{"id":"fcb9dc2e-1705-4018-a76e-8a394003f69b","arxiv_id":"2608.11792","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"partial","parameter_count":0,"one_line_summary":"Pair-zero and degree conditions admit extra six-dimensional hook solutions at seven points and a two-dimensional plane at eight; Bose symmetry and one normalized physical boundary condition single out the Hodges numerator.","lead":"This paper shows that the MHV gravity numerator is not fixed by its degree and zero locations alone: extra hook solutions appear at seven points, and a plane remains at eight. Physical constraints, including Bose symmetry and one normalized boundary limit, select the known Hodges formula.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"All-multiplicity injectivity rests on an unproved domain/prime-divisor property of Q_{n,Q}; if <ab> is not prime/radical, Lemma 6.1 and Theorem 6.2 fail.","rationale":"The reader's weakest assumption identifies exactly this: the all-multiplicity theorem assumes Q_{n,Q} is an integral domain at every multiplicity so that vanishing on <ab>=0 forces divisibility by <ab> and the angle factors can be extracted successively. I agree this is the most load-bearing assumption. It underpins Theorem 6.2, Corollary 6.3, and the all-multiplicity physical uniqueness claim, Theorem 6.4. The paper does state, in Section 3.1, that Q_{n,k} is the multihomogeneous coordinate ring of Fl(2,n−2;k^n); that statement, if made explicit with the standard straightening-law consequences, would justify the domain property for all n. But the paper never connects this to the divisor argument in Lemma 6.1, and the only place the domain property is explicitly invoked is Appendix D.3 for n=7. The eight-point calculations and the three physical selectors are well-structured and mutually consistent, and the exact rank/ideal-membership computations are substantial independent evidence. The Lean verification covers only the finite-dimensional consequence layer, so it does not fill this gap. Because the same conditional verdict already accounts for this assumption and for reproducibility issues, I do not change the reader's verdict.","tokens_in":26813,"tokens_out":44576,"duration_ms":504187,"concrete_test":"Provide a proof or explicit citation that for every n≥5 the ring Q_{n,Q} is the normal integral multihomogeneous coordinate ring of Fl(2,n−2;k^n) and that each Plücker divisor (<ab>) is prime/radical. As a computational check, use Macaulay2 or Singular for n=8 to verify that Q_{8,Q} is a domain and that Q_{8,Q}/(<12>) is reduced and irreducible. If the ring is a domain but some <ab> is not prime, recompute the successive factor extraction in Theorem 6.2; if the flag-variety identification fails, the injectivity of all marked restrictions is in doubt.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The all-multiplicity injectivity statement (Theorem 6.2, then Corollary 6.3 and Theorem 6.4) rests on Lemma 6.1. Its proof asserts that membership in K_ab(c) forces vanishing on D_ab = V(<ab>), and then that f = <ab>g 'because this is a reduced Cartier divisor'; later, distinct angle factors are said to be extractable successively. These steps require Q_{n,Q} to be an integral domain and each <ab> to be a prime/radical Schubert divisor, with no zero-divisor obstruction when extracting several factors. The paper states the domain property only for n=7, in Appendix D.3 ('Since Q_7,Q is a domain'), and does not prove it for general n. Section 3.1 identifies Q_{n,k} with the multihomogeneous coordinate ring of Fl(2,n−2;k^n), which would supply the domain property, but the paper does not spell out that this makes each <ab> a prime divisor. If the flag-variety identification is correct, the issue is a presentation gap; if it is not, or if some <ab> fails to be prime, Lemma 6.1 collapses and the all-n uniqueness theorem is unsupported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper studies whether a tree-level MHV gravity numerator is determined by its degree and by vanishing on the simultaneous pair loci <ij>=[ij]=0 for every pair, in the reduced form A_n=N_n/D_n with D_n=∏<ij>. The authors replace the original quotient-ring calculation by a flag-variety standard-monomial basis and an S_n-isotypic decomposition. They prove at seven points that W_7,ℚ ≅ S^{(2,1^5)}⊕S^{(1^7)}, so the pair-ideal conditions leave a six-dimensional hook in addition to the alternating Hodges line; this disproves the original one-dimensional algebraic uniqueness conjecture. Bose