{"id":"ff082e08-4f1a-43b7-99be-cd6e31317ded","arxiv_id":"2506.09493","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"A self-described non-research thesis summarizing nine papers, with new conjectures on TRAP-based Feynman rules and accelero-summation of asymptotically free theories.","lead":"This habilitation thesis compiles the author's mathematical work in quantum field theory, covering TRAPs, generalized zeta values, locality structures, and resurgence. It is a synthesis of nine prior papers, adding two new conjectures toward rigorous Feynman rules and accelero-summation.","discovery_kind":"review","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The Feynman-rule program hinges on Conjecture 1.8.11; the TRAP operations on distribution-valued meromorphic germs are not defined and stability under partial traces is unproved, so the construction remains conditional.","rationale":"I read the central claim as conditional: the freeness theorem for solar corolla oriented graphs is proved in detail and appears internally coherent, but the advertised application to Feynman rules requires a target TRAP of distribution-valued meromorphic germs. The reader's weakest-assumption analysis identifies exactly the right risk: Conjecture 1.8.11 is unproved, and Proposition 1.8.14 depends on it. My stress-test adds that the conjecture is not even fully specified: the partial trace maps on the inductive limit are not defined in the text, so a proof of stability is needed before the conjecture can be tested. This does not change the verdict: the thesis is a transparent habilitation synthesis that labels its conjectures as conjectures, and the provided proofs where they exist appear sound. The appropriate verdict remains CONDITIONAL, and no adjustment to the reader's verdict is required.","tokens_in":67697,"tokens_out":5757,"duration_ms":70663,"concrete_test":"Work in the simplest nontrivial case of Section 1.8. Let p and q be two elements of P(1,1), i.e. distribution-valued meromorphic germs with one input and one output variable, each having only linear poles in the regularisation parameters. Form p\\u2297q in P(2,2) by horizontal concatenation and compute the candidate partial trace t_{1,1}(p\\u2297q), defined by contracting the two middle variables through the distributional pairing, and then test: (1) is the result again a distribution-valued meromorphic germ with only linear poles; (2) does the result lie in the declared inductive limit; (3) does the commutativity axiom t_{1,1}\\u2218t_{2,2} = t_{1,1}\\u2218t_{1,1} of Definition 1.4.1 hold on a two-loop example? If any of these fails, Conjecture 1.8.11 is false and the Feynman-rule construction collapses.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Section 1.8 is sound as a conditional programme only if Conjecture 1.8.11 holds: the inductive limit of spaces of distribution-valued meromorphic germs with linear poles must carry a TRAP structure. The manuscript itself states that the horizontal concatenation and partial traces can only be 'conjecturally defined' on this limit. This is not a merely technical gap: Proposition 1.8.14, which gives the expected analytic and algebraic properties of the canonical Feynman rules, is derived from the conjecture, and Definition 1.8.12 uses the TRAP morphism from Theorem 1.6.8. A partial trace maps a germ depending on k+l variables to one depending on k+l\\u22122 variables by identifying an input and an output variable; for distribution-valued germs this requires a pullback under a diagonal map that need not be well-defined on arbitrary distributions, and preservation of the 'linear poles' condition is not automatic. If this stability fails, the universal property of Theorem 1.6.8 produces a map into a space that is not a TRAP, so the proposed rigorous Feynman rules have no target structure. The concern is load-bearing but it is frankly declared by the author; the thesis does not claim a proof.