q into a calibration statement for bPmiss at the stopping time.
minor comments (4)
  1. [Section 4.2.1, Eq. (23)] The expression 'z^n = L^m_{j=1} 1_{B_j}' should use a sum or disjoint-union symbol rather than 'L', because the intervals B_j are disjoint and the sum of indicators is the standard notation for the union of runs.
  2. [Table IV and Fig. 3] The zero BLER entries are plotted at 1/(2N) but printed as 0 in Table IV; please add the Wilson upper endpoint in the table or in a footnote so that the finite-sample uncertainty of those zero-error observations is visible.
  3. [Theorem 5.5] The theorem statement uses P_full^e(n) and P_ab^e(n) without defining them; please define them as the block error probabilities of full and abandoned GRAND, respectively, in the statement itself.
  4. [Section 6.2, software-work proxy] The software-work proxy C_sw assigns unit weights to membership tests, metric evaluations, priority-queue removals, and preprocessing operations; please state whether the qualitative conclusions from Fig. 2 (bottom right) are robust to reasonable alternative weightings.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central estimator is derived from the random-codebook occupancy model, and the random-linear-code caveat is an acknowledged applicability gap, not a circular reduction.

full rationale

The derivation chain is self-contained. Theorem 5.1 is a direct consequence of defining the ambient posterior weights proportional to the likelihood: ordering noise effects by nondecreasing energy is equivalent to nonincreasing likelihood, so the first listed codeword maximizes the codebook posterior under that model; this is a proof, not a fitted prediction. Theorem 5.2 derives E_C[U_q | E_q(ell_q)] = p_q T_q from the fixed-size uniform random-codebook model by linearity of expectation and subset-inclusion combinatorics; no parameter is fitted to decoding outcomes. The plug-in stopping rule is exactly this derived estimator, and Theorem 5.4 bounds the effect of relative error, so the stopping rule is not an independently fitted prediction. The only self-citation, LP-GRAND [28], is referenced in related work for exact likelihood-ordered enumeration, but the present paper also cites standard methods (best-first/K-shortest-path [34] and BCJR-style recursions [35]) and the main estimator theorems do not depend on [28]. Remark 5.3 explicitly concedes that p_q T_q is not the exact conditional mean for random linear codes; that limits the reach of the unbiasedness and concentration guarantees into the RLC experiments, but it is an honest, unverified approximation rather than a circular step. No equation reduces to its own input, and no fitted parameter is renamed as a prediction.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The paper introduces no new physical or structural entities. The 'ambient posterior tail mass', 'unqueried codebook-restricted denominator', and occupancy estimator p_q T_q are defined quantities derived from the assumed energy model, not posited new objects. The free parameters listed are either channel-model inputs, a user-selected threshold, or an empirical fit in a secondary experiment.

free parameters (4)
  • Posterior energy potentials alpha_i and beta_i = channel-statistics-derived (true rho=0.5 in Gauss-Markov experiments)
    These functions define the finite-memory posterior model in Eq. (6). In the experiments they are taken from the assumed channel statistics rather than fitted to decoding outcomes. All tail-mass and ML claims are conditional on this energy model, and Section 6.3.4 shows mismatch degrades BLER.
  • First-order Markov transition probabilities for Gilbert-Elliott approximation = estimated from 2e5 simulated noise samples
    Used in the Gilbert-Elliott experiments (Section 6.4.6). The decoder approximates the hidden-state process by a first-order Markov chain whose parameters are estimated from separate simulated noise. This is an empirical fit to a simulation, not part of the central estimator.
  • Memory order m = m=1 in all experiments
    The finite-memory energy representation is assumed with a fixed order; the paper uses first-order models in simulations. The recursions generalize to memory m with 2^m states, but no higher-order experiments are reported.
  • Stopping tolerance eta = 1e-4 or 1e-5 in tail-stop variants
    User-selected threshold in the stopping rule (Section 4.4). It is an operating parameter, not fitted to data; the perturbation bound in Theorem 5.4 holds for any eta.
assumptions (5)
  • domain assumption The ambient full-space posterior pi_rn(z^n) is exactly of the finite-memory energy form (6) with known potentials.
    Assumed in Section 3.1. The theoretical guarantees for tail masses and the ML property are relative to this model. The paper notes in Section 6.3.4 that mismatch degrades performance.
  • domain assumption The noise effect Z^n is independent of the transmitted codeword X^n and the channel is binary additive with hard-decision map h.
    Standard GRAND condition used in Section 3 and Eq. (1)-(2). Needed for the equivalence between codeword likelihood and noise-effect likelihood.
  • domain assumption For the random-codebook theorems, C_n is uniformly distributed over all M-subsets of {0,1}^n.
    Assumed in Section 4.1 and Theorem 5.2. Holds for the fixed-size random-codebook validation (Section 6.1) but not for the random-linear-code experiments, as conceded in Remark 5.3.
  • domain assumption The enumerator returns distinct noise effects and, for exact claims, returns them in nondecreasing energy order.
    Required by Algorithm 1 and Theorem 5.1. If the enumerator repeats patterns, the tail-mass subtraction and missing-list estimates break.
  • standard math Standard probability and recursion facts: linearity of expectation, negative covariance of inclusion indicators under sampling without replacement, Chebyshev's inequality, and correctness of BCJR-style forward-backward recursions.
    Used in the proofs of Theorems 5.2 and 5.4 and in Sections 4.3 and 4.5. These are uncontroversial background results.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Tail-Calibrated Soft-Output GRAND for Finite-Memory Noise-Effect Posteriors." pith.science (2026). https://pith.science/paper/OKXNR44E

