Pith. sign in

REVIEW 4 major objections 5 minor 1 cited by

Computational Approaches to the Singer Transfer: Preimages in the Lambda Algebra and $G_k$-Invariant Theory

T0 review · 4 major / 5 minor · reviewed 2026-08-06 · deepseek-v4-flash

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

Pith's one-line read The paper turns preimage-finding for the Singer transfer into solvable linear systems over the two-element field, gives an explicit preimage for the indecomposable p0 that earlier work left implicit, and claims that a cited proof that d0…

desk verdict A serious computational paper with a real new preimage for p0, but the primitivity checks are incomplete as written and the code is not reproducible; deserves peer review with major revision. read the letter →

arxiv 2507.10108 v4 pith:VRYY55CX submitted 2025-07-14 math.AT

Dang Vo Phuc This is my paper · ORCID
classification math.AT MSC 55T1555S1055S05
keywords SingeralgebraictransferlambdaalgebraAdamsspectralsequencePetersonhitproblemSteenrodG_k-invariantspreimageSageMath
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

The paper is trying to turn the preimage problem for the Singer algebraic transfer into routine computation: given a class in the $\lambda$-algebra cohomology of the Steenrod algebra, equivalently a cell of the Adams spectral sequence, decide whether the transfer detects it and, if so, exhibit an explicit witness. The transfer is realized by a recursion transcribed from [5], and the search for a preimage x and a coboundary z is written as the linear equation $\varphi_k(x) + \delta(z) = y$ over $\mathbb{F}_2$, which a computer can solve and filter by the condition that x is annihilated by the positive-degree Steenrod operations. On this basis the paper exhibits an explicit preimage for the indecomposable p0 that earlier work detected only non-constructively, verifies c0 and d0 against the [5] model, and claims that the proof in [17] that d0 is detected is false because it runs inside a different recursive rule, where the claimed correction term does not even belong to the correct $\lambda$ space and the residual is not a cocycle. The same project produces a SageMath pipeline that recomputes the $G_4$-invariant spaces of the rank-4 Singer conjecture from scratch, confirming $\dim[(QP_4)_{33}]^{G_4} = 1$ with an explicit basis. If correct, the paper converts two previously manual procedures, transfer preimage construction and $G_k$-invariant computation, into reproducible tools and flags a concrete proof in the literature as invalid.

What carries the argument

The machinery has two stages. The first is the chain-level equation $\varphi_k(x) + \delta(z) = y$ in the $\lambda$ algebra $\Lambda$, the bigraded differential algebra with generators $\lambda_t$, $t \geq 0$, Adem relations (1) and differential (2), whose admissible monomials of length k form the chain complex computing $\mathrm{Ext}_{\mathcal A}(\mathbb{Z}/2,\mathbb{Z}/2)$. The Singer transfer is represented on $H_*(B(\mathbb{Z}/2)^k)$ as a map to $\Lambda^k$ through the generating-function correspondence of [5], and its recursive form (Corollary 2.3) expresses $\varphi_k$ of a divided-power monomial $a^{(t_k)}_k \cdots a^{(t_1)}_1$ as a sum of $\lambda_i$ terms feeding $\varphi_{k-1}$ on elements transformed by the right action of the dual Steenrod operations, with the single-generator rule $a^{(t)}(\mathrm{Sq}^j_*) = \binom{t-j}{j}a^{(t-j)}$. Writing the target cocycle y, the unknown x in the divided-power basis, and the unknown z in the length-$(k-1)$ admissible $\lambda$ basis makes $\varphi_k(x) + \delta(z) = y$ a finite $\mathbb{F}_2$-linear system; the algorithm then filters the solution space by the $à$-primitive condition $(x)\mathrm{Sq}^j_* = 0$ for $j = 2^t$, $t = 0, \dots, \lfloor \log_2 \deg x \rfloor$. The second stage is the $G_k$-invariant computation: a weight-vector filtration of $P_k = \mathbb{Z}/2[x_1, \dots, x_k]$ stratifies the cohit space $QP_k$ by weight, an admissible basis is read off from an echelon form of the hit matrix, $\Sigma_k$-components are found by adjacency under the adjacent transpositions $\rho_1, \dots, \rho_{k-1}$, and $G_k$-invariants are obtained by imposing the transvection $\rho_k(x_k) = x_k + x_{k-1}$ through kernel computations, with duality to the transfer domain supplied by the coefficient pairing between divided-power homology and polynomial cohomology.

