{"id":"17b2132a-0038-4919-8dbb-cb20b7ab9ad7","arxiv_id":"2501.18832","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Tree-level single-trace MHV Einstein-Yang-Mills amplitudes with any number of gravitons are re-expressed as sums of pure gluon Parke-Taylor amplitudes, with each graviton represented by a collinear gluon pair.","lead":"This paper gives a new formula for certain scattering amplitudes in a theory combining gluons and gravitons, rewriting them using only gluon amplitudes. If correct, it simplifies computations and may reveal hidden structure linking gauge and gravity theories.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Formula (4.1) likely overcounts: ordered root tuples plus coincident roots break the claimed one-to-one map from the forest sum (2.10), with no quotient or canonical ordering specified.","rationale":"The reader identified the load-bearing weak point as the lack of a proven bijection between the forest sum (2.10) and the collinear-splitting sum (4.1), specifically the ambiguity from coincident roots and branch exchange. My analysis sharpens this into a concrete overcounting mechanism: the ordered summation over root tuples and labeled trees counts each forest with i distinct roots i! times, and star-like forests with several gravitons attached to one gluon are counted once as a single tree and additionally as multiple trees sharing a root. The paper's own note that exchanging branches in (3.14)/(3.15) yields equivalent permutations confirms that the PT sums for different label orders are identical, so no cancellation occurs. Because the overcount factor depends on the forest topology rather than being a global normalization, the formula (4.1) cannot be correct up to an overall constant. This is a central-claim failure, not a cosmetic issue. However, I have not run the numerical check, so I recommend a conditional verdict: the paper can be accepted only if the authors either prove a bijective correspondence (e.g., by summing over unlabeled forests with a canonical insertion order) or demonstrate that the overcount cancels; otherwise the formula is incorrect as stated. The concrete test isolates a minimal case—the two-graviton star topology—where the multiplicity is unambiguous and the ratio is easy to compute.","tokens_in":8595,"tokens_out":26294,"duration_ms":250111,"concrete_test":"Compute the two-graviton MHV amplitude for N=5 gluons and M=2 gravitons at generic spinor momenta. Evaluate S via (2.10) and the full amplitude via (4.1) as written. Isolate the star topology where both gravitons a and b attach to the same gluon l: (2.10) contributes one copy of s_{la}s_{lb} times the PT sum; (4.1) contributes three copies (i=1 with the two-branch tree, and i=2 with l1=l2 and the two single-graviton trees in both label orders). If the ratio of the two coefficients is not 1, (4.1) overcounts and contradicts the SBDW formula (2.7).","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central formula (4.1) sums over ordered root tuples (l1,...,li) and labeled trees T1,...,Ti, and explicitly allows lj=lk for distinct labels. For a forest in (2.10) with i components, the same graph is generated i! times by permuting the tree labels; if two components attach to the same gluon, the graph is also generated both as one tree and as multiple trees sharing a root. Because the recursive insertion (4.2) is order-independent—as the authors themselves demonstrate via the equivalence of (3.14) and (3.15)—these multiple copies add identical PT sums and identical K products. The paper never divides by this symmetry nor fixes a canonical order, so (4.1) does not follow from (2.10) by a bijection. The proof in Section 4 is only a sketch (\"we sketch the proof\") and does not address this multiplicity. The three-graviton derivation is detailed, but the jump to arbitrary M relies precisely on the unproven uniqueness that the coincident-root allowance violates. This is not merely a cosmetic gap: wrong relative weights between different forest topologies would change the amplitude as a function of kinematics.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a general formula, Eq. (4.1), expressing tree-level single-trace MHV Einstein-Yang-Mills (EYM) amplitudes as a sum over spanning forests of pure-gluon MHV amplitudes, in which each graviton is replaced by a collinear gluon pair. After reviewing the spinor-helicity formalism and the SBDW/spanning-forest formula, the authors work out the three-graviton case in detail, explicitly rewriting the S3 factor into sums of PT factors with collinear insertions. They then state the general formula with a recursive insertion rule (4.2) and give a three-step proof sketch. They note that the formula reduces to the known one- and two-graviton results.","tokens_in":8797,"tokens_out":5035,"duration_ms":49173,"significance":"If the general formula is correct, it provides a compact and physically suggestive representation of single-trace MHV EYM amplitudes in terms of pure-gluon amplitudes, extending the collinear-gluon-pair picture to an arbitrary number of gravitons. The three-graviton example is worked out carefully, and the eikonal and Schouten identities used are standard. The paper is clearly written and the derivation is an algebraic rewriting of the established SBDW/spanning-forest