symmetry selects the sign component and gives the seven-point Hodges line. At eight points they compute (W_8,ℚ)_sgn = ℚA⊕ℚB, a two-dimensional plane, and show that same-helicity BCFW O(z^{-2}) falloff, normalized ++ collinear factorization, and the leading soft coefficient each impose the same linear condition 7c_A-6c_B=0, selecting the Hodges line. They also prove an all-multiplicity marked-collinear injectivity theorem and a recursive physical uniqueness theorem within the fixed-common-denominator ansatz, conditional on the standard factorization and soft theorems as external inputs. The finite-dimensional calculations are supported by exact integer/rational arithmetic, modular minors under two primes, explicit ideal-reduction lifts, and a Lean-verified consequence layer.","tokens_in":26924,"tokens_out":15109,"duration_ms":176112,"significance":"If the results hold, they cleanly separate algebraic, Bose-compatible, and physical uniqueness for MHV gravity numerators: the pair-ideal conditions are not sufficient, Bose symmetry removes the hook at seven points but not at eight, and one normalized boundary condition restores uniqueness. The paper's computational architecture is a genuine strength: the flag-variety tableau basis, S_n-resolved restriction blocks, exact rank bounds from modular minors, explicit 60-coefficient hook polynomial, global ideal-membership proofs with generator lifts, and the Lean formalization of the finite-dimensional consequence layer are all unusually well documented. The eight-point demonstration that BCFW scaling, normalized collinear factorization, and the leading soft coefficient reduce to the same linear condition is a valuable cross-check. The main all-multiplicity theorem, however, is conditional on a geometric domain/prime-divisor property that is stated only for n=7; filling that gap is necessary before the advertised all-n uniqueness statement is fully supported.","major_comments":[{"comment":"The all-multiplicity injectivity chain rests on an unproved domain/prime-divisor property. The proof of Lemma 6.1 passes from vanishing on the marked loci to f=<ab>g with the words \"because this is a reduced Cartier divisor,\" and Theorem 6.2 then says that the distinct angle-bracket factors may be extracted successively. Both steps require Q_{n,ℚ} to be an integral domain, each <ab> to be a prime divisor, and the successive quotient by several <ab>'s to remain a domain (or at least a regular-sequence-type property). The paper states the domain property only at n=7, in Appendix D.3 (\"Since Q_7,ℚ is a domain\"), and gives no proof or citation for general n. If the identification of Q_{n,k} with the multihomogeneous coordinate ring of Fl(2,n-2;k^n) in Section 3.1 is intended to supply this, then the authors should state it explicitly as a proposition — including that this ring is a domain and that the Schubert divisors V(<ab>) are prime with prime successive intersections — with a proof or a precise reference. As written, Theorem 6.2, Corollary 6.3, and Theorem 6.4 are unsupported for n>7. This is a presentation gap in a load-bearing step rather than a detected contradiction, but it must be fixed before the all-multiplicity claim is accepted.","section":"Lemma 6.1, Theorem 6.2; Section 3.1; Appendix D.3"},{"comment":"The all-multiplicity physical uniqueness theorem is conditional on the universal ++ splitting law and on the five-point Hodges seed, and the paper is honest about this. However, the main-text statement of Theorem 6.4 should explicitly list all hypotheses in the theorem environment: the fixed-common-denominator ansatz, the fixed multidegree, the alternating character, one complete normalized ++ marked-collinear boundary, and the external factorization law. Currently these assumptions are distributed between the surrounding prose and Appendix A.2, which makes it easy to over-read the theorem as an unconditional derivation of MHV gravity from pair zeros. This is a scope-of-claim issue rather than a mathematical error, but it matters for the paper's advertised conclusion.","section":"Theorem 6.4 and Section 6.3"}],"minor_comments":[{"comment":"The multidegree d(n) is written with entries n-5,...,n-5 without specifying how many entries there are or how the ordering aligns with the Hodge convention; please state this explicitly at first use, even though Appendix A later clarifies it.","section":"Equation (2.7)"},{"comment":"There are typographical artifacts such as \"T able 1,\" and the 60-coefficient hook table in Appendix