{"id":"45a12390-ee2f-4b68-b556-363ff51bc8df","arxiv_id":"2411.15071","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A new Hopf algebra of formal multiple polylogarithms is defined for every field, with Hodge and motivic realizations, proposed as the conjectural Hopf algebra of framed mixed Tate motives.","lead":"This paper builds a Hopf algebra, an algebraic structure that captures the identities of multiple polylogarithm functions over any field, and equips it with maps to both Hodge-theoretic and motivic objects. If its conjectural identification with the motivic Hopf algebra is correct, it would give an elementary model for mixed Tate motives over arbitrary fields.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Hodge realization depends on specialization formula (40) for colliding arguments; without it, Proposition 48 fails and the motivic realization cannot be shown well-defined.","rationale":"The reader's conditional verdict is appropriate. The construction of Lf(F) as A(F)/R(F) is largely self-contained: Lemma 7 and Corollary 8 give the Lie coideal property, and the coJacobi identity, though delegated, follows by a standard argument. The point that actually carries the weight of the central claim is the existence of the Hodge realization in Proposition 48; without it, Proposition 50 has no leverage because it uses the Hodge realization to kill R0−R1. Both realizations reduce to a single external identity, (40), whose hypotheses are not checked in the collapsed-point case. My proposed computation directly probes exactly that case: if (40) is valid as stated for arbitrary rational functions, the concern is resolved and the paper's conditional claim stands; if not, the realizations are not known to be well-defined, and the paper would need to add a limiting argument or restrict the definition of the relation spaces Rn(F). The reader flagged this same dependency, so my read does not change the conditional verdict.","tokens_in":31847,"tokens_out":25444,"duration_ms":257158,"concrete_test":"Take n=2, f0(t)=0, f1(t)=t, f2(t)=t^2, and t0=0. This is the simplest case where the tuple collapses (all three points coincide at t=0), so it tests the non-general-position regime. Compute both sides of (40) directly from the definition of Hodge correlators on P^1 minus {0,t,t^2}. The right-hand side is CorQ-Hod(Sp_{t→0}(0,t,t^2)) = CorQ-Hod(0,0,0) = 0 by (A5). The left-hand side is the specialization of a framed variation of mixed Hodge–Tate structures; verify it is the zero element of LQ-Hod_2. If the only available proof of (40) assumes the f_i(t0) remain distinct, this computation will expose the missing limiting argument.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is Proposition 48. Its induction proves that Hodge correlators kill R_n(C) by comparing Spt→0 and Spt→1 of a δ-closed element R via identity (40): Sp_{t→t0} CorQ-Hod(f0(t),...,fn(t)) = CorQ-Hod(Sp_{t→t0}((f0,...,fn))). This identity is quoted from [Mal20, Thm. 28], and the paper's specialization map on A_n is designed specifically to work without general-position assumptions, i.e., when several f_i(t0) coincide. But the paper does not verify that Malkin's theorem covers exactly this collapsed case, nor how the limiting mixed Hodge–Tate structure on the left is framed. If (40) only holds when the map C∖S → Conf_m(C) is étale at t0 (all f_i(t0) distinct), or if a hidden framing is needed for the specialization, the induction in Proposition 48 collapses: statement (ii) for R_n(C) cannot be concluded, and Proposition 50 inherits the failure because it uses rσ(R0−R1)=0 via Proposition 48. All downstream claims that identities in Hf(F) imply identities among framed mixed Hodge–Tate structures depend on this. This is not an objection to using deep theorems; it is a request to spell out the exact hypothesis of [Mal20, Thm. 28] and confirm the non-étale case.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper defines, for every infinite field F, a graded connected commutative Hopf algebra Hf(F) of formal multiple polylogarithms, obtained as the universal coenveloping coalgebra of a Lie coalgebra Lf(F). The Lie coalgebra is generated by correlator symbols modulo an inductively defined space of relations Rn(F) that is designed to enforce homotopy invariance under pure transcendental extensions. The paper proves basic structural properties of correlators and iterated integrals, establishes shuffle, reversal, distribution, and quasi-shuffle relations, identifies the weight-two piece with the rational Bloch group, and then constructs