What would settle it

Run the paper's own SageMath routine (Listing 1) on the 36-term polynomial $q_{4,3}$ using recursion (4): if the Adem-reduced image matches $d_0^* + R$ and $\delta(R) = 0$, or if a legitimate $z \in \Lambda^3$ with $\delta(z) = R$ can be produced, the refutation of [17] fails. Independently, run the paper's linear-system search for the [17] representative of d0 under the Corollary 2.3 recursion: if any pair $(x, z)$ with $x \in PAH_{14}(B(\mathbb{Z}/2)^4)$ and $z \in \Lambda^3$ satisfies $\varphi_4(x) + \delta(z) = y$, the claim that the proof is irreparable is wrong. The same search for p0 is also checkable by substitution: the displayed polynomial x must satisfy $(x)\mathrm{Sq}^1_* = (x)\mathrm{Sq}^2_* = (x)\mathrm{Sq}^4_* = (x)\mathrm{Sq}^8_* = 0$ and $\varphi_4(x) = \lambda_7^2\lambda_5\lambda_{14} + \lambda_7^2\lambda_9\lambda_{10} + \lambda_7\lambda_{11}\lambda_9\lambda_6$.

Watch

Extended reading notes

Core claim

Stated as the author would state it to a fair reader, the central claim is that the transfer's preimage problem is a solvable linear-algebra problem in the $\lambda$-algebra model: a class $[y] \in \mathrm{Ext}^{k,*}_{\mathcal A}(\mathbb{Z}/2,\mathbb{Z}/2)$ lies in the image of $\varphi_k$ exactly when $\varphi_k(x) + \delta(z) = y$ admits a solution with $x$ an $\mathcal A$-annihilated element of $H_*(B(\mathbb{Z}/2)^k)$ and $z$ a length-$(k-1)$ $\lambda$ element. The recursive description of the transfer from [5] (Corollary 2.3) turns the equation into a finite matrix system over $\mathbb{F}_2$, and the paper solves it for three indecomposables. The outputs are explicit: a preimage $x \in PAH_8(B(\mathbb{Z}/2)^3)$ for c0 with $z = 0$; a preimage $x \in PAH_{14}(B(\mathbb{Z}/2)^4)$ with $z = \lambda_3^2\lambda_9 + \lambda_3\lambda_9\lambda_3$ for d0; and a preimage $x \in PAH_{33}(B(\mathbb{Z}/2)^4)$ with $z = 0$ for p0, the p0 witness being a preimage that the earlier detection arguments of [8] did not make explicit. The paper further claims that the proof in [17] that d0 lies in the image of the fourth transfer is false: that paper's recursion (4) places the $\lambda$ factor differently from the model of [5], its correction term $z = \lambda_1\lambda_3\lambda_9\lambda_3 + \lambda_1\lambda_3^2\lambda_9$ does not lie in $\Lambda^3$, and with its representative $d_0 = d_0^*$ the image of its own $q_{4,3}$ is $d_0^* + R$ with $\delta(R) \neq 0$, so no $z \in \Lambda^3$ can close the equation. The paper is explicit that the statement that d0 is detected is not what fails, since it gives a working preimage and notes an earlier alternative proof in [7], and that what collapses is the published argument. Alongside this, the paper claims its SageMath algorithm reproduces the $G_4$-invariant spaces previously computed by hand for the rank-4 Singer conjecture, including $\dim[(QP_4)_{33}]^{G_4} = 1$, and prints an explicit basis polynomial for each invariant space.

