{"id":"a6b2b0e2-4a78-4dcd-895e-1b1f6cba4d05","arxiv_id":"2505.24533","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":2.0,"correctness_risk":"high","formal_verification":"none","parameter_count":2,"one_line_summary":"A framework of multi-axis affine compositions is proposed; its DFT special case works, but the Hadamard and Walsh special cases are invalid as written.","lead":"This preprint proposes an algebraic framework for composing vector-matrix pairs along multiple axes and claims it unifies the DFT, Hadamard, and Walsh transforms as special cases. The core algebra is elementary, and the Hadamard construction contains a concrete arithmetic error, so the main unification claim does not hold.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Hadamard construction in §7 fails numerically at n=4 and the inductive step is not derived, so the central unification claim is unsupported.","rationale":"Reading the paper in good faith, the algebraic framework itself is coherent: the one-dimensional composition is associative, and the two-dimensional interchange law is correctly proven under the commutativity assumption on R_x and R_y. The DFT representation in §6 is a valid re-encoding of the DFT sum using block-diagonal rotations. However, the paper's principal advertised contribution is the unification of DFT, Hadamard, and Walsh transforms, and the Hadamard case is the load-bearing component because the Walsh case is derived from it. Section 7's construction fails arithmetically at n=4, and the inductive step is an assertion rather than a derivation. The reader's rationale identifies exactly this failure, but the reader's stated weakest_assumption is the interchange-law commutativity condition, which is not where the central claim actually collapses. In addition, the representation Σ R^{i−1}v_i is permissive enough that any linear map can be encoded trivially, e.g., R=I and v_1=Mx, so without further constraints on the v_i the claimed unification is not structurally substantive. The decisive and concrete issue remains the incorrect Hadamard construction, which is sufficient to reject the paper's central claim.","tokens_in":7499,"tokens_out":4167,"duration_ms":49804,"concrete_test":"Evaluate the §7 construction at n=4 for x=e1. Compute R=diag(1,−1,1,−1), v1=e1+e3, and v2=v3=v4=0, then compare Σ R^{i−1}v_i with H_4 e1. The construction yields (1,0,1,0)^T while H_4 e1=(1,1,1,1)^T, so the claimed identity fails. An independent symbolic re-derivation of the inductive step should also be run; it will show that no definition of R_{2n} or v′_i is supplied, so the step is not a proof.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is that the DFT, Hadamard transform, and Walsh transform each admit a representation as Σ R^{i−1}v_i within the proposed framework. The DFT construction (§6) is algebraically valid. The Hadamard construction (§7) is not. For n=4, the paper defines R=diag(1,−1,1,−1) and v_i with nonzero entries at positions (i,i+n/2), (i−n/2,i), etc. Direct computation gives Σ_{i=1}^4 R^{i−1}v_i = (x1+x3, −x2−x4, x1−x3, x2+x4)^T, whereas H_4 x = (x1+x2+x3+x4, x1−x2+x3−x4, x1+x2−x3−x4, x1−x2−x3+x4)^T. These vectors are already unequal for x=e1: the construction produces (1,0,1,0)^T, while H_4 e1=(1,1,1,1)^T. The claimed inductive step only asserts that H_n(x(1)+x(2)) and H_n(x(1)−x(2)) can be written using the inductive hypothesis with R_n; it never defines the 2n-dimensional R′, the corresponding v′_i, or proves that they satisfy the required sum identity. Because the Walsh construction in §8 is explicitly built from the Hadamard construction via conjugation by a permutation matrix (R′=P R P−1, v′_i=P v_i), it inherits the Hadamard failure. The paper's advertised unification therefore rests on a false special case, regardless of whether the algebraic framework itself is internally consistent.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes an algebraic framework whose basic objects are pairs (a, A) composed as (a, A) ∘ (b, B) = (a + Ab, AB), generalized to D-dimensional structures with one composition operator per axis and an interchange law enforced by pairwise commuting per-axis operators. It proves the 2D interchange law and sketches the higher-dimensional generalization. The central advertised application is that the DFT, the Hadamard transform, and the Walsh transform can each be represented as Σ R^{i-1} v_i within this framework. The DFT construction is explicit and arithmetically coherent, but the Hadamard construction fails already at n = 2, and the inductive step never defines the required higher-dimensional data. The Walsh construction inherits the Hadamard failure.","tokens_in":7870,"tokens_out":5305,"duration_ms":65533,"significance":"If the claimed unification were correct, it would provide a suggestive organizing principle for structured embeddings and for viewing classical linear transforms as compositions of simple algebraic objects; the explicit DFT construction and the careful 2D interchange-law proof are concrete strengths. However, the