@misc{pith2026260804068,
  author       = {Pith},
  title        = {Pith review of: Tail-Calibrated Soft-Output GRAND for Finite-Memory Noise-Effect Posteriors},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/OKXNR44E}},
  note         = {Machine review of arXiv:2608.04068}
}
abstract

In guessing random additive noise decoding (GRAND), memory in the hard-decision noise effect changes the likelihood order of candidate noise effects. In soft-output decoding, the same memory also affects the finite-block quantity determining the missing-list probability: the codebook-restricted posterior mass outside the current list. Existing correlation-aware GRAND methods exploit local dependence without interleaving, but their stopping and soft-output rules are not derived from finite-memory posterior tails. Soft-output GRAND (SOGRAND) derives random-codebook a posteriori probability (APP) estimates for GRAND lists, but does not provide finite-memory algorithms for posterior weights, partition functions, tail masses, or bitwise tail marginals for correlated noise-effect posteriors. We introduce Tail-Calibrated SOGRAND for binary additive channels whose ambient hard-decision noise-effect posterior, conditioned on received soft information, is represented by a finite-memory energy. The decoder enumerates candidate noise effects in nondecreasing posterior energy, queries codebook membership as in GRAND, computes posterior weights and tail masses by finite-state recursions, and estimates the unqueried codebook-restricted denominator as $p_qT_q$, where $T_q$ is the ambient posterior tail mass and $p_q$ is the remaining random-codebook occupancy probability. With exact enumeration and no abandonment, the first listed codeword is ML under the likelihood model defining the posterior energy. We also prove an ambient posterior-tail abandonment bound and, separately, conditional unbiasedness, variance, and concentration bounds for the random-codebook missing-list estimator. The same posterior-tail decomposition gives blockwise APP estimates, missing-list probabilities, and bitwise APP log-likelihood ratios (LLRs) for finite-memory noise-effect posteriors.

Figures

Figures reproduced from arXiv: 2608.04068 by the authors.