C.3 would be easier to verify if the row-major basis IDs were explained in one sentence right before the table.","section":"Table 1 and Appendix C.3"},{"comment":"The BCFW selection uses one displayed kinematic point; the argument is logically sufficient because the Hodges line is known to have O(z^{-2}) falloff, but the text should state explicitly that [z^6]N is a nonzero linear functional on U8 whose kernel is the Hodges line, so the reader does not wonder whether the single-point evaluation is meant to prove a global coefficient identity.","section":"Appendix G.1"},{"comment":"The sentence \"Since Q_7,ℚ is a domain\" appears without justification; even if this follows from the flag-variety realization, give a citation or a one-line argument so that the localization step at n=7 is self-contained.","section":"Appendix D.3 and Section 5"}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe short version: this paper refutes a published conjecture, and the refutation looks right. At seven points the pair-ideal intersection is seven-dimensional (hook plus sign line), not one-dimensional; at eight points Bose symmetry leaves a two-dimensional plane, and three independent physical selectors — same-helicity BCFW scaling, normalized collinear residue, and the leading soft coefficient — all cut out the Hodges line. These are genuinely new numbers: the W7 decomposition, the explicit 60-coefficient hook, and the eight-point alternating sector.\n\nWhat the paper does well: all the load-bearing algebra is exact. Ranks are certified by nonzero minors over two primes, ideal membership by explicit reductions with lifts to Plücker and momentum-conservation generators, and the finite-dimensional consequence layer is checked in Lean. The authors are unusually disciplined about scope: they say plainly what the Lean certificate does and does not cover, and they do not dress up the AI-assisted exploration as evidence.\n\nWhere the soft spots are: the all-multiplicity injectivity theorem rests on Q_{n,Q} being a domain and the pair loci <ab>=0 being prime divisors, but the paper proves the domain property only at n=7, in Appendix D.3. The stress-test note is right about this. I do not think it is fatal, because Section 3.1 identifies Q_{n,k} with the multihomogeneous coordinate ring of Fl(2,n−2;k^n), which is a domain, and the <ab> loci are Schubert divisors. The identification needs to be spelled out in the proof of Lemma 6.1; as written, the all-n theorem is conditional on an unstated geometric fact. The external physics (BCFW O(z^{-2}), universal ++ splitting, Weinberg soft theorem) is standard, and I do not count it as a flaw. Reproducibility points at an unpinned repo handle rather than a commit hash — minor.\n\nWho gets value: anyone working on amplitude bootstrap, numerator reconstruction, or zero-locus constraints in gauge and gravity amplitudes. I would send this to a serious referee. The seven- and eight-point results are worth citing, and the all-multiplicity gap is fixable without changing the main conclusions.","headline":"Refutes the Ref. [2] uniqueness conjecture with exact seven- and eight-point calculations; the all-multiplicity theorem has a fixable presentation gap.","tokens_in":27577,"tokens_out":3084,"would_cite":true,"duration_ms":29778,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"At seven and eight points, MHV gravity numerators are underdetermined by degree and pair-zero conditions alone; Bose symmetry and one normalized boundary condition restore uniqueness.","keywords":["scattering amplitudes","MHV gravity","numerator bootstrap","pair-ideal intersection","flag varieties","semistandard tableaux","Bose symmetry","BCFW scaling"],"falsifier":"Under the $[1,2\\rangle$ shift at the rational eight-point kinematics recorded in section G.1, compute the $z^6$ coefficient of $N/D_8$ for $N=7A-6B$: the paper predicts a nonzero value proportional to $7c_A-6c_B$, so a vanishing coefficient would overturn the large-$z$ selection. Independently, at $n=9$, search for a nonzero alternating fixed-degree numerator that lies in every pair ideal and in one marked collinear ideal $K_{ab}(2)$; theorem 6.2 predicts no such element exists.","tokens_in":26477,"feed_emoji":"🌌","tokens_out":14404,"duration_ms":148115,"temperature":0.7,"pith_summary":"Tree-level maximally-helicity-violating (MHV) gravity amplitudes are represented as a polynomial numerator divided by a fixed product of angle brackets, and this paper asks whether the numerator's degree together with