Hadamard representation is the load-bearing example for the paper's central claim, and it is false as stated. Since the Walsh claim is derived from the Hadamard claim, the advertised unification of DFT, Hadamard, and Walsh transforms is not established. The remaining framework, while coherent, is too thinly developed on its own to support the paper's stated significance.","major_comments":[{"comment":"The base case for n = 2 is arithmetically false. With R = diag(1, -1), v1 = (x1, x1)^T and v2 = (x2, -x2)^T, one obtains R^0 v1 + R^1 v2 = v1 + R v2 = (x1 + x2, x1 + x2)^T, not (x1 + x2, x1 - x2)^T = H_2 x. Thus the claimed representation fails at the smallest nontrivial size, invalidating the base case of the proposed induction.","section":"Section 7, 'Base Case' and the definition of v_i"},{"comment":"The inductive step is not a proof. After writing H_{2n} x = (H_n(x^(1) + x^(2)), H_n(x^(1) - x^(2)))^T and observing that each half can use the n-dimensional hypothesis, the passage asserts that 'the inductive step holds' without defining the 2n × 2n matrix R' or the vectors v'_i whose sum should equal H_{2n} x, and without proving that the block structure of the n-dimensional representation combines consistently. Consequently, the representation for n = 2^k is not established for any k ≥ 1.","section":"Section 7, 'Inductive Step'"},{"comment":"The Walsh construction is defined by R' = P R P^{-1} and v'_i = P v_i, so it inherits every failure of the Hadamard construction in Section 7. In addition, the text is internally inconsistent about the role of P: it states both W_n = P H_n and W_n x = P H_n(P^{-1} x), and the displayed equality Σ R'^{i-1} v'_i = P H_n(P^{-1} x) requires the v_i to be constructed from the permuted input P^{-1}x, which is not stated in the definition v'_i = P v_i. The Walsh representation is therefore unsupported even conditional on a correct Hadamard construction.","section":"Section 8, Walsh transform construction"}],"minor_comments":[{"comment":"The sentence 'We show that the framework that unifies several well-known linear transforms' is missing a main verb and should be rephrased.","section":"Abstract"},{"comment":"The construction uses the positive-exponent Fourier kernel e^{+j 2πk(i-1)/n}, which corresponds to the inverse DFT under the common negative-exponent convention; the paper should state which convention is intended.","section":"Section 6.1"},{"comment":"The two statements W_n = P H_n and W_n x = P H_n(P^{-1} x) describe different operations (a row permutation versus a conjugation); the paper should clarify which definition of the Walsh matrix is used.","section":"Section 8"},{"comment":"Reference [12] spells the author name as 'Godavart' while the title page spells it 'Godavarti'; the reference list should be corrected.","section":"References"}],"recommendation":"reject","confidential_remarks":"The core algebraic framework in Sections 3–5 is coherent, and the DFT construction is explicit and correct up to sign convention. However, the paper's advertised unification rests on the Hadamard construction, which is already false at n = 2 and is not repaired by the underspecified induction. Since the Walsh claim is parasitic on the Hadamard claim, the central contribution as stated cannot be salvaged by local revision; a correct construction, if one exists, would constitute a substantial new result rather than a repair of the present argument."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe punchline: the paper's central claim—that the DFT, Hadamard, and Walsh transforms all fit one compositional framework—is unsupported because the Hadamard construction is wrong at the base case. The algebraic shell is a minor re-packaging of affine compositions; the DFT case is correct but trivial.\n\nWhat is actually new: not much. The 1D composition (a,A)∘(b,B)=(a+Ab,AB) is the standard affine group (semidirect product). The D-dimensional version with per-axis operators and a commutativity assumption is a natural extension, and the interchange-law proof in 2D is correct under that assumption. The DFT construction arithmetic checks out: block-diagonal rotations reproduce the real and imaginary parts of the DFT sum. That part is fine but it is a recognition, not a derivation—the R and v_i are chosen so the sum equals the target transform.\n\nThe soft spot is fatal. Section 7 claims a Hadamard construction with R=diag(1,-1,...). For n=2, v1=(x1,x1) and v2=(x2,-x2). Then R v2 = (x2,x2), so v1+R v2 = (x1+x2, x1+x2), not (x1+x2, x1-x2) as printed. The base case fails on its own equation. The inductive step is not a proof: it says H_{2n}x = [H_n(x(1)+x(2)); H_n(x(1)-x(2))] and then asserts both halves can be written as sums without ever defining the 2n-dimensional R' or the v'_i. That is a non-sequitur. The Walsh construction in §8 inherits the failure because it merely conjugates the Hadamard one. So two of the three advertised special cases are either trivial or wrong.