{"id":"c91a1113-da8e-47ac-9147-65b18d477422","arxiv_id":"2505.24546","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Corrected necessary and sufficient coefficient conditions for q-Weil polynomials of degrees 6, 8 and 10, fixing earlier errors by Haloui, Haloui-Singh and Sohn, with proofs based on a derivative-root criterion.","lead":"This paper corrects flawed published descriptions of q-Weil polynomials, the coefficient objects that classify abelian varieties over finite fields, for dimensions 3, 4 and 5. It provides new if-and-only-if coefficient inequalities and locates exactly when the corresponding polynomials have real roots.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The proof of Proposition 2.4 relies on an unproved root-counting lemma in Appendix A, so the central criterion—and hence Main Theorems A, B, and C—may rest on an unverified assertion.","rationale":"The reader's weakest_assumption correctly locates Proposition 2.4 as the structural premise, citing the perturbation argument in Section 2 as only sketched. My stress-test agrees but goes further: the appendix's alternate proof, which is supposed to remove the doubt, itself contains an unproved lemma—Lemma A.2(VI)—and an only-sketeched summation argument leading to (15). Since the main theorems are claimed corrections of previous flawed characterizations, a gap at this central lemma is decisive. The reported Magma tests provide practical evidence for the final inequalities, but they do not prove the coefficient space is fully covered, and the paper's contribution is precisely a proof, not merely numerical validation. Therefore the verdict should move to REJECT, with the caveat that a targeted computational stress-test and a completed appendix proof would likely restore confidence.","tokens_in":23487,"tokens_out":2626,"duration_ms":29985,"concrete_test":"Implement Proposition 2.4 in a computer algebra system and test it on random real polynomials of degree 3 through 10, while forcing multiple derivative roots by adding repeated factors. Check condition (⋄) against the true real-root count from a high-precision root finder, with special attention to (a) derivative multiple roots that are not roots of f, (b) polynomials with non-real roots whose derivative-root values alternate in sign, and (c) the families used in Propositions 3.2 and 3.3. If the search produces a counterexample, the main theorems are at risk; if it produces none over a large sample, the practical risk is reduced, though a formal proof of Lemma A.2(VI) would still be needed.","verdict_should_be":"REJECT","load_bearing_attack":"Proposition 2.4 is the engine of the paper: it converts real-rootedness of h± into the coefficient inequalities in Main Theorems A, B, and C. Section 2 gives only a sketchy perturbation argument; Appendix A is meant to supply a direct proof. However, Proposition A.3(a) depends on Lemma A.2(VI), which asserts that f(x) has only real roots if and only if S(−∞, +∞) = 1. The proof of that lemma is essentially one line and does not justify that the sum of root-counts over the real intervals detects the presence of non-real roots of f. Moreover, the combinatorial argument leading to (15) is asserted rather than fully demonstrated: the case analysis in (i)–(iii) is plausible, but the conclusion that a −1 in the sequence Si is never balanced when SIGN fails is only sketched. Because the paper's main theorems are corrections of previously flawed results, and because this exact criterion is the structural premise used in every section, the unproved steps in Appendix A are load-bearing.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper gives explicit inequalities characterizing the q-Weil polynomials of degrees 6, 8, and 10, stated as Main Theorems A, B, and C. The strategy is to reduce the problem, via Proposition 4.7, to checking that two auxiliary real polynomials h+ and h− have only real nonnegative roots, and then to determine real-rootedness by a criterion (Proposition 2.4) involving the roots of the derivative. The derived coefficient conditions use Cardano and Ferrari formulas and are tested by the author’s Magma implementation against the known root-unitary algorithm. The paper also claims to correct earlier flawed classifications by Haloui, Haloui–Singh, and Sohn.","tokens_in":23616,"tokens_out":35043,"duration_ms":417573,"significance":"If correct, the main theorems provide useful, explicit corrections to previously