a Hodge realization for F ⊂ C and a motivic realization for number fields. It closes with Goncharov-type conjectures relating Hf(F) to algebraic K-theory and to the conjectural Hopf algebra of framed mixed Tate motives.","tokens_in":32195,"tokens_out":6455,"duration_ms":69585,"significance":"If the central construction is sound, this is a valuable explicit candidate for Goncharov's motivic Hopf algebra, and the realization theorems give a concrete mechanism by which functional equations proved in Hf(F) become identities among framed mixed Hodge–Tate structures and, for number fields, among mixed Tate motives. The paper is commendably explicit about its conjectural status and about which properties are built into the definition rather than derived; the weight-two identification with the Bloch group is a concrete nontrivial result, and the treatment of specializations without general-position assumptions is an improvement over earlier constructions. However, the overall claim rests on two kinds of load-bearing support that are not fully supplied: the exact hypotheses of the external specialization theorem used for the Hodge realization, and the proofs of two structural identities that make Lf(F) a Lie coalgebra and make the shuffle product well-defined.","major_comments":[{"comment":"The Hodge realization depends on the identity Sp_{t→t0}(Cor^{Q-Hod}(f0(t),...,fn(t))) = Cor^{Q-Hod}(Sp_{t→t0}((f0,...,fn))), quoted from [Mal20, Theorem 28]. The paper's specialization map Sp_{t→t0} is deliberately defined without general-position assumptions, and the inductive proof of Proposition 48 needs this identity precisely in the case where several of the f_i(t0) coincide. The manuscript does not state the hypotheses of [Mal20, Theorem 28] and does not verify that they cover this collapsed, potentially non-étale case, nor does it explain how the limiting framed mixed Hodge–Tate structure on the left-hand side is defined. If Malkin's theorem requires the map to the configuration space to be étale at t0, or requires an auxiliary framing that is not tracked here, then statement (ii) of Proposition 48 does not follow, and Proposition 50 inherits the failure through its use of Proposition 48. This is a request for a precise quotation of the theorem and a verification of the non-generic case, not an objection to the use of deep external results.","section":"§5.1, Eq. (40), Proposition 48"},{"comment":"Two structural identities on which the definition of Lf(F) rests are not proved in the manuscript. Lemma 2 leaves the coJacobi identity as an exercise, and Proposition 15 refers to [Gon01a, pp. 437–438] for the combinatorial part; moreover the base case in Proposition 15 currently reads 'The base of induction n = 2 follows from .' with the reference missing. Since Hf(F) is defined as the universal coenveloping coalgebra of Lf(F), the coJacobi identity is part of the definition of the object being studied, and the shuffle relation is used in Lemma 24 to construct the Hopf algebra. These checks should either be carried out in the paper or replaced by a precise statement of the cited result together with a verification that its hypotheses apply to the present relations (A1)–(A6).","section":"§2.1, Lemma 2, and §2.4, Proposition 15"},{"comment":"The homotopy invariance statement Corollary 11 is true by construction, because Rn(F) is defined as the span of Spt→0(R) − Spt→1(R) for δ-closed R. This is a feature of the axiomatic construction, but it should be labelled more explicitly as a built-in property rather than as a derived theorem, since it is the main input that makes the relation spaces mirror the conjectural K-theoretic identification (3). The surrounding discussion does say the definition is guided by Beilinson–Soulé vanishing, but a reader could mistake Corollary 11 for independent evidence for Conjecture 39. Clarifying this distinction would strengthen the paper's conceptual honesty without changing the mathematics.","section":"§2.3 and Corollary 11"}],"minor_comments":[{"comment":"In the displayed computation for the reversal symmetry, the text says 'in the second equality we used the induction assumption, in the second we changed summation indices'; the second occurrence should presumably be 'in the third' or 'in the fourth', depending on the intended numbering of the equalities.","section":"§2.4, Proposition 14 proof"},{"comment":"The notation log^H is used in the identity I(0;0,...,0;x_{n+1}) = (log^H(x_{n+1}))^n/n! but it is not defined before this point; the authors should either define it or replace it with an