{"id":"f350d3ce-0452-4f1b-bf63-a925c9c9598f","arxiv_id":"1908.04545","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"It gives sufficient conditions, based on a notion called n-simplicity and stable range, for diagonal reduction of matrices over Ore domains.","lead":"The paper proves conditions under which matrices over a class of noncommutative rings can be simplified to diagonal form using row and column operations. A generalist might read it because these results extend a classical part of algebra to a broader family of rings.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 2's proof assumes without argument that every full matrix A admits a right diagonal reduction AT = diag(...); this is not implied by the hypotheses and is essential to Theorems 2 and 3.","rationale":"We agree with the reader that the unproved existence of T with AT diagonal in Theorem 2 is the most load-bearing flaw. It is not a cosmetic issue: the proof cannot proceed without it, and it is not a known consequence of the stated hypotheses. The concrete counterexample candidate [[a,b],[0,1]] shows that the assertion is equivalent to every nonzero element generating R as a right ideal, i.e., R being a division ring. Since the paper gives no example of a 2-simple Ore domain of stable range 1 outside division rings, the assertion is at best a missing lemma and at worst false for the intended generality. The proof also relies on Proposition 2(iv) for Bezout rings while only assuming an Ore domain, another gap, but the AT step alone is sufficient to invalidate Theorems 2 and 3. We note that the paper does cite relevant literature and states definitions, but the central derivations lack support. Therefore the REJECT verdict stands.","tokens_in":6012,"tokens_out":25858,"duration_ms":242655,"concrete_test":"Test the AT step with A = [[a,b],[0,1]] over a ring R satisfying the hypotheses of Theorem 3 (2-simple Ore domain, stable range 1). If AT = diag(ε1, ε2) for T ∈ GL_2(R), then the (2,1) and (2,2) entries force T = [[t11,t12],[0,ε2]] with ε2 a unit, and the (1,2) entry gives b ∈ aR. Thus the asserted existence of T for every full A implies b ∈ aR for every nonzero a and every b, equivalent to R being a division ring. Check whether a 2-simple Ore domain of stable range 1 that is not a division ring exists (e.g., known examples of simple Ore domains with stable range 1). If such a ring exists, choose a nonzero nonunit a and b ∉ aR; then A is full but has no right diagonal reduction, directly refuting the first line of the proof of Theorem 2. If no such ring exists, the theorem is vacuous for domains and the unsupported step still needs replacement.","verdict_should_be":"UNCHANGED","load_bearing_attack":"In the proof of Theorem 2 (Section 3), the first line asserts without proof: 'According to the restrictions imposed on R and A we have AT = diag(ε1,...,ε_{n+1}) ≠ 0 for some matrix T.' This is the load-bearing step. The stated hypotheses — R an (n+1)-simple Ore domain of stable range n, A a non-zero-divisor (full) — do not imply that A admits a right multiplication to a nonzero diagonal matrix. For n=1, if such T exists, then for A = [[a,b],[0,1]] with a ≠ 0 (full in a domain), the (1,2) entry of AT forces b ∈ aR. Thus the assertion implies that every nonzero a generates R as a right ideal, i.e., R is a division ring. If R is not a division ring, pick a nonunit a and b ∉ aR; then the full matrix [[a,b],[0,1]] has no right diagonal reduction, contradicting the asserted step. The proof also later invokes Proposition 2(iv), a Bezout-specific completion result, for an arbitrary Ore domain. Without the AT justification, equation (8) and the completion argument cannot start, so Theorem 2 and its consequence Theorem 3 are unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies diagonal reduction of matrices over simple Ore domains. It introduces the notion of n-simple rings and claims four theorems: Theorem 1 reduces diag(a,b) to diag(1,c) in a 2-simple ring of stable range 1; Theorem 2 asserts that every full (n+1)x(n+1) matrix over an (n+1)-simple Ore domain of stable range n is equivalent to a block matrix with a 1 in the upper-left corner; Theorem 3 specializes this to 2x2 matrices; and Theorem 4 extends the result to n-simple Bezout domains. The