its vanishing on every pairwise locus $\\langle ij\\rangle = [ij] = 0$ determines the amplitude. The answer is no in general: at seven points the algebraic solutions form a seven-dimensional space $S^{(2,1^5)}\\oplus S^{(1^7)}$, consisting of the Hodges numerator line plus a six-dimensional hook representation. Requiring Bose symmetry (an alternating numerator) removes the hook and leaves the Hodges line. At eight points Bose symmetry leaves a two-dimensional alternating plane, and three physical conditions—the large-$z$ falloff, normalized collinear factorization, and the leading soft coefficient—impose the same linear condition and select the Hodges line. Within the fixed-common-denominator ansatz, the paper proves that one complete normalized marked-collinear boundary determines the numerator up to scale at any multiplicity.","feed_headline":"Pair-zero conditions do not pin down MHV gravity numerators","feed_subtitle":"At seven and eight points, Bose symmetry plus one normalized boundary limit selects the standard gravity amplitude.","key_machinery":"The load-bearing construction is the flag-variety realization of the spinor-helicity quotient: angle brackets become Plücker columns of height two, square brackets become complementary columns of height $n-2$, and the pair ideals $\\langle ij\\rangle,[ij]$ become column-restriction conditions in the coordinate ring of $\\mathrm{Fl}(2,n-2)$. A standard-monomial basis indexed by semistandard tableaux of shape $\\Lambda_n=(A_n+B_n,A_n+B_n,B_n,\\ldots,B_n)$ makes the ansatz finite-dimensional, and Young projectors decompose the simultaneous restriction map into irreducible $S_n$ blocks, shrinking the seven-point calculation from 65,870 tableaux to fifteen blocks whose largest multiplicity space has dimension 475. The same machinery supports the eight-point character calculation and the marked collinear ideals $K_{ab}(\\alpha)=\\langle\\langle ab\\rangle,\\langle ia\\rangle-\\alpha\\langle ib\\rangle\\rangle$, whose joint injectivity—proved by successively extracting distinct angle-bracket divisors—is the engine of the all-multiplicity uniqueness theorem.","core_discovery":"The central discovery is that algebraic and physical uniqueness of MHV gravity numerators separate at higher multiplicity. Over $\\mathbb{Q}$, the seven-point pair-ideal intersection at the target degree is $W_{7,\\mathbb{Q}}\\simeq S^{(2,1^5)}\\oplus S^{(1^7)}$, so the degree-and-zero conditions admit six hook directions in addition to the Hodges line and the all-multiplicity conjecture that held at five and six points fails. Bose symmetry is an independent constraint that retains only the sign component, giving $W^{\\mathrm{Bose}}_{7,\\mathbb{Q}}=\\mathbb{Q}\\,N^{\\mathrm{Hodges}}_7$. At eight points the Bose-compatible sector is a two-dimensional plane $QA\\oplus QB$, and the conditions of $O(z^{-2})$ large-$z$ behavior, normalized $++$ collinear factorization, and the leading positive-helicity soft coefficient each reduce to the same linear condition $7c_A-6c_B=0$, whose solution is the Hodges line $Q(6A+7B)=QN^{\\mathrm{Hodges}}_8$. For arbitrary $n$, an alternating fixed-degree numerator is injectively determined by one marked collinear restriction, so together with universal $++$ splitting and the five-point seed it is fixed up to normalization inside the common-denominator ansatz.","pith_inferences":["A natural next test is $n=9$: computing the full $S_n$-resolved pair-ideal kernel would show whether hook-type algebraic directions grow with multiplicity, and whether the physical selectors still agree on the Bose-compatible sector.","The all-multiplicity theorem assumes the bracket ring has no zero divisors at every $n$; locating the first multiplicity where this fails, or exhibiting a nonzero alternating numerator in the kernel of one marked restriction, would mark the boundary of the method.","Because the large-$z$, collinear, and soft selectors coincide exactly on the eight-point plane, a plausible extension is that every normalized boundary condition fixing the leading soft or collinear coefficient defines the same linear functional on the Bose-compatible sector, a statement not proved beyond eight points."],"forward_implications":["The original conjecture that degree plus pair-zero conditions fix the numerator at every multiplicity is false; five- and six-point uniqueness was a low-multiplicity accident.","Bose symmetry is not implied by the pair-ideal conditions, but at seven points it restores uniqueness by killing the six hook