explicit description in terms of Cor(x0,x1).","section":"§3.2, Proposition 27 proof"},{"comment":"The sentence 'It follows that δ(M2(R)) = 0 and so δ(M2(R)) lies in the kernel ...' should read 'M2(R) lies in the kernel ...', since the condition δ(M2(R)) = 0 is exactly what places M2(R) in the kernel.","section":"§4.4, Proposition 45 proof"},{"comment":"The notation 'C = P1 \\ {∞}' is slightly confusing because C is then used both as the complex plane and as the punctured projective line; the intended meaning is clear, but a brief gloss would help.","section":"§5.1, paragraph after Eq. (38)"},{"comment":"The proof of injectivity of the regulator map (42) is stated as a consequence of Borel's theorem; it would be helpful to indicate explicitly that the comparison of Beilinson and Borel regulators is being used to identify the real regulator with the Borel regulator, rather than merely asserting injectivity of the sum of Hodge realizations.","section":"§5.2, paragraph before Eq. (42)"}],"recommendation":"major_revision","confidential_remarks":"The paper is substantial and likely publishable after the identified points are addressed. The main risk is not the depth of the external results but whether their hypotheses match the non-generic specializations used in Proposition 48; this should be resolved before acceptance. The omitted proofs of Lemma 2 and Proposition 15 are less risky but should be completed or precisely referenced, especially since the current text contains an incomplete sentence in Proposition 15. The axiomatic status of Corollary 11 is acceptable for a conjectural candidate, but the paper should make that status unmistakable."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThe short version: this paper is a serious, carefully built candidate for an elementary model of Goncharov's conjectural Hopf algebra of framed mixed Tate motives, and it deserves refereeing. What is genuinely new is not the broad idea—Goncharov and GKLZ already had similar Lie coalgebras—but the specialization maps that work without general-position assumptions, which is what lets the authors construct Hodge and motivic realization maps. The weight-two proof recovering the Bloch group is clean, and the paper is honest about what is theorem and what is conjecture.\n\nThe main soft spot is the one flagged by the stress test. Proposition 48, and therefore Proposition 50, rests on identity (40) quoted from Malkin's Theorem 28, which lets specialization pass through the Hodge correlator. The paper does not check that Malkin's theorem covers the collapsed, non-étale case where several f_i(t0) coincide. Since the whole point of the algebraic specialization map is to handle exactly that case, this is load-bearing. I do not think it is fatal—Malkin's work is aimed at such specializations—but the authors should spell out the hypothesis and either prove (40) or give a precise reference covering the non-étale case. Without that, the induction in Proposition 48 does not close.\n\nThe other soft spots are minor. Lemma 2 leaves the coJacobi identity as an exercise, and Proposition 15 refers shuffle relations to Goncharov. Both are probably true, but they are part of the foundation and should be supplied. I also want to flag the circularity noted in the reader's report: R_n(F) is defined so that homotopy invariance under pure transcendental extensions holds by construction, guided by the conjectured K-theory isomorphism. That is an axiomatic choice, not a derivation. The paper states this clearly, so it is a limitation rather than a hidden flaw, but it does mean part of the desired structure is built into the definition.\n\nBottom line: this is a well-built, serious contribution for people working on polylogarithms and mixed Tate motives. Send it to a knowledgeable referee, and ask them to check the Malkin input and the delegated identities. I would want to see those gaps closed before I would bet on the realization theorems, but the construction itself is worth engaging with.","headline":"A serious candidate for Goncharov's Hopf algebra with genuinely new specialization maps; the realizations hinge on whether Malkin's theorem covers the collapsed case.","tokens_in":32688,"tokens_out":2461,"would_cite":true,"duration_ms":25514,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11G55","11M32","11R70","19F27","14F42","16T05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper builds an explicit Hopf algebra of formal multiple polylogarithms over any field and maps it to mixed Tate motives and Hodge structures.","keywords":["multiple