published descriptions of W_q(3), W_q(4), and W_q(5), with an accompanying reproducible Magma implementation. The overall derivation is self-contained and does not fit parameters to data: the inequalities come from root-location conditions, Rolle’s theorem, and radical formulas. The Appendix A proof of Proposition 2.4 is, in substance, a valid direct proof, and the stress-test concern about Lemma A.2(VI) does not actually land: since f′ has only real roots, summing over real roots gives the total degree of f′, so S(−∞,+∞)=1 forces all roots of f to be real. However, as printed, Section 3 contains a serious sign error in a load-bearing proposition, so the manuscript cannot be accepted in its current form.","major_comments":[{"comment":"The stated conditions for real nonnegative roots are sign-reversed. For a quadratic a2x^2+a1x+a0 with a2>0, the correct condition is a1≤0 and 0≤a0≤a1^2/(4a2); the polynomial x^2−5x+6 has roots 2 and 3 and yet violates the printed condition a1≥0. For the cubic f(x)=a3x^3+a2x^2+a1x+a0 with a3>0, the correct condition is a2≤0, 0≤a1≤a2^2/(3a3), and a0≤0; the polynomial x(x−1)(x−2)=x^3−3x^2+2x has nonnegative roots but violates the printed condition a2≥0. This is not a local typo: the proof of Proposition 3.3 invokes Proposition 3.2 on f′ and obtains condition (i) a3≤0, which is the opposite of what the printed Proposition 3.2 would give (3a3≥0). The main theorems appear to use the corrected signs, but the intermediate statements and the proof chain in Section 3 must be repaired before the paper is publishable.","section":"Section 3, Propositions 3.1 and 3.2"},{"comment":"Relatedly, the derivation of Proposition 3.3 from Proposition 3.2 cannot be followed as written. After substituting f′(x)=4a4x^3+3a3x^2+2a2x+a1 into the printed Proposition 3.2, one would obtain a3≥0, contradicting condition (i) of Proposition 3.3. The author should restate Propositions 3.1 and 3.2 with the correct inequalities and re-derive Prop 3.3 and the main theorems from the corrected statements. As it stands, the proof of the central characterization is internally inconsistent.","section":"Section 3, Proposition 3.2 and proof of Proposition 3.3"},{"comment":"Because Main Theorems A–C are deduced from Propositions 3.2, 3.3, and 3.8, the sign errors in Section 3 affect the proof of record for all three main theorems. The final inequalities appear to be consistent with the corrected sign convention, and I verified several representative examples, but the manuscript must clearly state the corrected propositions and confirm that the subsequent translations into the (a)–(i) inequalities are unchanged. In particular, the authors should re-check the proofs of Propositions 3.3 and 3.8 and the coefficient substitutions in Sections 5–7 against the corrected root conditions.","section":"Section 5–7"}],"minor_comments":[{"comment":"The perturbation argument in the short proof is only sketched: it asserts without proof the existence of a perturbation preserving the number of real roots, making the derivative roots distinct, and making the inequalities strict. Since the direct Appendix A proof is available, the author should either expand this argument or explicitly designate the appendix as the proof of record.","section":"Section 2, proof of Proposition 2.4"},{"comment":"The combinatorial step leading to equation (15) is correct but too compressed. In particular, the claim that a −1 in the sequence (S_{k+1},…,S_1) produced by case (i) is never balanced by a +1 deserves one or two sentences of explanation, since the balancing argument in case (iii) is the key to the proof.","section":"Appendix A, proof of Proposition A.3"},{"comment":"The one-line proof of Lemma A.2(VI) is terse. It would be clearer to state explicitly that, because f′ has only real roots, the contribution of f′ over all real roots is K−1; hence S(−∞,+∞)=1 forces the real roots of f to account for all K roots, so f has no non-real roots.","section":"Appendix A, Lemma A.2(VI)"},{"comment":"The real-root classifications are described as “mutually exclusive,” but within Case (I) of Main Theorem B, for example, a polynomial with both (x+√q)^2 and (x−√q)^2 factors can be represented in either form when the auxiliary factor h0 itself has a real root. The wording should either require h0 to have no real roots in the subcases or state that the listed forms are a cover rather than a disjoint partition.","section":"Main Theorems B and C"},{"comment":"There are several typographical and wording issues: in the proof of Proposition 4.7, “0 ∈ 0 ∈ S−” should read “0 ∈ S−”; in the introduction, “acting of the ℓ-adic Tate module” should be “acting on”; and the title of Section 2 (“Polynomials with only real positive roots”) does not match its content, which treats arbitrary real roots as well.