\n\nThe interchange law itself is a modeling assumption: order-independence holds only when the axis operators commute, which is stated but not discussed as a limitation. The paper overclaims in the conclusion, and the prior work reference [12] appears to contain the same framework, so the novelty is thin.\n\nWho is this for? Perhaps a reader curious about algebraic embeddings, but they would already know the affine group and the DFT-as-rotation-sum. The arithmetic error is easily checked and kills the main claim. My recommendation: desk reject. No referee time needed for a paper whose central special case fails at n=2.","headline":"The Hadamard construction fails at n=2, so the claimed unification of DFT, Hadamard, and Walsh is unsupported; the framework itself is a standard affine composition.","tokens_in":8410,"tokens_out":3266,"would_cite":false,"duration_ms":34157,"reading_group":"no","serious_thinker":"no","would_accept_peer_review":false},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["18M05"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper tries to show that the DFT, the Hadamard transform, and the Walsh transform all arise as special cases of one directional, axis-wise composition algebra based on commuting operator powers.","keywords":["monoidal categories","interchange law","non-commutative composition","compositional embeddings","discrete Fourier transform","Hadamard transform","Walsh transform","learnable transformations"],"falsifier":"Run the paper's Section 6 construction for a signal of length $n=5$: build $R$ as the block-diagonal $2\\times2$ rotation matrix with the stated angles, build $v_i$ by repeating $(a_i,0)$ across all blocks, compute $V=\\sum_{i=1}^{n}R^{i-1}v_i$, and compare each block's two entries against $(\\mathrm{Re}\\,X_k, \\mathrm{Im}\\,X_k)$ of the FFT. The claimed representation is settled by exact agreement for random $a_i$; any mismatch in sign, exponent, or block order refutes the construction.","tokens_in":7272,"feed_emoji":"🧮","tokens_out":9191,"duration_ms":115624,"temperature":0.7,"pith_summary":"The paper proposes a class of algebraic structures for composing data along several directions at once, built from vector-matrix pairs and a recursive composition rule for each axis. Its central claim is that three classical transforms, the discrete Fourier transform, the Hadamard transform, and the Walsh transform, are special cases of the same compositional embedding expression $V = \\sum_{i=1}^n R^{i-1} v_i$, obtained by specific choices of the matrix $R$ and vectors $v_i$. The structural guarantee is an interchange law: for independent axes, composing horizontally then vertically gives the same result as composing vertically then horizontally, provided the per-axis operators commute pairwise. If the claim is right, standard signal-processing transforms become instances of a single learnable, axis-aware composition scheme rather than separate algorithms.","feed_headline":"One algebra yields DFT, Hadamard, and Walsh transforms","feed_subtitle":"Composing embeddings along independent axes reproduces the Fourier, Hadamard, and Walsh spectra from one rule.","key_machinery":"The central object is the composed embedding $(a,R_1^{n_1},\\ldots,R_D^{n_D})$ together with the axis-wise composition rule $(a,A)\\circ(b,B)=(a+Ab,AB)$ in one dimension and its per-axis generalization. The operator exponents record how many times each axis has been traversed, the vector accumulates direction-specific contributions, and the identity $V=\\sum_i R^{i-1}v_i$ is the bridge to classical transforms. Its defining structural property is the interchange law $(a\\circ_x b)\\circ_y(c\\circ_x d)=(a\\circ_y c)\\circ_x(b\\circ_y d)$, which holds exactly because the per-axis generators $R_x,R_y$ are assumed to commute.","core_discovery":"Define an element in one dimension as $(a,A)$ with $a$ a vector and $A$ an invertible matrix, and compose by $(a,A)\\circ(b,B)=(a+Ab,AB)$. In $D$ dimensions an element carries one exponent per axis, $(a,R_1^{n_1},\\ldots,R_D^{n_D})$, and composing along axis $i$ adds the exponent on that axis while applying $R_i^{n_i}$ to the incoming vector. When the $R_i$ commute pairwise, the horizontal-then-vertical and vertical-then-horizontal orders of composition agree, which is the interchange law. On a grid of embeddings this yields $E=(\\sum_{i,j} R_x^{i-1}R_y^{j-1}v_{ij}, R_x^H R_y^W)$, and the paper shows that setting $R$ to a block-diagonal rotation matrix and $v_i$ to repeated scalar blocks reproduces the DFT coefficients exactly; setting $R$ to an alternating-sign diagonal matrix and $v_i$ to paired basis vectors reproduces the Hadamard transform by an inductive doubling argument; and applying a Gray-code permutation to that construction reproduces the Walsh transform.","pith_inferences":["The paper's recipe suggests a sufficiency criterion: any transform whose matrix can be written as a linear combination of powers of a single operator $R$, and in Walsh's case a conjugation by a permutation, should admit the same embedding without needing a separate derivation.","The commuting-operator condition points to a practical