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This habilitation thesis synthesises results from nine papers into four largely independent chapters. Chapter 1 introduces PROPs and TRAPs, proves that the PROP of generalised graphs is free over indecomposable graphs (Theorem 1.3.10), and proves that solar corolla ordered decorated graphs form the free TRAP generated by a family of sets (Theorem 1.6.8). It then proposes a program to construct Feynman rules canonically from the TRAP universal property, conditional on Conjecture 1.8.11. Chapter 2 develops arborified zeta values, shuffle and stuffle AZVs, and tree zeta values, proving in particular that AZVs are finite rational linear combinations of MZVs and that shuffle AZVs admit a multiple-series representation; applications to Mordell-Tornheim and conical zeta values are given. Chapter 3 develops locality structures, proves a locality Birkhoff-Hopf factorisation (Theorem 3.4.12), and applies it to the multivariate renormalisation of Kreimer's toy model (Theorem 3.5.21). Chapter 4 performs a resurgent analysis of the Wess-Zumino model, proving 1-Gevrey and resurgence properties of the solution to the truncated RGE and deriving an asymptotic bound; its headline result, Corollary 4.4.11, is however conditional on the unproven Claim 4.2.2, as the text itself acknowledges.","tokens_in":1682,"tokens_out":1799,"duration_ms":86908,"significance":"The free TRAP theorem is a useful structural result: it provides a canonical TRAP morphism from decorated graphs into any TRAP, so that the construction of Feynman-like amplitudes reduces to the choice of a TRAP-valued decoration. The arborified-zeta-value results are concrete and testable, with explicit algebraic identities, convergence statements, and series representations. The locality Hopf factorisation, including the observation that the renormalised value reduces to minimal subtraction under the locality assumptions, is a genuine contribution. The resurgence chapter is valuable as a detailed conditional program, but its full value depends on an external physics input. The algebraic proofs are written out in detail, and the conditional nature of the analytic applications is mostly acknowledged in the prose; the main work for publication is to make that conditional status visible in the theorem statements and abstract.","major_comments":[{"comment":"The proof of Corollary 4.4.11, the main resummability statement of Chapter 4, rests on Claim 4.2.2, which the manuscript explicitly says is 'taken for granted' (General Introduction, p. 13, and Section 4.2). The claim is used to establish that the Borel transform of the two-point function is resurgent; without it, Theorem 4.3.12 gives resurgence only under an external physics input. I recommend stating Corollary 4.4.11 as a conditional theorem, for instance as 'Assuming Claim 4.2.2, the solution... is Borel-Ecalle resummable', and moving this caveat from the introduction into the theorem environment. If the claim is regarded as an imported result from the physics literature, a precise citation and an explicit statement of its status as an unproved input would remove the current ambiguity about what has been proved in the thesis.","section":"Claim 4.2.2 / Corollary 4.4.11"},{"comment":"The proposed rigorous construction of Feynman rules is conditional on Conjecture 1.8.11, and the manuscript is candid about this. However, the abstract and the chapter introduction say that TRAPs 'could be used' to define Feynman rules, which can be read as a stronger claim. Proposition 1.8.14 is not a theorem in the current state of knowledge; it is a consequence of Conjecture 1.8.11. I recommend that the opening of Section 1.8 and the abstract state explicitly that Definition 1.8.12 and Proposition 1.8.14 are conditional on the conjecture, and that the main open difficulty of the conjecture, namely stability of the inductive limit of distribution-valued meromorphic germs under partial trace maps, be restated as a required step before the universal property of Theorem 1.6.8 can be applied to Feynman rules.","section":"Section 1.8, Conjecture 1.8.11, Definition 1.8.12, Proposition 1.8.14"}],"minor_comments":[{"comment":"The index ranges in Axiom 3(c) appear to be misprinted: for p in P(k,l) and p' in P(k1,l1), the partial trace t_{i,j} should be considered for i in [k+k1] and j in [l+l1], not for i in [k+l] and j in [k1+l1].","section":"Definition 1.4.1, item 3(c)"},{"comment":"The cross-references to Theorem 1.5.7 are inconsistent: the General Introduction describes it first as a PROP statement and later as a TRAP statement. The PROP structure on continuous morphisms is Theorem 1.2.15, while the TRAP structure is Theorem 1.5.7; the text should be harmonized.","section":"General Introduction and Chapter 1 introduction"},{"comment":"The compressed 'resp.' formulation ('stuffle (resp. starred stuffle, shuffle) AZVs') is hard to parse. I suggest splitting the theorem into separate statements for the stuffle, starred stuffle, and shuffle cases, each with