formula (2.10), so the main claim is a reformulation rather than an independent computation. However, the general proof is explicitly only a sketch, and the load-bearing uniqueness of the forest/root sum is not established; this limits the present significance of the result.","major_comments":[{"comment":"The equality between (4.1) and the spanning-forest formula (2.10) is not established because the summation in (4.1) is over ordered tuples of roots l1,...,li and labeled trees T1,...,Ti, with coincident roots explicitly allowed in the three-graviton case (Section 3, after Eq. (3.16)). In a forest, each component has a unique root and roots of distinct components are distinct; a configuration with l_j = l_k corresponds to two trees sharing a vertex and hence to a single component, not to a forest. In addition, for a forest with distinct roots, permuting the labels of the trees gives the same graph i! times. Since neither a quotient by this symmetry nor a canonical ordering is specified, the right-hand side of (4.1) generically overcounts the terms in (2.10). This affects the relative weights of different topologies and would change the amplitude as a function of kinematics.","section":"Section 4, Eq. (4.1)"},{"comment":"The proof sketch only treats the insertion of the first tree T1 in detail and then asserts that trees T2,...,Ti are inserted 'in turn' by repetition of Step 2. It does not show that the set of permutations produced by the recursive rule (4.2) is independent of the order in which the trees are processed, nor that each insertion order yields the same sum of PT factors. The three-graviton example demonstrates the equivalence of (3.14) and (3.15) for two roots, but that is a single low-order case; the general case with M > 3, and especially with coincident roots, requires an argument that each forest term in (2.10) maps to exactly one term in (4.1). This is the load-bearing step of the paper.","section":"Section 4, Steps 2-3"},{"comment":"Because the proposed identity is purely algebraic, it can be checked numerically without any integration. The paper provides no cross-check for M = 3 or higher against the known formula (2.10). Such a check for a few random kinematic points would either confirm the formula and the absence of overcounting, or expose the multiplicity issue raised above. Its absence is conspicuous given that the general proof is only sketched.","section":"Section 4 (no numerical check)"}],"minor_comments":[{"comment":"The word 'formalsim' near the beginning of the section should be 'formalism'.","section":"Section 2.1"},{"comment":"The statement that 'lj and lk with distinct labels may be identical' should be reconciled with the forest interpretation in Section 2.2, where roots of different components are distinct; the present phrasing is confusing and may itself indicate the source of the overcounting.","section":"Section 3, after Eq. (3.16)"},{"comment":"The claim that the formula reduces to the known results for one and two gravitons is not demonstrated; a brief verification or a reference to an explicit derivation would be helpful.","section":"Introduction"},{"comment":"The word 'gravitions' should be 'gravitons'.","section":"Section 4"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is a clearly written note, but the central unsolved issue is the uniqueness of the forest/root sum in Eq. (4.1). If the authors can fix the overcounting by specifying a canonical ordering or by summing over unlabeled forests with distinct roots, and provide a numerical check for M=3 or M=4, the paper could be suitable for JHEP. The novelty is limited because the result is an algebraic consequence of the known SBDW formula, but the collinear-gluon-pair representation is a useful packaging. The paper should not be accepted in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the three-graviton case is solid; the general formula (4.1) is not established as written because it sums over ordered root tuples and labeled trees, and the authors' own sketch doesn't fix the resulting i! redundancy. The overcounting concern in the stress-test note is real.\n\nWhat's new: Ref [18] gave the collinear-splitting form for one and two gravitons; this paper states a candidate for M gravitons. The eikonal rewrite in Section 3 is careful and correct as far as it goes. The idea of expressing each graviton as a collinear pair and writing the MHV amplitude as a sum of gluon MHV amplitudes is attractive, and the three-graviton derivation is a legitimate working example.\n\nWhere it bends: The jump from the spanning forest formula (2.10) to (4.1) is not a bijection. In (2.10) a forest is an unordered collection of trees rooted on gluons. In (4.1) you sum over ordered tuples l1,...,li and labeled trees T1,...,Ti, and you allow lj=lk. For a given forest with i components, the same graph appears i! times by permuting the labels, and when two roots coincide you also get collisions between different tree structures. The authors demonstrate order-independence for the two-root example ((3.14) vs (3.15)) by summing over all l1,l2, but that doesn't show each order contributes once; it shows the two sums are equal. The general formula, as written, has no quotient by the symmetric group and no canonical ordering. Unless the sum happens to be invariant under all label permutations, which is not proven, (4.1) overcounts. The proof in Section 4 is explicitly a sketch (their words) and doesn't address this multiplicity. The three-graviton case is worked out in full, so it may well be correct; the general all-multiplicity claim needs either a proof that the label-permuted terms are identical in total, or a reformulation as a sum over unordered forests.