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The sign errors in Propositions 3.1 and 3.2 are severe enough that the current version cannot be accepted, but they appear to be repairable and the final main theorems are plausible. I would ask the author to correct the statements, re-run the derivations in Section 3 and Sections 5–7, and add explicit verification of the corrected propositions. Given the paper’s reliance on long algebraic reductions, a computer-algebra check of the corrected propositions would be appropriate before resubmission."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Main take: the corrections are real and the core criterion holds up; the real risk is the compressed algebra, not the root-counting lemma. I'd send it to review and ask for the long computations to be opened up.\n\nWhat's new: Main Theorems A, B, and C give explicit coefficient inequalities for W_q(3), W_q(4), and W_q(5), fixing genuine mistakes in Haloui, Haloui–Singh, and Sohn, and they pin down exactly when the polynomial has real roots, which the older statements missed or misstated. The derivative-root criterion (Proposition 2.4) is a clean tool, and having two proofs is a good sign. The real-root classification is a useful addition, not just a correction.\n\nWhere I'd push back: the stress-test says Lemma A.2(VI) is unproved and load-bearing. I don't think that holds up. Since f' is assumed to have only real roots, S(−∞, +∞) is just (# real roots of f) − (K−1); that equals 1 iff all roots of f are real. The proof is one line because the statement is one line. The combinatorial argument around (15) is compressed, but it's correct — a +1 can only be created next to a −1, and a failed SIGN leaves an unmatched −1. Terse, not wrong.\n\nThe actual soft spot is the repeated 'a computation shows' in Sections 5–7, especially the identities u2±, u4±, u3+ = −u3−, and the relation between h− and h+ in Section 7. Those are exactly where a sign typo or a missing 50q term could hide, and the paper's own history shows that this literature is where such typos live. I'd want the referee to request a fully expanded derivation of those identities, or a short computer-algebra certificate. The Magma tests are a good sanity check, but they're downstream and shouldn't substitute for the algebra.\n\nBottom line: the central argument is sound as far as I can see, the prior corrections are real, and the paper belongs in the literature. It should go to a serious referee with a request to verify the long algebraic steps; conditional accept is the right outcome.","headline":"The corrections to W_q(3,4,5) look right and the derivative-root criterion holds up; the long algebraic simplifications in Sections 5–7 are the part a referee should check.","tokens_in":24187,"tokens_out":5641,"would_cite":true,"duration_ms":63210,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14K15","12D10","26C10"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the coefficient inequalities in Main Theorems A, B, and C are necessary and sufficient for q-Weil polynomials of degrees 6, 8, and 10, correcting earlier flawed descriptions.","keywords":["q-Weil polynomials","abelian varieties over finite fields","Honda–Tate classification","hyperbolic polynomials","real-root criterion","isogeny classes","small-degree polynomial inequalities"],"falsifier":"The central claim would be settled by finding a monic integer polynomial satisfying inequalities $(a)$–$(e)$ of Main Theorem A whose complex roots are not three conjugate pairs of absolute value $\\sqrt{q}$, or by finding a real polynomial whose derivative roots satisfy the alternating sign condition of Proposition 2.4 while the polynomial itself has a non-real root.","tokens_in":23211,"feed_emoji":"🔢","tokens_out":8412,"duration_ms":96122,"temperature":0.7,"pith_summary":"This paper corrects the published descriptions of $q$-Weil polynomials of degree $2g$ for $g=3,4,5$, the characteristic polynomials of Frobenius endomorphisms of abelian varieties of those dimensions over a finite field. The Honda–Tate classification says that these