proof strategy is to use n-simplicity to construct a unimodular row, use stable range to complete it to an invertible matrix, and then eliminate off-diagonal entries by elementary transformations.","tokens_in":6259,"tokens_out":18085,"duration_ms":179440,"significance":"If the theorems were correct, Theorem 3 would provide a clean sufficient condition for diagonal reduction of 2x2 matrices over simple Ore domains, a result of genuine interest in noncommutative ring theory. The paper is rooted in established work by Cohn, Henriksen, and Zabavsky, and the overall strategy is plausible. However, the manuscript does not contain machine-checked proofs or reproducible code, and the proof of the central theorem contains unproved assertions and uses an inapplicable proposition; the claimed results are therefore not established by the present write-up.","major_comments":[{"comment":"The proof of Lemma 1 does not verify identity (3) with the substitutions as written. For i=2, u2 a2 v2 = (x2 a2) a2 (a3...an y2) = x2 a2^2 a3...an y2, whereas the term needed to reconstruct the sum x2(a1...an)y2 is x2 a1 a2...an y2. The definition should have u2 := x2 a1, not u2 := x2 a2. As written, the displayed factorization fails in a noncommutative ring, so the proof of Theorem 1, which relies on Lemma 1, is not sound.","section":"Section 3, Lemma 1"},{"comment":"The proof begins with the assertion that AT = diag(epsilon_1,...,epsilon_{n+1}) is nonzero for some matrix T, but no proof or reference is given. This is a one-sided diagonal reduction of a full matrix and is not a formal consequence of (n+1)-simplicity and stable range n as stated; it is essentially a weak form of the diagonalization result being proved. The subsequent construction of u, v, and equation (8) depends entirely on this step. In addition, the definition of w just before (8) is incorrect: from AT = diag(epsilon_i) one needs w = T v, not w^T = T^T u^T, so the displayed identity (8) does not follow as written.","section":"Section 3, Theorem 2"},{"comment":"The proof invokes Proposition 2(iv) to complete the unimodular row (u1,...,u_{n+1}) and the column (w1,...,w_{n+1})^T to matrices in GE_m(R). However, Proposition 2(iv) is stated only for right Bezout rings of finite stable range, while the hypotheses of Theorem 2 assume only that R is an (n+1)-simple Ore domain. No argument is given that such an Ore domain is right Bezout, so the completion step is unsupported. Even if a stable-range completion to GL_{n+1}(R) were available, the proof as written relies on a proposition whose hypotheses are not met.","section":"Section 3, Theorem 2 (Proposition 2(iv))"},{"comment":"The proof of Theorem 4 consists of a single sentence asserting that each A_i is triangular because each commutative Bezout domain is a Hermite ring. This does not address the noncommutative setting of the theorem, and it does not explain how Theorem 2 yields the displayed block form with triangular blocks. The claimed result is therefore not proved.","section":"Section 3, Proof of Theorem 4"}],"minor_comments":[{"comment":"The proof refers to 'Lemma 6', but no Lemma 6 exists; the intended reference is Lemma 3.","section":"Section 3, Lemma 4"},{"comment":"The text assigns 'v1 := p12 and v1 := p22'; the second assignment should be v2 := p22. Additionally, in a domain with a nonzero, the equalities a p12 = a p22 = 0 force p12 = p22 = 0, contradicting the invertibility of P, so the case analysis needs to be revisited.","section":"Section 3, Lemma 2, Case 2"},{"comment":"The displayed sum involving u1 epsilon1 v1 through u_{n+1} epsilon_{n+1} v_{n+1} is written as equal to 0; it should be equal to 1 for the subsequent conclusion to hold.","section":"Section 3, Theorem 2"},{"comment":"The term 'FI-ring' is used without definition, which makes Proposition 1 difficult to interpret.","section":"Section 2, Proposition 1"},{"comment":"There are numerous typographical errors (e.g., 'Clorollary' in Proposition 2(iv), inconsistent renditions of 'Bezout'), and the citation '[9, p. 29-30]' in the proof of Theorem 4 is too vague to identify the relevant result.","section":"Throughout"}],"recommendation":"reject","confidential_remarks":"The paper is not ready for publication in its current form. The central proof of Theorem 2 contains an unproved one-sided diagonal reduction step and applies Proposition 2(iv) outside its stated hypotheses. These are load-bearing gaps, not mere presentation issues. The manuscript also has many typographical errors and mislabeled references, suggesting that it would need substantial rewriting before it could be fairly evaluated."