\n\nAlso, there is no numerical cross-check. For a formula of this type, a few random momentum evaluations would have caught the overcount immediately. The lack of such a check makes the unproven uniqueness harder to forgive.\n\nBottom line: this is a small-step work, honestly labeled by the authors. The three-graviton result is useful and likely correct; the general formula is a conjecture, not a theorem, and as stated it is probably wrong. It deserves a serious referee because the core idea is worth developing, but the referee should have the liberty to demand a fix for the combinatorics before publication.","headline":"Three-graviton case is solid; the general formula (4.1) likely overcounts ordered forests, so the main claim is unproven as written.","tokens_in":9348,"tokens_out":3347,"would_cite":false,"duration_ms":36378,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Tree-level MHV Einstein-Yang-Mills amplitudes can be rewritten as sums of pure-gluon amplitudes, with each graviton split into a collinear gluon pair.","keywords":["Einstein-Yang-Mills amplitudes","MHV amplitudes","single-trace amplitudes","collinear gluon pairs","spanning forests","Parke-Taylor formula","spinor-helicity formalism","SBDW formula"],"falsifier":"Compute the right-hand side of Eq. (4.1) for a four-graviton, four-gluon MHV amplitude at a generic kinematic point and compare it with the directly evaluated left-hand side obtained from the spanning-forest expansion (2.10) or from the CHY formula; any mismatch, in particular from forests whose exchanged branches produce the same permutation, would falsify the claimed identity.","tokens_in":8352,"feed_emoji":"⚛️","tokens_out":8670,"duration_ms":74716,"temperature":0.7,"pith_summary":"This paper seeks to establish a general graviton-as-collinear-gluon-pair formula for tree-level single-trace maximally-helicity-violating (MHV) amplitudes in Einstein-Yang-Mills theory. The claimed result, Eq. (4.1), writes the amplitude with gluons $1,\\ldots,N$ and any number of gravitons as a sum, over spanning forests planted at gluon positions, of a product of kinematic factors $K(T)$ times a pure-gluon Parke-Taylor amplitude $\\mathrm{PT}(1,\\rho(l_1,\\ldots,l_i),N)$. Each graviton $n$ becomes two collinear gluons $n$ and $\\tilde n$ carrying the same momentum and helicity, inserted inside the gluon ordering according to the tree structure. If the formula is correct, graviton degrees of freedom in MHV EYM amplitudes reduce to a purely combinatorial insertion into gluon amplitudes, recovering the known one- and two-graviton cases and extending them to arbitrary graviton number. The three-graviton case is derived in full detail and the general proof is sketched by recursive insertion.","feed_headline":"Every graviton becomes a collinear gluon pair in MHV EYM","feed_subtitle":"Tree-level MHV Einstein-Yang-Mills amplitudes reduce to sums of pure-gluon Parke-Taylor amplitudes.","key_machinery":"The machinery that carries the argument is the spanning-forest expansion of the SBDW formula, together with the eikonal identity of spinor-helicity formalism. Equation (2.10) writes the factor $S(H;G)$ in the MHV amplitude as a sum over forests whose roots are gluons and whose edges are dressed by $\\psi_{ab}=[ab]\\langle a\\xi\\rangle\\langle a\\eta\\rangle/(\\langle ab\\rangle\\langle b\\xi\\rangle\\langle b\\eta\\rangle)$. Using the eikonal identity (2.5), each $\\psi$ factor is converted into a sum over positions where an inserted collinear gluon pair splits the Parke-Taylor denominator, and the recursive rule (4.2) organises these insertions into a permutation $\\rho(l_1,\\ldots,l_i)$ for every tree. The named object at the centre is the collinear gluon pair: a graviton $n$ splits into two gluons $n$ and $\\tilde n$ with identical momentum and helicity, one placed to the left and one to the right of the root gluon or of the previously inserted pair. Each edge of a tree contributes a Mandelstam invariant $s_{ab}$, so the full forest weight is a product of kinematic invariants times a single Parke-Taylor factor.","core_discovery":"The central claim is that the $(g^-g^-)$ single-trace MHV amplitude satisfies $$ A(1,\\ldots,N|H)\\sim \\sum_{l_1,\\ldots,l_i\\in G}\\;\\sum_{\\text{spanning forests }\\{T_1,\\ldots,T_i\\}} K(T_1)\\cdots K(T_i)\\;\\mathrm{PT}\\bigl(1,\\rho(l_1,\\ldots,l_i),N\\bigr), $$ where $G$ is the gluon set, each tree $T_j$ is planted at a gluon $l_j\\in G$ (distinct labels may point to the same gluon), $K(T_j)=\\prod_{ab\\in E(T_j)}s_{ab}$ is the product of Mandelstam invariants along the tree's edges, and the permutation $\\rho(l_1,\\ldots,l_i)$ is defined recursively by Eq. (4.2): the left part of the previous permutation receives the tree's gluon insertions $\\sigma_{T_k}$, the right part receives the reversed set $\\tilde\\sigma_{T_k}^T$. Each graviton $n_a$ is thereby replaced by a pair of collinear