directions.","At eight points the large-$z$, collinear, and soft conditions are interchangeable selectors within the Bose-compatible plane: any one normalized boundary condition fixes the amplitude.","At arbitrary multiplicity, reconstructing the numerator reduces to evaluating one complete marked collinear boundary, provided the common-denominator ansatz and the standard splitting law hold.","Higher-multiplicity calculations can be organized by permutation sectors, so the full $S_n$ decomposition, not just the sign sector, is the natural computational target."],"supporting_citations":[{"why":"This reference defines the numerator ansatz, the target multidegree, and the pair-ideal conditions, and states the uniqueness conjecture that the seven-point calculation refutes.","marker":"[2]"},{"why":"It gives the determinant formula whose numerator is the Hodges line, the object selected by Bose symmetry and by the eight-point physical conditions.","marker":"[9]"},{"why":"It identifies the spinor-helicity variety with a flag variety and supplies the straightening law behind the standard-monomial basis.","marker":"[13]"},{"why":"It provides the standard-monomial theory used to index the tableau basis and to transport the pair restrictions into the flag realization.","marker":"[14]"},{"why":"It gives the Young-symmetrizer theorem used to decompose the restriction map into irreducible $S_n$ blocks and to separate hook from alternating components.","marker":"[21]"},{"why":"It supplies the universal gravity splitting law used as the normalized collinear boundary data in the all-multiplicity theorem and in the eight-point selection.","marker":"[17]"},{"why":"It documents the $O(z^{-2})$ large-$z$ behavior of gravity tree amplitudes used by the large-momentum selector at eight points.","marker":"[19]"},{"why":"It gives the leading soft-graviton theorem whose normalized coefficient provides the soft selector.","marker":"[20]"},{"why":"It provides the finite-field modular-rank method used to obtain exact upper bounds on the kernel dimensions.","marker":"[22]"}],"fun_headline_variants":["Bose symmetry picks Hodges line from extra gravity numerators","Pair-zero conditions leave extra MHV gravity solutions","Physical constraints single out Hodges numerator","At eight points, one linear condition restores uniqueness","Algebraic vs physical: Bose symmetry decides MHV gravity"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The all-multiplicity result rests on assuming that at every point count $n$ the ring of spinor bracket polynomials has no zero divisors, so a numerator that vanishes wherever one angle bracket vanishes must contain that bracket as a factor; the paper verifies this property only at seven points, and the physical selection statements also take the standard large-$z$ falloff, the universal $++$ splitting law, and the leading soft-graviton theorem as inputs rather than deriving them.","fun_headline_variants_meta":{"raw":{"variants":["Bose symmetry picks Hodges line from extra gravity numerators","Pair-zero conditions leave extra MHV gravity solutions","Physical constraints single out Hodges numerator","At eight points, one linear condition restores uniqueness","Algebraic vs physical: Bose symmetry decides MHV gravity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000442,"raw_usage":{"total_tokens":2293,"prompt_tokens":1051,"completion_tokens":1242,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":667,"completion_tokens_details":{"reasoning_tokens":1168}},"tokens_in":667,"tokens_out":1242,"duration_ms":10301,"temperature":1.0,"reasoning_tokens":1168,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-16T00:27:42.869653+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Under the $[1,2\\rangle$ shift at the rational eight-point kinematics recorded in section G.1, compute the $z^6$ coefficient of $N/D_8$ for $N=7A-6B$: the paper predicts a nonzero value proportional to $7c_A-6c_B$, so a vanishing coefficient would overturn the large-$z$ selection. Independently, at $n=9$, search for a nonzero alternating fixed-degree numerator that lies in every pair ideal and in one marked collinear ideal $K_{ab}(2)$; theorem 6.2 predicts no such element exists.","supporting_citations":[{"cited_title":"A geometric approach to Standard Monomial Theory","cited_arxiv_id":"math/0111054","evidence_quote":"It provides the standard-monomial theory used to index the tableau basis and to transport the pair restrictions into the flag realization."}],"review_version":1}