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper attacks a legitimate problem: extending Zabavsky's characterization of elementary divisor rings from Bezout domains to simple Ore domains. Theorem 1, about reducing diag(a,b) over a 2-simple ring of stable range 1, is a reasonable new statement, and Lemma 3 (existence of a + xay as a unit) has the right flavor. But the main result, Theorem 2, is not proven.\n\nThe first line of Theorem 2's proof asserts, without argument, that every full matrix A over an (n+1)-simple Ore domain of stable range n admits a nonzero right multiplication AT = diag(...). That is a weak form of the theorem itself, and it is not a consequence of the stated hypotheses. The stress-test's specific counterexample does not quite work as written, because T = E11 often produces a nonzero diagonal matrix. But the underlying gap is real: in a free algebra, the row (x,y) has no nontrivial relation, and a matrix whose rows are such elements need not be right-multipliable to a diagonal form. So the proof collapses at that point unless the authors supply a genuine argument.\n\nBeyond that, the paper has several concrete problems. Lemma 1's proof uses incorrect substitutions (e.g., u2 := x2a2 rather than x2a1), so the displayed sum does not telescope to 1. The proof of Theorem 2 then writes a sum of u_i epsilon_i v_i equal to 0, which is surely a typo for 1; but even if corrected, a diagonal matrix with zero entries does not automatically support such a representation. Proposition 2(iv) is a Bezout-specific completion result, yet it is applied to an arbitrary Ore domain. Lemma 4 cites a non-existent Lemma 6, and Theorem 4's proof is a non sequitur about commutative Bezout domains.\n\nSome of these are minor and fixable. The loads-bearing problem is the missing justification for AT being diagonal. If the authors can prove that step, the paper might be salvageable; if not, Theorems 2 and 3 are unsupported. The citation pattern is honest, and the topic deserves attention, but the current manuscript is too rough for acceptance.\n\nFor peer review: I would send it to a knowledgeable referee rather than desk-reject, because the questions are real and the flaws are specific enough that a referee can tell whether the approach is repairable. Expect heavy revision if it comes back.","headline":"A plausible extension of diagonal reduction to simple Ore domains, but the main theorem's proof contains an unproven right-diagonalization step and multiple errors; not publishable as is.","tokens_in":6793,"tokens_out":16353,"would_cite":false,"duration_ms":147633,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["19B10","16E50","16U10","16U20"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper establishes that every non-zero-divisor 2×2 matrix over a 2-simple Ore domain of stable range 1 is equivalent to diag(1,a), with extensions to larger matrices over (n+1)-simple Ore domains and to Bézout domains.","keywords":["Ore domain","Bezout domain","stable range","n-simple ring","diagonal reduction","full matrix","Hermite ring","elementary divisor ring"],"falsifier":"Exhibit a 2-simple Ore domain R of stable range 1 and a non-zero-divisor 2×2 matrix A over R that is not equivalent to diag(1,a). A more direct test targets the unproved step in Theorem 2: find such R and A for which no matrix T makes AT diagonal; since the proof rests on that diagonal-multiple assertion, a single example of that kind would refute the reduction claim. Concretely, one could examine small Ore domains like skew polynomial rings with stable range 1 and enumerate 2×2 matrices to check whether every full matrix is equivalent to a two-term diagonal.","tokens_in":5800,"feed_emoji":"🔢","tokens_out":8098,"duration_ms":71512,"temperature":0.7,"pith_summary":"This paper asks when matrices over a noncommutative integral domain can be brought to diagonal form by left and right multiplication by invertible matrices—the