Figure 1
Figure 1. Reliability diagrams for the fixed-size random-codebook [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗
Figure 2
Figure 2. Gauss–Markov RLC experiment with [n, k] = [128, 116], ρ = 0.5, and qmax = 20000. The tolerance η = 10−4 is active only for the Exact-Markov-tailStop variant and is inactive for first-hit BLER decoding. Top row: BLER, found-codeword rate, and average number of codebook membership queries. Bottom row: 99th-percentile membership-query count, mean estimated missing-list probability, and software-work proxy Csw. The mean… view at source ↗
Figure 3
Figure 3. Correlation sweep at Eb/N0 = 2 dB for the [128, 116] RLC, with 5000 frames per point. Left: BLER. Middle: average membership-query count. Right: software-work proxy Csw. Zero observed BLER values are displayed at 1/(2N) on logarithmic axes, where N is the number of simulated frames; the corresponding CSV entries remain equal to zero. values are approximately 3.26×103 , 4.52×103 , and 4.35×103 . Thus, in this softwar… view at source ↗
Figures from the paper (7 more)
Figure 4
Figure 4. Figure 4: Model-mismatch sweep for the Exact-Markov decoder. [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Missing-list reliability diagrams. Left: Exact-Markov list-output decoder. Middle: Exact-Markov tail-stopping variant. [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Missing-list stopping tolerance sweep for the Exact-Markov-tailStop decoder with [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]
Figure 7
Figure 7. Figure 7: Membership-query-limit sweep at Eb/N0 = 3 dB for the [128, 116] RLC with ρ = 0.5. For Exact-Markov-tailStop, η = 10−5 . Left: BLER. Middle: average membership-query count. Right: found-codeword rate. dependent channel variance, codebook density, and query-limit truncat…
Figure 8
Figure 8. Figure 8: Fixed-query-limit rate sweep for length- [PITH_FULL_IMAGE:figures/full_fig_p017_8.png]
Figure 9
Figure 9. Figure 9: Hard-decision binary Markov noise experiment for the [PITH_FULL_IMAGE:figures/full_fig_p017_9.png]
Figure 10
Figure 10. Figure 10: Gilbert–Elliott hard-decision experiment for the [PITH_FULL_IMAGE:figures/full_fig_p017_10.png]

Review history (2 revisions) →

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

39 extracted references · 31 canonical work pages

  1. [1]

    Low-density parity-check codes,

    R. Gallager, “Low-density parity-check codes,”IRE Transactions on information theory, vol. 8, no. 1, pp. 21–28, 1962

    1962

  2. [2]

    Factor graphs and the sum-product algorithm,

    F. R. Kschischang, B. J. Frey, and H.-A. Loeliger, “Factor graphs and the sum-product algorithm,”IEEE Transactions on Information Theory, vol. 47, no. 2, pp. 498–519, 2001

    2001

  3. [3]

    Near shannon limit error- correcting coding and decoding: Turbo-codes. 1,

    C. Berrou, A. Glavieux, and P. Thitimajshima, “Near shannon limit error- correcting coding and decoding: Turbo-codes. 1,” inIEEE International Conference on Communications, vol. 2. IEEE, 1993, pp. 1064–1070

    work page 1993

  4. [4]

    Channel polarization: A method for constructing capacity- achieving codes for symmetric binary-input memoryless channels,

    E. Arikan, “Channel polarization: A method for constructing capacity- achieving codes for symmetric binary-input memoryless channels,”IEEE Transactions on information Theory, vol. 55, no. 7, pp. 3051–3073, 2009

    2009

  5. [5]

    Crc-aided decoding of polar codes,

    K. Niu and K. Chen, “Crc-aided decoding of polar codes,”IEEE communications letters, vol. 16, no. 10, pp. 1668–1671, 2012

    work page 2012

  6. [6]

    List decoding of polar codes,

    I. Tal and A. Vardy, “List decoding of polar codes,”IEEE transactions on information theory, vol. 61, no. 5, pp. 2213–2226, 2015

    2015

  7. [7]

    Ordered reliability bits guessing random additive noise decoding,

    K. R. Duffy, W. An, and M. Médard, “Ordered reliability bits guessing random additive noise decoding,”IEEE Transactions on Signal Process- ing, vol. 70, pp. 4528–4542, 2022

    work page 2022

  8. [8]

    Keep the bursts and ditch the interleavers,

    W. An, M. Médard, and K. R. Duffy, “Keep the bursts and ditch the interleavers,”IEEE Transactions on Communications, vol. 70, 2022

    work page 2022

Show all 39 references
  1. [9]

    Using channel correlation to improve decoding-ORBGRAND-AI,

    K. R. Duffy, M. Grundei, and M. Médard, “Using channel correlation to improve decoding-ORBGRAND-AI,” inGLOBECOM 2023-2023 IEEE Global Communications Conference. IEEE, 2023, pp. 3585–3590

    2023

  2. [10]

    Capacity-achieving guessing random additive noise decoding,

    K. R. Duffy, J. Li, and M. Médard, “Capacity-achieving guessing random additive noise decoding,”IEEE Transactions on Information Theory, vol. 65, no. 7, pp. 4023–4040, 2019

    2019

  3. [11]