its own product and algebra-morphism assertion.","section":"General Introduction, statement of Theorems 2.3.24 and 2.4.15"},{"comment":"In the proof of compatibility of Phi with the partial trace, the notation e = {e1, f1} is used for the newly created internal edge while G^1_e refers to the cut graph; a different symbol for the edge would avoid confusion between the set of two glued edges and the resulting single edge.","section":"Section 1.6.4, proof of Lemma 1.6.17"}],"recommendation":"major_revision","confidential_remarks":"To the editor: this is an habilitation thesis, and the standards for such documents differ from those for a research article. The main issue is that the Chapter 4 headline result is conditional on Claim 4.2.2, and the Section 1.8 Feynman-rule construction is conditional on Conjecture 1.8.11. Neither is hidden, but the framing in the abstract and in the theorem statements should be sharpened so that conditional results are labelled as conditional. I do not see evidence of circularity or of deliberate overclaiming; the author is explicit about what is conjecture. If the venue accepts conditional theorems and clearly marked programmatic sections, the manuscript is close to acceptable after the requested revisions."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nIf you're expecting a new research result, this won't be it. The thesis says up front that it is 'essentially not a research text'—it's a synthesis of nine published papers. The only new content is two conjectures (1.8.11 and 4.5.2) and a conditional proposition (1.8.14). That is consistent with it being an HDR thesis, but it changes what you should ask of it.\n\nWhat the thesis does well is give a complete, readable tour of the author's program. The freeness of the PROP of graphs and the free TRAP theorem (1.6.8) are proved in full detail; I don't see obvious gaps. The examples, especially the TRAP of smoothing kernels on a closed manifold, are instructive. In Chapter 3, the main locality result—that the Birkhoff-Hopf factorisation collapses to minimal subtraction under the locality assumptions—is a genuinely clean theorem. Chapter 2 on arborified zeta values is substantial, even if its results are already in the literature.\n\nThe soft spots are real, but they're openly declared. Conjecture 1.8.11 is load-bearing: the inductive limit of distribution-valued meromorphic germs with linear poles is conjectured to carry a TRAP structure, and the author says the partial traces and horizontal concatenation can only be 'conjecturally defined' on it. That is not a minor technicality—stability under partial traces is exactly the hard analytic step. If the conjecture fails, the Feynman-rule map in Definition 1.8.12 has no target structure. Proposition 1.8.14 is explicitly conditional on it. Likewise, the Chapter 4 resummation theorem (Corollary 4.4.11) depends on Claim 4.2.2, that the Borel-transformed anomalous dimension of the Wess-Zumino model is resurgent. The author says this is 'taken for granted,' not proved. So the two headline applications are conditional, not established.\n\nWho is this for? Someone who wants a single reference for the TRAP framework, the locality construction, and the resurgence work, and who can handle conjectures being labelled as such. The conjectures are concrete enough to work on; a referee could usefully push on whether the partial trace stability in Conjecture 1.8.11 is plausible and whether the WZ resurgence claim has independent support. I would not publish this as an original research article—the theorems it proves are already in the primary literature. But I would send it to a referee if the venue accepts review/synthesis pieces. It deserves serious engagement rather than a desk rejection based on novelty alone. The honest framing is a plus, not a minus.","headline":"An honest HDR synthesis whose only new content is two conjectures; the Feynman-rule program is conditional on a TRAP structure that remains unproven.","tokens_in":68484,"tokens_out":4115,"would_cite":true,"duration_ms":46959,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81Q30","81T18","81T15","11M32"],"pacs":[],"model":"deepseek-v4-flash","headline":"This thesis proves that solar corolla-ordered graphs decorated by X are the free TRAP generated by X, and uses that universal property to propose a rigorous route to Feynman rules, conditional on a stated conjecture.","keywords":["TRAP","wheeled PROPs","Feynman rules","multiple zeta values","rooted