noncommutative analogue of Smith normal form. Its central claim is that a simplicity condition called n-simplicity, combined with finite stable range, guarantees such diagonal reduction. Concretely, Theorem 3 states that over a 2-simple Ore domain of stable range 1, every non-zero-divisor 2×2 matrix is equivalent to diag(1,a), and Theorem 2 gives a block-diagonal reduction for (n+1)×(n+1) matrices over (n+1)-simple Ore domains of stable range n. A sympathetic reader cares because diagonal reduction is a strong structural property, and the paper offers a clean sufficient condition in terms of two standard ring-theoretic invariants.","feed_headline":"Non-zero-divisor 2x2 matrices diagonalize over 2-simple Ore domains","feed_subtitle":"In stable range 1, full matrices reduce to diag(1,a), the noncommutative analogue of Smith normal form.","key_machinery":"The load-bearing object is the n-simple ring: a simple ring R in which every nonzero a admits a two-sided expression sum_{i=1}^n u_i a v_i = 1 with u_i, v_i in R, and n minimal. Stable range n is the second ingredient: every unimodular (n+1)-row can be shortened to a unimodular n-row by adding a suitable multiple of the last entry. Lemma 1 converts n-simplicity into a two-sided linear combination for products a_1 ... a_n. Lemma 3 uses stable range 1 to turn a nonzero a into a unit of the form a + x a y, and Lemma 4 turns diag(a,a) into diag(1,b). For Theorem 2, the crucial step is equation (8): after right-multiplying A to a nonzero diagonal matrix, (n+1)-simplicity gives a rank-one identity u A w = 1, and stable range n lets the row u and column w be completed to invertible matrices, placing a 1 in the upper-left corner. This corner-1 creation is the mechanism that carries the whole argument.","core_discovery":"The paper's core discovery is that n-simplicity and stable range n combine to place a 1 in the corner of any non-zero-divisor matrix, after which the matrix reduces to a simple block form. The main theorem (Theorem 3) reads: if R is a 2-simple Ore domain of stable range 1, then for every non-zero-divisor A in $R^{{2×2}}$ there exist invertible P,Q with PAQ = diag(1,a). Theorem 2 states the analogous result in dimension n+1: for an (n+1)-simple Ore domain of stable range n, any non-zero-divisor (n+1)×(n+1) matrix is equivalent to a block matrix (1 0; 0 A0) with A0 an n×n matrix. The paper also derives a consequence for Bézout domains (Theorem 4), where the reduction yields triangular blocks instead of diagonal ones.","pith_inferences":["If the missing assertion in the proof of Theorem 2 can be supplied, the same machinery might push diagonal reduction from 2×2 matrices to all full matrices over 2-simple Ore domains of stable range 1, not just the stated sizes.","The unproved step—that A can be right-multiplied to a nonzero diagonal matrix—could be added as an explicit hypothesis to obtain a conditional theorem, in which case the paper's contribution would be the corner-1 reduction rather than the existence of diagonal multiples.","A testable extension is to check small concrete Ore domains, such as skew polynomial rings or simple Artinian-like domains, for the n-simple and stable-range conditions and verify whether all full matrices reduce to diagonal form by explicit algorithms."],"forward_implications":["Over a 2-simple Ore domain of stable range 1, every non-zero-divisor 2×2 matrix has a two-term diagonal form diag(1,a), so such rings behave like noncommutative principal ideal domains at the 2×2 level.","Theorem 1 gives the same conclusion for diagonal matrices diag(a,b) whenever ab ≠ 0 or ba ≠ 0 in a 2-simple ring of stable range 1, covering rings that need not be domains.","Theorem 2 reduces non-zero-divisor (n+1)×(n+1) matrices over (n+1)-simple Ore domains of stable range n to a block form with a 1 in the corner, providing a uniform reduction in all dimensions.","Theorem 4 shows that n-simple Bézout domains admit a block triangular reduction for every larger square matrix, with an identity block and n×n triangular blocks."],"supporting_citations":[{"why":"Supplies the standard facts on stable range, Hermite rings, and elementary divisor rings, including the result that each Bézout domain is an Ore domain, used throughout and in Theorem 4.","marker":"[9]"},{"why":"Provides