    Soft maximum likelihood decoding using GRAND,

    A. Solomon, K. R. Duffy, and M. Médard, “Soft maximum likelihood decoding using GRAND,” inICC 2020-2020 IEEE International Con- ference on Communications (ICC). IEEE, 2020, pp. 1–6

    2020

  4. [12]

    Multi-code multi-rate universal maximum likelihood decoder using GRAND,

    A. Riaz, V . Bansal, A. Solomon, W. An, Q. Liu, K. Galligan, K. R. Duffy, M. Medard, and R. T. Yazicigil, “Multi-code multi-rate universal maximum likelihood decoder using GRAND,” inESSCIRC 2021-IEEE 47th European Solid State Circuits Conference (ESSCIRC). IEEE, 2021, pp. 239–246

    2021

  5. [13]

    A sub-0.8-pj/bit universal soft-detection decoder using ORBGRAND,

    A. Riaz, A. Yasar, F. Ercan, W. An, J. Ngo, K. Galligan, M. Médard, K. R. Duffy, and R. T. Yazicigil, “A sub-0.8-pj/bit universal soft-detection decoder using ORBGRAND,”IEEE Journal of Solid-State Circuits, vol. 60, no. 7, pp. 2645–2659, 2024

    2024

  6. [14]

    Soft-output (so) grand and iterative decoding to outperform LDPC codes,

    P. Yuan, M. Médard, K. Galligan, and K. R. Duffy, “Soft-output (so) grand and iterative decoding to outperform LDPC codes,”IEEE Transactions on Wireless Communications, vol. 24, no. 4, pp. 3386–3399, 2025

    2025

  7. [15]

    Near-optimum decoding of product codes: Block turbo codes,

    R. M. Pyndiah, “Near-optimum decoding of product codes: Block turbo codes,”IEEE Transactions on Communications, vol. 46, no. 8, pp. 1003– 1010, Aug. 1998

    1998

  8. [16]

    Block turbo decoding with ORBGRAND,

    K. Galligan, M. Médard, and K. R. Duffy, “Block turbo decoding with ORBGRAND,” in2023 57th Annual Conference on Information Sciences and Systems (CISS). IEEE, 2023, pp. 1–6

    2023

  9. [17]

    Iterative guessing random additive noise decoder for universal decoding of product codes,

    A. Riaz, K. Galligan, A. Yasar, V . Bansal, K. R. Duffy, M. Medard, and R. T. Yazicigil, “Iterative guessing random additive noise decoder for universal decoding of product codes,”IEEE Transactions on Circuits and Systems I: Regular Papers, 2025

    2025

  10. [18]

    Exponential error bounds for erasure, list, and decision feedback schemes,

    G. Forney, “Exponential error bounds for erasure, list, and decision feedback schemes,”IEEE Transactions on Information Theory, vol. 14, no. 2, pp. 206–220, 1968

    1968

  11. [19]

    Upgrade error detection to prediction with GRAND,

    K. Galligan, P. Yuan, M. Médard, and K. R. Duffy, “Upgrade error detection to prediction with GRAND,” inGLOBECOM 2023-2023 IEEE Global Communications Conference. IEEE, 2023, pp. 1818–1823

    2023

  12. [20]

    GRAND for gaussian intersymbol interference channels,

    Z. Li and W. Zhang, “GRAND for gaussian intersymbol interference channels,”arXiv preprint arXiv:2603.08325, 2026

    2026

  13. [21]

    Error-free coding,

    P. Elias, “Error-free coding,”Transactions of the IRE Professional Group on Information Theory, vol. 4, no. 4, pp. 29–37, 1954

    1954

  14. [22]

    Guesswork, large deviations, and shannon entropy,

    M. M. Christiansen and K. R. Duffy, “Guesswork, large deviations, and shannon entropy,”IEEE transactions on information theory, vol. 59, no. 2, pp. 796–802, 2012

    2012

  15. [23]

    On universal decoding over discrete additive channels by noise guessing,

    H. K. Miyamoto and S. Yang, “On universal decoding over discrete additive channels by noise guessing,” in2025 IEEE Information Theory Workshop (ITW). IEEE, 2025, pp. 1–6

    2025

  16. [24]

    Universal decoding over finite-state additive channels via noise guessing,

    ——, “Universal decoding over finite-state additive channels via noise guessing,”arXiv preprint arXiv:2501.12971, 2025