Load-bearing premise

Everything hinges on the recursive model of the transfer: every preimage found and every no-solution verdict is computed inside the recursion of Corollary 2.3 (transcribed from [5]) as the true chain-level model and the differing recursion (4) attributed to [17] as a false one, so a misprint or misapplication in either model, or an unchecked cocycle representative imported from [10], would void the witnesses and the refutation alike.

Editorial extensions

If this is right

  • All three target classes, c0, d0, and p0, are detected by the Singer transfer, with explicit primitive witnesses that a reader can check by substitution into the identity $\varphi_k(x) + \delta(z) = y$.
  • The published proof in [17] that d0 lies in the image of the fourth transfer is asserted to be false: its recursion (4) is not the chain-level model of the transfer, its auxiliary element z falls outside $\Lambda^3$, and with its representative of d0 the residual R is not a cocycle, so no correcting z exists.
  • The detection statement for d0 is not itself overturned, since the paper supplies a valid preimage under the [5] recursion and notes an alternative proof in [7]; the correction concerns the proof, not the theorem.
  • The preimage of p0, previously known only through the non-constructive detection arguments of [8], is now given explicitly in $PAH_{33}(B(\mathbb{Z}/2)^4)$, making the membership of p0 in the transfer's image independently checkable.
  • The SageMath pipeline reproduces the manual rank-4 computations, including $\dim[(QP_4)_{33}]^{G_4} = 1$ with an explicit 16-term basis, so the $G_4$-invariant results of the author's earlier papers are verifiable by any reader with enough memory.

Reading between the lines

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

  • Model-dependence is the paper's sharpest edge but also its narrowest: the no-solution verdict is computed with one cocycle representative under one recursion, so it refutes that specific argument rather than establishing that no representative could be found in either model; the cohomology class, not the cocycle, is the stable object.
  • Testable extension: the same matrix formulation applies unchanged to any Ext class, so the method's limit is only the size of the admissible lambda basis and the divided-power basis; a natural next check is a class at a bidegree where the two recursive models would predict different membership.
  • Editorial reading: the divergence between recursion (4) and Corollary 2.3 is a warning about hand computations of the transfer more broadly, since any published preimage computed without fixing the chain-level model is exposed to the same kind of collapse documented here for [17].
  • The pairing $\langle x, f \rangle = 1$ between the p0 preimage and the 16-term $G_4$-invariant gives a concrete template: each global invariant should pair with a dual primitive, so the invariant-space algorithm doubles as a generator of explicit Ext classes, which could populate further cells of the Adams chart.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

4 major / 5 minor

Summary. The paper develops a computational framework for the preimage problem of the Singer algebraic transfer inside the lambda algebra. It formulates the search for x and z satisfying φ_k(x)+δ(z)=y as a linear system, applies it to the classes c0, d0, and p0, and claims to exhibit explicit preimages for c0 and p0, and a preimage for d0 under the Chon-Ha recursion. It further asserts that the proof in Nguyen Sum's paper [17] that d0 lies in the image of the fourth transfer is false, both because the z used there lies in the wrong lambda-algebra length and because Sum's recursion (4) is allegedly not a model of the transfer. The final part of the paper presents a SageMath algorithm for computing dimensions and bases of the G4-invariant spaces [(QP4)d]^{G4}, with a detailed sample run for d=33 and a claim that this verifies the author's earlier manual rank-4 results.

Significance. If the missing verifications are supplied, the explicit polynomial x for p0 would be a concrete new contribution: a previously unpublished preimage for an indecomposable element in Ext^{4,37}_A, with the duality pairing against the G4-invariant f displayed in detail. The d0 computation also provides an explicit witness that the class is in the image of φ4. The SageMath framework and the extensive output for k=4, d=33 are valuable because they make the hit-problem and invariant computations reproducible in principle, and the paper explicitly ships machine-checkable witness polynomials and a sample execution. However, the central positive claims are not yet fully established: the primitive checks for p0 and d0 omit required Steenrod operations, the refutation of [17] rests on an unproved assertion about formula (4), and the extrapolation from finitely many degrees to a general family of degrees is not justified.