design: parameterize $R_1,\\ldots,R_D$ as simultaneously diagonalizable matrices so the interchange law holds by construction, giving a differentiable, order-robust positional encoding for attention layers on grids.","Testing the construction on non-power-of-two Hadamard-like transforms or on Walsh variants with different sequency orders would show how far the representation extends, since the paper's inductive proof explicitly relies on $n=2^m$."],"forward_implications":["The DFT, Hadamard, and Walsh transforms are all expressible as $V = \\sum_{i=1}^n R^{i-1} v_i$, so any system that can compose embeddings of this form can compute all three transforms with the same algebraic machinery.","Because composition along each axis is associative and, under commuting operators, order-independent across axes, grid and tensor data can be encoded without ambiguity about which axis was processed first.","The 2D DFT construction follows by applying the same 1D embedding along rows then columns, reproducing the standard separable two-step Fourier transform and extending by the same argument to $N$ dimensions.","For the Hadamard case the representation is recursive, using the doubling construction behind Sylvester's Hadamard matrices, so it scales with $n=2^m$ and carries over to tensor products for higher-dimensional Hadamard transforms.","Because $R$ and the $v_i$ are parameters of the composition, the same framework supports learnable replacements for the fixed classical transforms."],"supporting_citations":[{"why":"Supplies the standard definition of monoidal structures and categories that the framework generalizes to higher dimensions.","marker":"[1]"},{"why":"Introduces bimonoidal structures and the interchange law that the per-axis composition rules are designed to satisfy.","marker":"[2]"},{"why":"Provides the n-fold and iterated monoidal categorical setting that motivates the D-dimensional composition framework.","marker":"[8]"},{"why":"Supplies the Hadamard matrix definition and the recursive doubling construction used in Section 7 to build $R$ and $v_i$.","marker":"[13]"},{"why":"Supplies the Walsh transform and its sequency ordering, which Section 8 realizes by a Gray-code permutation applied to the Hadamard embedding.","marker":"[14]"}],"fun_headline_variants":["Single algebra unifies DFT, Hadamard, and Walsh","One composition rule reproduces three classic transforms","Directional algebra subsumes Fourier, Hadamard, Walsh","Unified structure yields DFT, Hadamard, Walsh spectra","From one algebraic rule: DFT, Hadamard, Walsh"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The interchange law, and hence the claim that multi-axis composition order does not matter, stands on the assumption that the per-axis operators $R_x,R_y,\\ldots$ commute pairwise; if they do not, the two composition orders give different results and the framework loses its central structural guarantee.","fun_headline_variants_meta":{"raw":{"variants":["Single algebra unifies DFT, Hadamard, and Walsh","One composition rule reproduces three classic transforms","Directional algebra subsumes Fourier, Hadamard, Walsh","Unified structure yields DFT, Hadamard, Walsh spectra","From one algebraic rule: DFT, Hadamard, Walsh"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000778,"raw_usage":{"total_tokens":3446,"prompt_tokens":961,"completion_tokens":2485,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":577,"completion_tokens_details":{"reasoning_tokens":2402}},"tokens_in":577,"tokens_out":2485,"duration_ms":22309,"temperature":1.0,"reasoning_tokens":2402,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:20:08.273291+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the paper's Section 6 construction for a signal of length $n=5$: build $R$ as the block-diagonal $2\\times2$ rotation matrix with the stated angles, build $v_i$ by repeating $(a_i,0)$ across all blocks, compute $V=\\sum_{i=1}^{n}R^{i-1}v_i$, and compare each block's two entries against $(\\mathrm{Re}\\,X_k, \\mathrm{Im}\\,X_k)$ of the FFT. The claimed representation is settled by exact agreement for random $a_i$; any mismatch in sign, exponent, or block order refutes the construction.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the standard definition of monoidal structures and categories that the framework generalizes to higher dimensions."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces bimonoidal structures and the interchange law that the per-axis composition rules are designed to satisfy."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the n-fold and iterated monoidal categorical setting that motivates the D-dimensional composition framework."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Hadamard matrix definition and the recursive doubling construction used in Section 7 to build $R$ and $v_i$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the Walsh transform and its sequency ordering, which Section 8 realizes by a Gray-code permutation applied to the Hadamard embedding."}],"review_version":1}