forests","locality structures","renormalisation","resurgence"],"falsifier":"Take a Feynman integrand of a scalar quantum field theory on $\\mathbb{R}^d$, view it as a distribution-valued meromorphic germ in the regularisation parameters $z_1,\\dots,z_E$, and apply one partial trace $t_{i,j}$ that identifies an input and an output. If for some graph the result acquires a pole that is not linear, or leaves the space of linear-pole germs, then Conjecture 1.8.11 is false and the proposed rigorous definition of Feynman rules is invalid.","tokens_in":1891,"feed_emoji":"📐","tokens_out":2384,"duration_ms":83981,"temperature":0.7,"pith_summary":"This habilitation thesis argues that a universal algebraic structure called a TRAP — a family of vector spaces with horizontal concatenation and partial trace maps — is the right setting to make Feynman rules rigorous. Its central result is that solar, corolla-ordered, decorated generalised graphs form the free TRAP generated by their decorations: any assignment of decorations to a target TRAP extends uniquely to a TRAP morphism on all such graphs. That universal property gives a canonical route from Feynman graphs to analytic spaces, provided the target space carries a TRAP structure. The thesis makes this concrete by stating Conjecture 1.8.11, that the inductive limit of spaces of distribution-valued meromorphic germs with linear poles carries a TRAP structure, which would turn Feynman rules into a well-defined map. The same algebraic philosophy also organises the other chapters: rooted forests generalise multiple zeta values, locality structures encode multivariate renormalisation, and resurgence theory builds analytic two-point functions for the Wess-Zumino model.","feed_headline":"Free 'TRAP' algebra could make Feynman rules rigorous","feed_subtitle":"If a conjectured trace structure holds on meromorphic germs, Feynman amplitudes become canonical maps instead of recipes.","key_machinery":"The central object is the TRAP, a family of vector spaces $P(k,l)$ carrying a symmetric-group action, an associative and commutative horizontal concatenation, and partial trace maps $t_{i,j}: P(k,l) \\to P(k-1,l-1)$ that close an input to an output. The load-bearing identity is the freeness theorem: solar corolla-ordered graphs — graphs with no through-edges and with totally ordered half-edges at each vertex — decorated by $X$ are the free TRAP on $X$, so they admit a unique morphism to any TRAP. This is what lets graph amplitudes be defined canonically once vertex decorations are chosen, and it is also the mechanism behind the generalised trace and the amplitude map from decorated graphs to a TRAP.","core_discovery":"On its own terms, the paper establishes Theorem 1.6.8: if $X=(X(k,l))$ is any family of sets, the TRAP \\mathsf{solCGr}(X) of solar corolla-ordered generalised graphs decorated by $X$ is the free TRAP generated by $X$. In plain terms, every coherent way of assigning vertex decorations into a TRAP extends uniquely to an entire graph amplitude that respects horizontal concatenation, permutations, and partial traces. This converts the problem of defining Feynman rules into the problem of equipping the target analytic space with a TRAP structure and specifying the vertex data. The thesis also proves the folklore statement that generalised graphs form the free PROP over indecomposable graphs, and explains why that PROP-level result is insufficient for QFT: closed loops require trace-like operations, which PROPs do not provide and TRAPs do.","pith_inferences":["Editorial inference: the free-TRAP theorem is a template that could be applied beyond QFT; any setting where a combinatorial class of wired graphs is universal would automatically produce canonical trace-compatible evaluations once a target TRAP is specified.","Editorial inference: if Conjecture 1.8.11 fails only because the spaces are not stable under partial traces, one could enlarge the space or define partial traces on a completion, preserving the universal graph-level arguments while modifying the analytic target.","Editorial inference: since unital TRAPs are wheeled PROPs, the freeness result should transfer to wheeled PROPs, potentially giving a universal construction of traces in invariant theory and other fields where wheeled PROPs are already used.","Editorial inference: the accelero-summation conjecture for asymptotically free theories could be tested in simpler toy models by computing the Borel transform's singularity structure