polynomials classify isogeny classes, so a concrete description of them is the practical route to enumerating those classes. The main theorems give necessary and sufficient coefficient inequalities for membership in $W_q(3)$, $W_q(4)$, and $W_q(5)$, and they also identify exactly when such a polynomial has a real root. Earlier descriptions for these three dimensions contained mistakes of varying severity, and the paper replaces them with corrected statements. The arguments build on a real-root criterion for polynomials of low degree.","feed_headline":"Corrected inequalities classify q-Weil polynomials in degrees 6, 8, 10","feed_subtitle":"New necessary and sufficient conditions fix the earlier descriptions and also detect every polynomial with a real root.","key_machinery":"The load-bearing object is Proposition 2.4, a hyperbolicity criterion: for a real polynomial $f(x)$ of degree $K>1$ with positive leading coefficient, if the roots of $f'(x)$ are real and sorted as $\\beta_{K-1}\\le\\cdots\\le\\beta_1$, then all roots of $f(x)$ are real exactly when $f(\\beta_i)\\le 0$ for odd $i$ and $f(\\beta_i)\\ge 0$ for even $i$. Proposition 4.7 then converts the $q$-Weil condition into the requirement that two sets $S^+$ and $S^-$ of $g$ real numbers formed from the coefficients are subsets of $\\mathbb{R}_{\\ge0}$. The main theorems follow by applying the criterion recursively to the cubic, quartic, and quintic polynomials obtained this way, using radical formulas for roots up to degree four.","core_discovery":"The central claim is that for $g=3,4,5$, a monic integer polynomial of the palindromic form $h(x)=x^{2g}+a_1x^{2g-1}+\\cdots+a_gx^g+\\cdots+q^g$ is a $q$-Weil polynomial if and only if its coefficients satisfy the explicit inequality lists $(a)$–$(e)$, $(a)$–$(g)$, and $(a)$–$(i)$ of Main Theorems A, B, and C, respectively. Each theorem also characterizes the boundary case of real roots: a real root occurs exactly when one of certain inequalities is an equality, and the resulting polynomials are listed explicitly, for example $(x^2-q)^2h_0(x)$ when $q$ is not a square. The proof route is to reduce the $2g$-degree condition to checking that two degree-$g$ polynomials built from the same coefficients have only real non-negative roots, and then to apply derivative-root criteria recursively.","pith_inferences":["If the corrected inequalities are right, previously published counts or tables derived from the flawed descriptions may need revision for square-$q$ cases, since the paper shows the old statements missed real-root polynomials there.","The derivative-root criterion of Proposition 2.4 is independent of Weil polynomials and could be reused for other hyperbolicity problems, such as determining when a totally real algebraic integer has all conjugates real.","The same recursion is unlikely to produce explicit coefficient inequalities for $g=6$: the derivative roots of a degree-6 polynomial generally involve solving a quintic, which has no radical formula, so a different method would be needed beyond these dimensions.","A natural stress test would be to implement the three theorems independently and compare against exhaustive root-unitary search over a wider range of prime powers, concentrating on the equality cases where real roots appear."],"forward_implications":["For every prime power $q$, the corrected inequalities give a finite, explicit test for membership in $W_q(3)$, $W_q(4)$, and $W_q(5)$, so these sets can be enumerated by coefficient search without computing or factoring roots.","The equality conditions describe exactly which of these $q$-Weil polynomials have real roots, completing the classification in the square-$q$ cases that earlier statements missed.","The same inequalities yield algorithms for all dimensions up to five, and the paper reports that their implementation matched the output of the exhaustive search algorithm over the tested prime powers.","Since the Honda–Tate image condition depends on factorizations and valuations, these corrected descriptions provide the coefficient-level input needed to list actual characteristic polynomials of abelian varieties of dimensions 3–5."],"supporting_citations":[{"why":"It supplies the description of $W_q(3)$ that Main Theorem A revises.","marker":"[5]"},{"why":"It supplies the description of $W_q(4)$ that Main Theorem B revises.","marker":"[6]"},{"why":"It supplies the description of $W_q(5)$ that Main Theorem C