major comments (4)
  1. [§3.2 (p0 and d0 cases), Remark 3.2, Eq. (3)] The manuscript verifies (x)Sq^1_* = (x)Sq^2_* = (x)Sq^4_* = (x)Sq^8_* = 0 for the degree-33 element x and says the instability condition makes this sufficient. According to the paper's own criterion in Remark 3.2, one must check (x)Sq^{2^t}_* for t = 0, ..., floor(log2(deg x)), which for deg x = 33 means also Sq^16_* and Sq^32_*. Instability of the cohomology action does not imply vanishing of the dual homology action when 2^t ≤ deg x; the standard example Sq^{deg u}(u) = u^2 shows the distinction. The same gap occurs for the degree-14 d0 witness: floor(log2 14) = 3, so Sq^8_* must be checked, but the text says only 0 ≤ j ≤ 2. Without these checks the displayed polynomials are not proved to lie in PAH_*, and Lemma 3.1 cannot be applied to conclude membership in the image of the transfer.
  2. [Remark 3.3] The refutation of Nguyen Sum's proof depends on two load-bearing assertions: that recursion (4) is not a valid chain-level model of the Singer transfer while Corollary 2.3 is, and that under (4) there is no pair (x,z) with x ∈ PAH14, z ∈ Λ^3, and φ4(x)+δ(z)=y for Sum's representative. Neither assertion is proved in the manuscript. No derivation or citation is given showing that (4) is incompatible with the Chon-Ha formula, and the 'no solution' verdict is reported as a computer result without a rank or consistency certificate. The observation that Sum's z = λ1λ3λ9λ3 + λ1λ3^2λ9 has length four and hence is not in Λ^3 already invalidates the displayed proof; the stronger claim about formula (4) is therefore not needed and should either be proved rigorously or removed.
  3. [§4.1, paragraph on d = 2^{s+3}+2^{s+1}-3] The claim that it suffices to compute s = 1,2,3,4 and then 'match the monomial patterns' to obtain the general G4-invariant formula is an unproved induction. The uniform structure of the admissible basis for all s is asserted from four observed cases, and no theorem or algorithmic certificate establishes that the pattern persists. Since this inference is used to claim verification of the rank-4 Singer conjecture results in [11,12,13], the paper should either prove the pattern by induction or explicitly restrict the claim to the finitely many computed degrees.
  4. [§3.1 and Appendix] The central computational results are not fully auditable as presented. The preimage search is described by pseudocode (Algorithm 0), but the actual SageMath code in the appendix implements the G4-invariant computation; the code used to produce the c0, d0, and p0 witnesses and to establish the 'no solution' verdict for Sum's recursion is not provided. Moreover, the asserted identity φ4(x) = y for p0 is not displayed, unlike the d0 case, and the c0 primitive check is asserted rather than shown. For a computational paper, the generating scripts and machine-readable certificates for these specific claims should be included or linked.
minor comments (5)
  1. [§4.2, Example Computational Workflow] The sample workflow states 'Polynomial ring: Z/2[x1,x2,x3,x4,x5]' for k=4; the ring should be Z/2[x1,x2,x3,x4].
  2. [§3.2, c0 case] The sentence 'For z ≠ 0, − z ∈ {λ2λ7, λ7λ2}' is malformed; the minus sign is spurious and the intended meaning appears to be 'z ∈ {λ2λ7, λ7λ2}'.
  3. [Remark 3.2 and Algorithm 0] The bound in the IsAnnihilated helper function is written as max_t = floor(log2(degree)) + 1 followed by range(max_t - 1); this is consistent with Remark 3.2 but should be stated uniformly to avoid confusion.
  4. [§3.2, p0 case] The representative y = λ7^2λ5λ14 + λ7^2λ9λ10 + λ7λ11λ9λ6 is imported from Lin [10] with no check that it is a cocycle; a short verification would remove reliance on external transcription.
  5. [§2, differential δ] The differential in (2) increases the length of a monomial by one, but this is not stated explicitly; the confusion is visible in Remark 3.3, where the length of z is the key point.