and checking whether the predicted logarithmic acceleratrix form appears."],"forward_implications":["If the conjectured TRAP structure on distribution-valued meromorphic germs with linear poles exists, Feynman rules become a genuine map from Feynman graphs to that analytic space, with amplitudes automatically compatible with horizontal and vertical concatenation and with partial traces.","The generalised trace recovers integration along small diagonals for smooth kernels and the usual trace for finite-rank operators, so the framework covers standard QFT operations such as contraction and convolution in one algebraic package.","Because solar graphs are free, loops in Feynman graphs are handled by partial traces rather than by ad hoc orientation choices, removing the obstruction that made the free-PROP approach impractical for perturbative QFT.","In the zeta-value chapters, the same universal-property method constructs arborified zeta values and tree zeta values as algebra morphisms, yielding explicit rational-coefficient expressions of these generalisations in terms of ordinary multiple zeta values.","In resurgence theory, the truncated Wess-Zumino Schwinger-Dyson and renormalisation-group system has a Borel-Ecalle resummable two-point function analytic in a disk that escapes Dyson's argument."],"supporting_citations":[{"why":"Introduces TRAPs and the solar corolla-ordered graph construction whose freeness the thesis proves in Theorem 1.6.8.","marker":"[PCP20]"},{"why":"Establishes the relation between unital TRAPs and wheeled PROPs and provides Theorem 5.3.1, which the thesis cites to identify TRAPs with wheeled PROPs.","marker":"[CLS22]"},{"why":"Supplies the interpretation of Feynman rules as distribution-valued multivariate meromorphic germs that motivates Conjecture 1.8.11.","marker":"[Riv91]"},{"why":"Proves a version of the regularised-Feynman-amplitude conjecture for scalar QFT on closed Riemannian manifolds, the closest existing support for Conjecture 1.8.8.","marker":"[DZ21]"},{"why":"Gives recent analytic results on spaces related to distribution-valued meromorphic germs, which the thesis cites as reason to hope Conjecture 1.8.11 can be tackled.","marker":"[DPS22]"},{"why":"Introduced wheeled PROPs, the unitary counterpart of TRAPs, providing the reference point that locates TRAPs in the literature.","marker":"[Mer06]"},{"why":"Uses wheeled PROPs under the name 'ordinary graphs' and supplies the alternative viewpoint on solar graphs used in Remark 1.6.12.","marker":"[YJ15]"}],"fun_headline_variants":["Free TRAPs turn Feynman rules into canonical maps","TRAP algebra gives rigorous foundation for Feynman rules","How to make Feynman rules rigorous with TRAPs","Free TRAP theorem: canonical Feynman amplitudes","Replacing Feynman recipes with TRAP-traced maps"],"cache_read_input_tokens":70656,"weakest_assumption_plain":"The proposed construction of Feynman rules collapses if Conjecture 1.8.11 fails: the inductive limit of spaces of distribution-valued meromorphic germs with linear poles must be stable under the partial trace maps and hence carry a TRAP structure. The thesis identifies exactly this stability under partial traces as the main technical difficulty.","fun_headline_variants_meta":{"raw":{"variants":["Free TRAPs turn Feynman rules into canonical maps","TRAP algebra gives rigorous foundation for Feynman rules","How to make Feynman rules rigorous with TRAPs","Free TRAP theorem: canonical Feynman amplitudes","Replacing Feynman recipes with TRAP-traced maps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.00049,"raw_usage":{"total_tokens":2335,"prompt_tokens":797,"completion_tokens":1538,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":413,"completion_tokens_details":{"reasoning_tokens":1455}},"tokens_in":413,"tokens_out":1538,"duration_ms":11877,"temperature":1.0,"reasoning_tokens":1455,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-07T04:46:32.883859+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a Feynman integrand of a scalar quantum field theory on $\\mathbb{R}^d$, view it as a distribution-valued meromorphic germ in the regularisation parameters $z_1,\\dots,z_E$, and apply one partial trace $t_{i,j}$ that identifies an input and an output. If for some graph the result acquires a pole that is not linear, or leaves the space of linear-pole germs, then Conjecture 1.8.11 is false and the proposed rigorous definition of Feynman rules is invalid.","supporting_citations":[],"review_version":1}