revises.","marker":"[24]"},{"why":"It establishes that the characteristic polynomial determines the isogeny class, making $W_q(g)$ the right classification object.","marker":"[26]"},{"why":"It supplies the converse construction showing which Weil polynomial data are realized by abelian varieties, underpinning the classification.","marker":"[9]"},{"why":"It identifies mistakes in the earlier descriptions and provides the computational context for testing the corrected criteria.","marker":"[4]"},{"why":"It provides the exhaustive root-unitary search algorithm used as a benchmark for the paper's implementation.","marker":"[13]"},{"why":"It contains an attempted correction of the degree-8 case that still has a small error, motivating the present statement.","marker":"[3]"}],"fun_headline_variants":["New inequalities fix q-Weil classification for degrees 6, 8, 10","Explicit criteria correct q-Weil polynomial lists for g up to 5","Corrected root criteria determine q-Weil polynomials in degrees 6, 8, 10","Errors fixed: q-Weil polynomial classification for degrees 6, 8, 10"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole proof rests on Proposition 2.4, which claims that a real polynomial has only real roots precisely when the values at the roots of its derivative alternate in sign by parity; if that criterion failed, the coefficient inequalities in all three main theorems would not follow.","fun_headline_variants_meta":{"raw":{"variants":["New inequalities fix q-Weil classification for degrees 6, 8, 10","Explicit criteria correct q-Weil polynomial lists for g up to 5","Corrected root criteria determine q-Weil polynomials in degrees 6, 8, 10","Errors fixed: q-Weil polynomial classification for degrees 6, 8, 10"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001371,"raw_usage":{"total_tokens":5516,"prompt_tokens":862,"completion_tokens":4654,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":478,"completion_tokens_details":{"reasoning_tokens":4562}},"tokens_in":478,"tokens_out":4654,"duration_ms":41342,"temperature":1.0,"reasoning_tokens":4562,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T12:19:11.991223+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"The central claim would be settled by finding a monic integer polynomial satisfying inequalities $(a)$–$(e)$ of Main Theorem A whose complex roots are not three conjugate pairs of absolute value $\\sqrt{q}$, or by finding a real polynomial whose derivative roots satisfy the alternating sign condition of Proposition 2.4 while the polynomial itself has a non-real root.","supporting_citations":[{"cited_title":"The characteristic polynomials of abelian varieties of dimensions 3 over finite fields","cited_arxiv_id":null,"evidence_quote":"It supplies the description of $W_q(3)$ that Main Theorem A revises."},{"cited_title":"The characteristic polynomials of abelian varieties of dimension 4 over finite fields","cited_arxiv_id":null,"evidence_quote":"It supplies the description of $W_q(4)$ that Main Theorem B revises."},{"cited_title":"The Newton polygons of the characteristic polynomials for abelian varieties of dimension 5 over finite fields","cited_arxiv_id":null,"evidence_quote":"It supplies the description of $W_q(5)$ that Main Theorem C revises."},{"cited_title":"Endomorphisms of abelian varieties over finite fields","cited_arxiv_id":null,"evidence_quote":"It establishes that the characteristic polynomial determines the isogeny class, making $W_q(g)$ the right classification object."},{"cited_title":"Isogeny classes of abelian varieties over finite fields","cited_arxiv_id":null,"evidence_quote":"It supplies the converse construction showing which Weil polynomial data are realized by abelian varieties, underpinning the classification."},{"cited_title":"Isogeny classes of abelian varieties over finite fields in the LMFDB","cited_arxiv_id":null,"evidence_quote":"It identifies mistakes in the earlier descriptions and provides the computational context for testing the corrected criteria."},{"cited_title":"root-unitary","cited_arxiv_id":null,"evidence_quote":"It provides the exhaustive root-unitary search algorithm used as a benchmark for the paper's implementation."},{"cited_title":"Commutative endomorphism rings of simple abelian varieties over finite fields","cited_arxiv_id":null,"evidence_quote":"It contains an attempted correction of the degree-8 case that still has a small error, motivating the present statement."}],"review_version":1}