Circularity Check

1 steps flagged · score 2.0 of 10

No central circularity: the preimage computations are exact and self-contained; only a minor, non-load-bearing self-citation loop appears in the G4-invariant verification section.

  1. other [Section 4.1, paragraph 'The algorithm reconstructs...' and Abstract]
    "This tool facilitates the verification of our results [11, 12, 13] that were previously computed manually in connection with Singer's conjecture for rank 4."

    The paper uses its own SageMath algorithm to verify invariant-space dimensions that were originally computed manually in the author's previous papers [11,12,13], and those same papers are cited as the established solution of the rank-4 Singer conjecture. This is a self-referential consistency check: the algorithm and the manual results share the same conceptual machinery, so agreement between them does not by itself provide fully independent confirmation. However, the invariant computation is an explicit, parameter-free calculation over F2 with displayed bases; no fitted parameter or equation is imported from [11-13] into the algorithm, and the preimage/refutation claims in Section 3 do not depend on this loop. Thus the self-citation is minor and not load-bearing.

full rationale

The central derivation chain is not circular. Lemma 3.1 reduces membership in the transfer image to an explicit linear system φ_k(x)+δ(z)=y with x A-annihilated, and the claimed witnesses (x,z) are displayed verbatim: for c0 and d0, the paper exhibits z and x and shows φ4(x)+δ(z)=y; for p0, it exhibits x with z=0 and shows φ4(x)=y. These are direct, machine-checkable equalities over F2, with no parameter fitted to force the equations to balance. The refutation of Sum's proof is also a concrete computation: with recursion (4) and Sum's representative d0*, the paper computes φ4(q4,3)=d0*+R and then directly verifies δ(R)≠0, so no z can correct the discrepancy. That argument is self-contained given the quoted recursion. The dependence on Chon-Ha's Corollary 2.3 and Lin's cocycle representatives is a citation to external prior work, not a reduction of the target claims to themselves. The G4-invariant algorithm verifies the author's own earlier manual results [11-13], a self-citation loop worth noting, but the computation is explicit and independently checkable, so it does not make the paper's central claims circular. A separate correctness concern, not circularity, is that the p0 primitivity check stops at Sq8* even though the paper's own Remark 3.2 and the stated condition require Sq16* and Sq32* for degree 33; this is a verification gap, not an equivalence of input and output by construction.

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

No numerical constants are fitted anywhere in the paper: all witnesses are exact elements over F2, and the search bound 'max_z_terms' in Algorithm 0 is a computational cap, not a mathematical parameter. The axioms listed are the imported formulas and identifications the central claims rely on. No new entities (particles, dimensions, conserved quantities) are postulated; the 'correction polynomial h'' is a computational device within the algorithm, not an invented object.

assumptions (5)
  • domain assumption The Chon-Ha formula (Theorem 2.2) and the recursion of Corollary 2.3 constitute a correct chain-level model of the Singer transfer φk.
    Every preimage found in Section 3 is a solution of φk(x) + δ(z) = y with this φk; if the recursion is mis-transcribed or misapplied, the witnesses are not transfer preimages. The paper imports the formula from [5] without proof.
  • domain assumption The quoted lambda-algebra representatives for c0, d0, p0 from Lin [10] are cocycles representing those Ext classes; in particular λ7λ1λ5λ1 reduces to λ3λ5λ3^2 in the d0 representative.
    These representatives are the target vectors y in the linear systems; the membership claims are only as good as the cohomology-class identification, which is not verified in this paper. Location: Section 3.2.
  • domain assumption An element of degree n is A-annihilated iff (x)Sq^{2^t}_* = 0 for all t with 2^t ≤ n.
    Stated in Remark 3.2 and used by IsAnnihilated; the paper's own verifications do not apply it fully (Sq^16 and Sq^32 are omitted for the degree-33 p0 witness, and Sq^8 for the degree-14 d0 witness).
  • standard math Standard Peterson hit problem facts: weight-vector filtration, admissible monomial bases, generation of Gk by ρ1,...,ρk, and invariant-coinvariant duality.
    Used throughout Section 4; either proven in the paper or referenced to Walker-Wood [19,20].
  • ad hoc to paper For d = 2^{s+3} + 2^{s+1} - 3, computing s = 1,2,3,4 and matching monomial patterns gives the general G4-invariant formula for all s.
    The paper explicitly argues by pattern extrapolation from degrees 17, 37, 77, 128 rather than proving stabilization; this supports the general formulas claimed in [11,12,13].