gluons $n_a,\\tilde n_a$ with the same momentum and helicity, so the original $N$-gluon, $M$-graviton amplitude is expressed as a combination of $N+2M$-point pure-gluon MHV amplitudes. The authors explicitly verify the formula for three gravitons and show that the one- and two-graviton cases reduce to known results; they also state the straightforward modifications for the $(h^-,g^-)$ helicity configuration.","pith_inferences":["If the representation is unique, it implies nontrivial identities among pure-gluon MHV amplitudes, because the same permutation can arise from different spanning forests and the total amplitude must be independent of those choices.","The collinear-pair picture may extend beyond single-trace MHV amplitudes, for instance to double-trace amplitudes or to other helicity configurations, where the same combinatorial insertion might hold with modified dressing factors; the paper lists double-trace amplitudes as future work.","A direct numerical check at a generic kinematic point for four gluons and four gravitons would test the recursive rule's handling of exchanged branches, the least explicit part of the proof."],"forward_implications":["The formula gives an explicit collinear-pair representation for MHV EYM amplitudes with an arbitrary number of gravitons, not just one or two.","An $N$-gluon, $M$-graviton MHV amplitude is expressed as a sum of pure-gluon MHV amplitudes with $N+2M$ external legs, each graviton pair sharing a momentum and helicity.","Applying the stated replacements extends the formula to the $(h^-,g^-)$ MHV configuration, with the positive-helicity graviton set and an overall minus sign.","The recursive insertion rule (4.2) gives a concrete algorithm: for each spanning forest, read the tree structure to build the permutation and multiply the Parke-Taylor factor by the product of $s_{ab}$ over tree edges."],"supporting_citations":[{"why":"Supplies the SBDW formula (2.7)–(2.8) that expresses the single-trace MHV amplitude through a generating function, the starting point of the paper's derivation.","marker":"[1–3]"},{"why":"Provides the spanning-forest expansion (2.10) of $S(H;G)$ that the paper converts into collinear-gluon pairs.","marker":"[10]"},{"why":"Establishes the one- and two-graviton collinear-gluon formulas that the general formula (4.1) extends and reduces to.","marker":"[18]"},{"why":"Gives the Parke-Taylor formula for tree-level MHV gluon amplitudes, the pure-gluon factors into which the graviton pairs are inserted.","marker":"[23]"},{"why":"Motivates the graviton-as-collinear-gluon-pair viewpoint that the paper makes explicit and systematic.","marker":"[17–21]"}],"fun_headline_variants":["Gravitons split into gluon pairs in MHV EYM amplitudes","MHV EYM: a forest formula with graviton-to-gluon splitting","Single-trace MHV amplitudes: gravitons become gluon pairs","Extending the graviton-splitting formula to general MHV EYM","All MHV EYM amplitudes from spanning forests of gluons"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the recursive insertion rule (4.2) converting the spanning-forest sum (2.10) into the collinear-gluon sum (4.1) with a one-to-one correspondence, so that no forest is double-counted when branches are exchanged or roots coincide; this is shown explicitly only for three gravitons and sketched for the general case.","fun_headline_variants_meta":{"raw":{"variants":["Gravitons split into gluon pairs in MHV EYM amplitudes","MHV EYM: a forest formula with graviton-to-gluon splitting","Single-trace MHV amplitudes: gravitons become gluon pairs","Extending the graviton-splitting formula to general MHV EYM","All MHV EYM amplitudes from spanning forests of gluons"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001156,"raw_usage":{"total_tokens":4803,"prompt_tokens":971,"completion_tokens":3832,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":587,"completion_tokens_details":{"reasoning_tokens":3734}},"tokens_in":587,"tokens_out":3832,"duration_ms":27407,"temperature":1.0,"reasoning_tokens":3734,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T22:16:17.904542+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the right-hand side of Eq. (4.1) for a four-graviton, four-gluon MHV amplitude at a generic kinematic point and compare it with the directly evaluated left-hand side obtained from the spanning-forest expansion (2.10) or from the CHY formula; any mismatch, in particular from forests whose exchanged branches produce the same permutation, would falsify the claimed identity.","supporting_citations":[{"cited_title":"Direct Evaluation of $n$-point single-trace MHV amplitudes in 4d Einstein-Yang-Mills theory using the CHY Formalism","cited_arxiv_id":"1608.00883","evidence_quote":"Provides the spanning-forest expansion (2.10) of $S(H;G)$ that the paper converts into collinear-gluon pairs."},{"cited_title":"Disk relations for tree amplitudes in minimal coupling theory of gauge field and gravity","cited_arxiv_id":"1001.0060","evidence_quote":"Establishes the one- and two-graviton collinear-gluon formulas that the general formula (4.1) extends and reduces to."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the Parke-Taylor formula for tree-level MHV gluon amplitudes, the pure-gluon factors into which the graviton pairs are inserted."}],"review_version":1}