polylogarithms","Hopf algebra","mixed Tate motives","Bloch group","correlators","iterated integrals","motivic cohomology","functional equations"],"falsifier":"Look for an explicit $R\\in\\mathrm{A}_n(\\mathbb{C}(t))$ with $\\delta(R)=0$ in the relation quotient for which the Hodge correlator of $\\mathrm{Sp}_{t\\to 0}R-\\mathrm{Sp}_{t\\to 1}R$ is nonzero; Proposition 48 predicts this difference always vanishes, so any such element would break the Hodge realization and with it the central claim.","tokens_in":31683,"feed_emoji":"🧮","tokens_out":12390,"duration_ms":112053,"temperature":0.7,"pith_summary":"Multiple polylogarithm functions satisfy many functional equations, but proving these identities and lifting them to motivic objects is traditionally laborious. This paper gives every infinite field $F$ an elementary Hopf algebra $\\mathrm{H}^f(F)$—a graded algebra whose coproduct formalizes how a polylogarithm decomposes into simpler pieces—built from formal iterated integrals, with relations imposed by comparing two limiting values of certain elements over $F(t)$. Inside this algebra the classical five-term, shuffle, quasi-shuffle, distribution, and inversion identities become theorems. The paper then constructs a Hodge realization into framed mixed Hodge–Tate structures when $F\\subset\\mathbb{C}$ and a motivic realization into mixed Tate motives when $F$ is a number field, and conjectures that the motivic realization is an isomorphism. The central claim, read in good faith, is that $\\mathrm{H}^f(F)$ is the conjectural Hopf algebra of framed mixed Tate motives.","feed_headline":"A Hopf algebra captures every polylogarithm identity","feed_subtitle":"Formal multiple polylogarithms map to Hodge structures and mixed Tate motives, so identities become motive relations.","key_machinery":"The machine is the pair $(\\mathrm{A}(F),\\delta)$ of formal correlators with the cut cobracket: an element $((x_0,\\dots,x_n))$ records $n+1$ points on a circle, and $\\delta$ sums over all ways to cut the circle at a point and split the remaining arc into two smaller circular configurations, exactly the combinatorics of motivic correlators. Relation spaces are then defined by a specialization trick: after passing to $F(t)$, any element whose cobracket vanishes modulo previously defined lower-weight relations is required to have equal specializations at $t=0$ and $t=1$, which encodes the idea that the relation is constant without any general-position assumptions. Finally, the universal coenveloping coalgebra construction turns the Lie coalgebra $\\mathrm{L}^f(F)$ into the commutative Hopf algebra $\\mathrm{H}^f(F)$, and the classical iterated integrals are lifted uniquely into it on the basis of the correlator cobracket.","core_discovery":"The central object is the graded Lie coalgebra $\\mathrm{L}^f(F)=\\mathrm{A}(F)/\\mathrm{R}(F)$: $\\mathrm{A}(F)$ is the rational vector space spanned by formal correlators $((x_0,\\dots,x_n))$ modulo cyclic, translation, homothety, and logarithm relations, with a cobracket that cuts a cyclic configuration into two smaller ones. The relations $\\mathrm{R}_n(F)$ are generated inductively by taking $R\\in\\mathrm{A}_n(F(t))$ whose cobracket lands in $\\bigoplus_{k=1}^{n-1}\\mathrm{A}_k\\wedge\\mathrm{R}_{n-k}$ and declaring the specializations at $t=0$ and $t=1$ equal. The Hopf algebra $\\mathrm{H}^f(F)$ is the universal coenveloping coalgebra of this Lie coalgebra—the free commutative Hopf algebra with a given Lie coalgebra of indecomposables—so its indecomposables are exactly $\\mathrm{L}^f(F)$; iterated integrals and multiple polylogarithms are defined in it and satisfy the expected coproduct, shuffle, and composition laws. The paper proves that the weight-one part is $F^\\times_\\mathbb{Q}$, the weight-two part is the rationalized Bloch group, and that the Hodge realization $\\mathrm{r}^{\\mathrm{Q}\\text{-}\\mathrm{Hod}}$ and the motivic realization $\\mathrm{r}^{\\mathrm{M}}$ are well-defined Hopf algebra morphisms. The announced conclusion is that every functional equation proven inside $\\mathrm{H}^f(F)$ is a genuine relation among framed mixed Hodge–Tate structures and, for number fields, among framed mixed Tate motives.","pith_inferences":["A natural next computation would be to implement the relation-space recursion for $F=\\mathbb{Q}$ in weights up to five and compare the resulting depth filtration with the known multiple-zeta-value table; the paper's explicit generators make this a finite linear-algebra problem.","The same recipe of taking a vanishing-cobracket