how reviews work

0 comments
Cite this review

Pith. "Pith review of Computational Approaches to the Singer Transfer: Preimages in the Lambda Algebra and $G_k$-Invariant Theory." pith.science (2026). https://pith.science/paper/VRYY55CX

@misc{pith2026250710108,
  author       = {Pith},
  title        = {Pith review of: Computational Approaches to the Singer Transfer: Preimages in the Lambda Algebra and $G_k$-Invariant Theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/VRYY55CX}},
  note         = {Machine review of arXiv:2507.10108}
}
abstract

We present a systematic, algorithmic method to compute the preimage of elements under the Singer algebraic transfer. Using the lambda algebra and the invariant-theoretic formula of P.H. Chon and L.M. Ha [5], we formulate the preimage search as a solvable problem in linear algebra. This framework is applied to study key indecomposable elements in the Adams spectral sequence. As a consequence, we show that the proof of the known result that the indecomposable element $d_0 \in \mathrm{Ext}^{4,18}_{\mathcal A}(\mathbb Z/2, \mathbb Z/2).$ lies in the image of the fourth Singer transfer, as given by Nguyen Sum in [17], is false. Furthermore, we provide the explicit description of a preimage for the indecomposable element $p_0 \in \mathrm{Ext}^{4,37}_{\mathcal A}(\mathbb Z/2, \mathbb Z/2).$ This preimage had not been explicitly determined in the previous work of N.H.V. Hung and V.T.N. Quynh [8]. Finally, our most significant contribution is the construction of a complete \textsc{SageMath} algorithm that fully automates the computation of both the dimension and an explicit basis for the $G_k$-invariant space $[(Q\mathcal{P}_k)_d]^{G_k}$. This tool facilitates the verification of our results [11, 12, 13] that were previously computed manually in connection with Singer's conjecture for rank 4.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

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

  1. Geometric realization via unoriented bordism and a counterexample to Singer's conjecture for the sixth algebraic transfer

    math.AT 2025-09 conditional novelty 6.0 of 10

    At rank 6 and degree 36 the source of Singer's algebraic transfer is 2-dimensional while the target is 1-dimensional, so the transfer cannot be injective and Singer's conjecture is false.

Reference graph

Works this paper leans on

20 extracted references · 20 canonical work pages · cited by 1 Pith paper

  1. [17]

    Sum, On the determination of the Singer transfer , Vietnam J

    N. Sum, On the determination of the Singer transfer , Vietnam J. Sci., Technol. Eng. 60 (2018), 3–16

    work page 2018

  2. [1]

    J. F. Adams, On the structure and applications of the Steenrod algebra , Comment. Math. Helv. 32 (1958), 180–214

    work page 1958

  3. [2]

    Boardman, Modular representations on the homology of power of real projective space , in Algebraic Topology: Oaxtepec 1991, ed

    J.M. Boardman, Modular representations on the homology of power of real projective space , in Algebraic Topology: Oaxtepec 1991, ed. M. C. Tangor; in Contemp. Math. 146 (1993), 49–70

    work page 1993

  4. [3]

    Bruner, L.M

    R.R. Bruner, L.M. Ha and N.H.V. Hung, On behavior of the algebraic transfer , Trans. Amer. Math. Soc. 357 (2005), 437–487

    work page 2005

  5. [4]

    A. K. Bousfield, E.B. Curtis, D. M. Kan, D. G. Quillen, D. L. Rector and J. W. Schlesinger, The mod-p lower central series and the Adams spectral sequence , Topology, 5 (1966), 331–342

    work page 1966

  6. [5]

    P. H. Chon and L. M. Ha, Lambda algebra and the Singer transfer , C. R. Math. Acad. Sci. Paris 349(1-2) (2011), 21–23

    work page 2011

  7. [6]