    2025 arXiv

  17. [25]

    Guessing random additive noise decoding with symbol reliability information (SRGRAND),

    K. R. Duffy, M. Médard, and W. An, “Guessing random additive noise decoding with symbol reliability information (SRGRAND),”IEEE Transactions on Communications, vol. 70, no. 1, pp. 3–18, 2021

    2021

  18. [26]

    ORBGRAND is almost capacity-achieving,

    M. Liu, Y . Wei, Z. Chen, and W. Zhang, “ORBGRAND is almost capacity-achieving,”IEEE Transactions on Information Theory, vol. 69, no. 5, pp. 2830–2840, 2022

    2022

  19. [27]

    Decoding in the presence of ISI without interleaving orbgrand ai,

    K. R. Duffy, M. Grundei, J. A. Millward, M. Rangaswamy, and M. Medard, “Decoding in the presence of ISI without interleaving orbgrand ai,”arXiv preprint arXiv:2510.14939, 2025

    2025 arXiv

  20. [28]

    Low-Pathwidth GRAND: Exact likelihood-ordered enumeration for BPSK transmission over correlated gaussian noise,

    B. Razeghi, “Low-Pathwidth GRAND: Exact likelihood-ordered enumeration for BPSK transmission over correlated gaussian noise,” arXiv preprint arXiv:2607.28363, 2026. [Online]. Available: https: //arxiv.org/abs/2607.28363

    2026 arXiv

  21. [29]

    Soft-input, soft- output joint data detection and GRAND: A performance and complexity analysis,

    H. Sarieddeen, P. Yuan, M. Médard, and K. R. Duffy, “Soft-input, soft- output joint data detection and GRAND: A performance and complexity analysis,” inIEEE International Symposium on Information Theory (ISIT). IEEE, 2023, pp. 1090–1095

    2023

  22. [30]

    Soft-output guessing codeword decoding,

    K. R. Duffy, P. Yuan, J. Griffin, and M. Médard, “Soft-output guessing codeword decoding,”IEEE Communications Letters, 2024

    2024

  23. [31]

    SOGRAND assisted guesswork reduction,

    L. Rapp, M. Médard, and K. R. Duffy, “SOGRAND assisted guesswork reduction,”IEEE Communications Letters, 2025

    2025

  24. [32]

    Leveraging code structure to improve soft output for grand, gcd, osd, and scl,

    J. Feng, K. R. Duffy, and M. Médard, “Leveraging code structure to improve soft output for grand, gcd, osd, and scl,”arXiv preprint arXiv:2503.16677, 2025

    2025 arXiv

  25. [33]

    SOGRAND decoding of LDPC codes,

    K. R. Duffy, J. Feng, L. Rapp, and M. Médard, “SOGRAND decoding of LDPC codes,”arXiv preprint arXiv:2607.04045, 2026

    2026 arXiv

  26. [34]

    List viterbi decoding algorithms with applications,

    N. Seshadri and C.-E. Sundberg, “List viterbi decoding algorithms with applications,”IEEE transactions on communications, vol. 42, no. 234, pp. 313–323, 1994

    1994

  27. [35]

    Optimal decoding of linear codes for minimizing symbol error rate (corresp.),

    L. Bahl, J. Cocke, F. Jelinek, and J. Raviv, “Optimal decoding of linear codes for minimizing symbol error rate (corresp.),”IEEE Transactions on information theory, vol. 20, no. 2, pp. 284–287, 1974

    1974

  28. [36]

    On calibration of modern neural networks,

    C. Guo, G. Pleiss, Y . Sun, and K. Q. Weinberger, “On calibration of modern neural networks,” inInternational conference on machine learning. PMLR, 2017, pp. 1321–1330

    2017

  29. [37]

    Probable inference, the law of succession, and statistical inference,

    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

    1927

  30. [38]

    Capacity of a burst-noise channel,

    E. N. Gilbert, “Capacity of a burst-noise channel,”Bell system technical journal, vol. 39, no. 5, pp. 1253–1265, 1960

    1960

  31. [39]

    Estimates of error rates for codes on burst-noise channels,

    E. O. Elliott, “Estimates of error rates for codes on burst-noise channels,” The Bell System Technical Journal, vol. 42, no. 5, pp. 1977–1997, 1963

    1977

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.