element over $F(t)$ and identifying its two specializations could plausibly construct Hopf algebras for other families of configuration-space integrals, such as Grassmannian polylogarithms.","If the isomorphism conjecture is true, the canonical real period map on the Hodge realization would attach to each formal identity a real number, giving a direct path from algebraic relations in $\\mathrm{H}^f(\\mathbb{C})$ to Borel-regulator values and special values of $L$-functions."],"forward_implications":["Every functional equation proven inside $\\mathrm{H}^f(F)$ (five-term relation, shuffle, quasi-shuffle, distribution, inversion) becomes a theorem about framed mixed Hodge–Tate structures, and about framed mixed Tate motives when $F$ is a number field.","Since $\\mathrm{H}^f_2(F)$ is the rationalized Bloch group, the algebra gives a uniform home for weight-two regulator and K-theory computations and a model for higher weights.","If the conjectured Chern-class isomorphisms hold, the higher cohomology of $\\mathrm{H}^f(F)$ computes the $\\gamma$-graded algebraic K-theory of $F$, turning the Hopf algebra into a concrete computational target for $K_{2n-i}(F)_\\mathbb{Q}$.","If the motivic realization is an isomorphism, $\\mathrm{H}^f(F)$ supplies an elementary presentation of the Hopf algebra of framed mixed Tate motives for number fields, independent of the full Tannakian category."],"supporting_citations":[{"why":"Supplies the specialization theorem for Hodge correlators, equation (40), used to prove the Hodge realization descends to the relation space.","marker":"[Mal20]"},{"why":"Provides the rigidity lemma and the cobracket formula for motivic correlators on which the relation construction and its Hodge proof rest.","marker":"[GR18]"},{"why":"States the conjectural framework of motivic multiple polylogarithms and mixed Tate motives that the paper formalizes.","marker":"[Gon01b]"},{"why":"Constructs mixed Tate motives over number fields, the target category of the motivic realization.","marker":"[DG05]"},{"why":"Gives existence of Tate motives over number fields, the background that makes the motivic target available.","marker":"[Lev93]"},{"why":"Proves the injectivity of the Borel regulator used to establish that the motivic realization is well-defined.","marker":"[Bor77]"},{"why":"Defines the Bloch group and its differential and specialization maps, used to prove $\\mathrm{L}^f_2(F)$ is the rationalized Bloch group.","marker":"[Sus90]"},{"why":"Introduces Hodge correlators and the canonical real period map, which the Hodge realization evaluates.","marker":"[Gon19a]"},{"why":"Supplies the correspondence between connected graded Hopf algebras and Lie coalgebras used to define $\\mathrm{H}^f(F)$ as universal coenveloping coalgebra.","marker":"[MM65]"}],"fun_headline_variants":["One Hopf algebra rules all polylog identities","Polylogarithm identities become motive relations","Formal polylogs: a Hopf algebra for mixed Tate motives","Every polylog identity is a motive relation","Hopf algebra unifies polylogarithm identities"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The construction depends on the inductively defined relation spaces forming a Lie coideal, and the realization theorems depend on two external inputs—the rigidity of variations of mixed Hodge–Tate structures and the injectivity of the Borel regulator—whose applicability is imported rather than proved here.","fun_headline_variants_meta":{"raw":{"variants":["One Hopf algebra rules all polylog identities","Polylogarithm identities become motive relations","Formal polylogs: a Hopf algebra for mixed Tate motives","Every polylog identity is a motive relation","Hopf algebra unifies polylogarithm identities"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000493,"raw_usage":{"total_tokens":2415,"prompt_tokens":934,"completion_tokens":1481,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":550,"completion_tokens_details":{"reasoning_tokens":1418}},"tokens_in":550,"tokens_out":1481,"duration_ms":9981,"temperature":1.0,"reasoning_tokens":1418,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:32:42.870546+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Look for an explicit $R\\in\\mathrm{A}_n(\\mathbb{C}(t))$ with $\\delta(R)=0$ in the relation quotient for which the Hodge correlator of $\\mathrm{Sp}_{t\\to 0}R-\\mathrm{Sp}_{t\\to 1}R$ is nonzero; Proposition 48 predicts this difference always vanishes, so any such element would break the Hodge realization and with it the central claim.","supporting_citations":[],"review_version":1}