the notions of inner rank, full matrices, their factorization theory, and the criterion for a square matrix over an Ore domain to be full, which underlie the non-zero-divisor assumption.","marker":"[3]"},{"why":"Establishes the reduction matrix property for right principal Bézout domains and the embedding of an Ore domain into a division ring, used to define row and column ranks.","marker":"[4]"},{"why":"Henriksen's theorem that unit-regular rings are elementary divisor rings, cited in Proposition 2(i) and used in the proof of Lemma 2.","marker":"[6]"},{"why":"Provides the quotient ring construction and the fact that each right Bézout domain is a right Ore domain, cited in Proposition 2(ii).","marker":"[8]"},{"why":"Each Bézout domain is a Hermite ring, a fact used in the introduction and in deriving the Bézout-domain consequence of Theorem 4.","marker":"[1]"}],"fun_headline_variants":["Every non-zero-divisor 2x2 matrix is diag(1,a) over 2-simple Ore domains","Noncommutative Smith normal form: 2-simple Ore domains, stable range 1","2-simple Ore domains of stable range 1 force diag(1,a) for non-zero-divisor matrices","Ore domains: stable range 1 and 2-simplicity yield canonical diagonal reduction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of Theorem 2 assumes, without proof, that every non-zero-divisor (n+1)×(n+1) matrix A over an (n+1)-simple Ore domain of stable range n can be right-multiplied by some matrix T to become a nonzero diagonal matrix; if this assumption fails, Theorems 2 and 3 do not follow.","fun_headline_variants_meta":{"raw":{"variants":["Every non-zero-divisor 2x2 matrix is diag(1,a) over 2-simple Ore domains","Noncommutative Smith normal form: 2-simple Ore domains, stable range 1","2-simple Ore domains of stable range 1 force diag(1,a) for non-zero-divisor matrices","Ore domains: stable range 1 and 2-simplicity yield canonical diagonal reduction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000756,"raw_usage":{"total_tokens":3265,"prompt_tokens":757,"completion_tokens":2508,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":373,"completion_tokens_details":{"reasoning_tokens":2410}},"tokens_in":373,"tokens_out":2508,"duration_ms":18046,"temperature":1.0,"reasoning_tokens":2410,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:39:53.517120+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a 2-simple Ore domain R of stable range 1 and a non-zero-divisor 2×2 matrix A over R that is not equivalent to diag(1,a). A more direct test targets the unproved step in Theorem 2: find such R and A for which no matrix T makes AT diagonal; since the proof rests on that diagonal-multiple assertion, a single example of that kind would refute the reduction claim. Concretely, one could examine small Ore domains like skew polynomial rings with stable range 1 and enumerate 2×2 matrices to check whether every full matrix is equivalent to a two-term diagonal.","supporting_citations":[{"cited_title":"Zabavsky","cited_arxiv_id":null,"evidence_quote":"Supplies the standard facts on stable range, Hermite rings, and elementary divisor rings, including the result that each Bézout domain is an Ore domain, used throughout and in Theorem 4."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the notions of inner rank, full matrices, their factorization theory, and the criterion for a square matrix over an Ore domain to be full, which underlie the non-zero-divisor assumption."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Establishes the reduction matrix property for right principal Bézout domains and the embedding of an Ore domain into a division ring, used to define row and column ranks."},{"cited_title":"Henriksen","cited_arxiv_id":null,"evidence_quote":"Henriksen's theorem that unit-regular rings are elementary divisor rings, cited in Proposition 2(i) and used in the proof of Lemma 2."},{"cited_title":"Stenstr¨ om","cited_arxiv_id":null,"evidence_quote":"Provides the quotient ring construction and the fact that each right Bézout domain is a right Ore domain, cited in Proposition 2(ii)."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Each Bézout domain is a Hermite ring, a fact used in the introduction and in deriving the Bézout-domain consequence of Theorem 4."}],"review_version":1}