    P. H. Chon and L. M. Ha, On the May spectral sequence and the algebraic transfer II , Topology Appl. 178 (2014), 372–383

    work page 2014

  8. [7]

    Ha, Sub-Hopf algebras of the Steenrod algebra and the Singer transfer , Geom

    L.M. Ha, Sub-Hopf algebras of the Steenrod algebra and the Singer transfer , Geom. Topol. Monogr. 11 (2007), 101-124

    work page 2007

Show all 20 references
  1. [8]

    N. H. V. Hung and V. T. N. Quynh, The image of Singer’s fourth transfer , C. R. Math. Acad. Sci. Paris 347 (2009), 1415–1418

    2009

  2. [9]

    Hung and G

    N.H.V. Hung and G. Powell, The A-decomposability of the Singer construction , J. Algebra 517 (2019), 186–206

    2019

  3. [10]

    W. H. Lin, Ext 4,∗ A (Z/2, Z/2) and Ext5,∗ A (Z/2, Z/2), Topology Appl. 155 (2008), 459–496. 100 PHUC VO DANG ∗

    2008

  4. [11]

    Phuc, On the algebraic transfers of ranks 4 and 6 at generic degrees , Rend

    D.V. Phuc, On the algebraic transfers of ranks 4 and 6 at generic degrees , Rend. Circ. Mat. Palermo, II. Ser. 74, 38 (2025). Corrected version (2025), 34 pages. Available online at https://www.researchgate. net/publication/382917122

    2025

  5. [12]

    Phuc, On Singer’s conjecture for the fourth algebraic transfer in certain generic degrees , J

    D.V. Phuc, On Singer’s conjecture for the fourth algebraic transfer in certain generic degrees , J. Homotopy Relat. Struct. 19, 431–473 (2024). Corrected version (2025), 32 pages. Available online at https://arxiv. org/abs/2506.10232

    2024

  6. [13]

    Phuc, The affirmative answer to Singer’s conjecture on the algebraic transfer of rank four , Proc

    D.V. Phuc, The affirmative answer to Singer’s conjecture on the algebraic transfer of rank four , Proc. Roy. Soc. Edinburgh Sect. A 153 (2023), 1529–1542. Corrected version (2025), 25 pages. Available online at https://www.researchgate.net/publication/352284459

    2023

  7. [14]

    Phuc, A matrix criterion and algorithmic approach for the Peterson hit problem: Part I , ArXiv:2506.18392, 47 pages, https://arxiv.org/abs/2506.18392

    D.V. Phuc, A matrix criterion and algorithmic approach for the Peterson hit problem: Part I , ArXiv:2506.18392, 47 pages, https://arxiv.org/abs/2506.18392

    arXiv

  8. [15]

    W. M. Singer, The transfer in homological algebra , Math. Z. 202 (1989), 493–523

    1989

  9. [16]

    Sum, The hit problem for the polynomial algebra of four variables , arXiv:14121709

    N. Sum, The hit problem for the polynomial algebra of four variables , arXiv:14121709

  10. [18]

    Tin, The hit problem for the polynomial algebra in five variables and applications , PhD

    N.K. Tin, The hit problem for the polynomial algebra in five variables and applications , PhD. thesis, Quy Nhon University, Vietnam, 2017

    2017

  11. [19]

    In: London Math

    Walker, G., Wood, R.M.W.: Polynomials and the mod 2 Steenrod Algebra: Volume 1, The Peterson hit problem. In: London Math. Soc. Lecture Note Ser., Cambridge Univ. Press, 2018

    2018

  12. [20]

    In: London Math

    Walker, G., Wood, R.M.W.: Polynomials and the mod 2 Steenrod Algebra: Volume 2, Representations of GL(n, F2). In: London Math. Soc. Lecture Note Ser., Cambridge Univ. Press, 2018. Department of AI, FPT University, Quy Nhon AI Campus, An Phu Thinh New Urban Area, Quy Nhon City,...

    2018

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.