{"id":"d8328d47-75ff-41c5-b16a-f967cd36a679","arxiv_id":"2501.19096","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":4,"one_line_summary":"Khovanov-Rozansky gl_N link homology carries a functorial action of the positive Witt algebra, making link cobordisms equivariant maps between twisted homology groups.","lead":"This paper builds an action of the positive Witt algebra on Khovanov-Rozansky gl_N link homology, a central knot invariant. The action is shown to be compatible with link cobordisms, giving the invariant extra algebraic structure and explicit computations for Hopf and torus links.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The central construction inherits all weight from the cited QRSW24 Theorem 4.4; if L_n fails to preserve any state-space relation, Reidemeister invariance and functoriality collapse.","rationale":"The reader's weakest_assumption identifies exactly the same load-bearing point: the W∞-action on state spaces is imported from [QRSW24, Theorem 4.4] and is not re-proved against the full set of foam relations, especially dot migration. Our stress-test confirms that this is the critical dependency: every later equivariance computation and Reidemeister-move proof in the paper assumes the action descends to FN(Γ). If that descent fails, the central claims collapse, and no amount of subsequent computation can repair them. The paper does include substantial independent work: explicit formulas for L_n, a proof of the Witt relation on saddles, the construction of twist data, and elaborate short-exact-sequence arguments. The relative homotopy category formalism (Section 2, Appendix A) is also sound. But these pieces only work if Theorem 3.7 is valid in the precise setting used here, including the generalized decorations Sym_{ℓ}⊗Sym_{N-ℓ} mentioned in the footnote. Our proposed test would directly settle the issue for the relations that are actually invoked. The abstract's unsupported claim about Lee homology and genus bounds is a separate, non-central issue and does not affect the main theorem's correctness risk. We therefore keep the reader's CONDITIONAL verdict unchanged.","tokens_in":136,"tokens_out":7921,"duration_ms":147824,"concrete_test":"Independently verify the descent of L_n for the state-space relations actually used. Concretely, fix N=2 and take a basic foam containing a decoration on one facet adjacent to a vertex. Compute L_n of the foam directly from (9)-(17) for n = -1,0,1,2. Then apply the dot-migration relation [RW20a, (11)] to move the decoration across the vertex, and compute L_n of the resulting foam. If the two results differ for any n, Theorem 3.7 fails for the relations used in this paper, so Theorems 5.1 and 6.4 are unsupported. If they agree, repeat for the theta relation used in Lemma 4.9 and the rotation relation behind Lemma 4.10; agreement on all three relations would settle the main correctness concern.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorems 5.1 and 6.4 assert invariance and functoriality in the relative homotopy category of W∞-modules. The objects KRWN(L) are chain complexes whose terms are state spaces FN(Γ) with a W∞-action. That action is defined by the local operators L_n in equations (9)-(17), extended to arbitrary foams by the Leibniz rule. For L_n to act on FN(Γ), it must be well-defined modulo every foam-evaluation relation that defines FN(Γ), in particular the dot-migration relation [RW20a, (11)] and the theta/rotation relations used in Lemmas 4.9-4.11. The paper does not prove this compatibility; it cites [QRSW24, Theorem 4.4] and restates it as Theorem 3.7. Proposition 3.6 only checks the Witt bracket on basic foams and explicitly refers to [QRSW24, p.24] for the cup. The later equivariance arguments (Lemmas 4.13-4.17, 5.2, 5.8, and the Reidemeister proofs in Section 5) all compute L_n on foams and use state-space identities such as dot migration to move twists. If any of those identities is not preserved by L_n, then the operators do not descend to FN(Γ), the braiding complexes (29)-(30) are not complexes of W∞-modules, and the short exact sequences used to prove Reidemeister invariance are invalid. The reduction to good-position foams via [QW24] is likewise cited, not proved. This is not an internal contradiction, but it is the single place where correctness risk is concentrated: the entire paper rests on one imported compatibility statement that is not re-derived here.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper constructs a W∞_{-1}-action on Khovanov–Rozansky gl_N link homology, extending the sl_2-action of Qi–Robert–Sussan–Wagner. The authors define local foam operators L_n following [QRSW24], introduce red-dot twists, build braiding complexes in a relative homotopy category of W∞_{-1}-modules, and prove Reidemeister invariance and functoriality for framed links. They also compute explicit bases and actions for T(2,m) torus links and give structural results for N=2, including irreducibility statements for certain W∞_{-1}-modules.","tokens_in":28868,"tokens_out":4987,"duration_ms":58294,"significance":"If correct, the paper establishes a genuinely new infinite-dimensional symmetry of Khovanov–Rozansky homology that goes beyond the previously known sl_2 action. The main theorems are substantial: every framed oriented link is assigned a W∞_{-1}-module structure that is invariant up to isomorphism in the relative homotopy category, and the construction is functorial for a suitable category of cobordisms with red-dot labels. The explicit computations for T(2,m) links and the N=2 structural analysis are valuable and consistent with known sl_2 structures; the paper is careful to present diagrammatic constructions and to credit the foam-action machinery to [QRSW24]. The main risk is concentration of correctness in an imported compatibility theorem and in several equivariance checks that are left to the reader.","major_comments":[{"comment":"The entire construction rests on the statement that the local operators L_n descend to the state space F_N(Γ), i.e. that they preserve every foam-evaluation relation defining F_N(Γ), in particular the dot-migration relation [RW20a, (11)] and the theta/rotation relations used in Lemmas 4.9–4.11. Proposition 3.6 only verifies the Witt bracket on basic foams before taking the quotient, and the cup case is referred to [QRSW24, p.24]. Since Theorems 5.1 and 6.4 depend on this compatibility, the paper should either prove it or give a precise citation to the exact statement and proof in [QRSW24] covering the full set of state-space relations. Without this, a failure of L_n to preserve even one relation would invalidate the braiding complexes (29)–(30) and the Reidemeister arguments in Section 5.","section":"Section 3.3, Theorem 3.7"},{"comment":"Several equivariance statements that are load-bearing for the invariance proof are asserted with proofs left to the reader or described as analogous: Lemmas 4.14–4.17, the W∞_{-1}-equivariance of all maps in the short exact sequences (32) and (33), the Maurer–Cartan twists (1)–(2) in Lemma 5.2, and the κ∘σ equivariance in Proposition 5.10. These morphisms are exactly the ones used to obtain the Reidemeister isomorphisms via Corollary 2.3, so their equivariance is not a cosmetic detail. The authors should include the missing checks or give a precise reduction to the computations already displayed.","section":"Section 4, Lemmas 4.14–4.17; Section 5, Lemma 5.2 and Proposition 5.10"},{"comment":"The proof of the sliding lemma introduces a short exact sequence (32) whose splitting is asserted to be W∞_{-1}-equivariant only after 'the verification is left to the reader.' The same applies to the null-homotopy of xC'' in the relative category. Because Lemma 5.2 is used repeatedly in Corollaries 5.4–5.6 and in the Reidemeister I and II arguments, the reader cannot fully verify Theorem 5.1 without reconstructing these computations. Please provide the missing details or an explicit reference to a place where they appear.","section":"Section 5.2, Lemma 5.2"}],"minor_comments":[{"comment":"The passage from framed to unframed invariance is very brief: the statement that a solid red dot of label 1/2 can be added to each crossing and then slid and combined into at most one twist per component needs a few more sentences or a diagrammatic justification, especially since the unframed complexes (57)–(58) are asserted to be invariant under all Reidemeister moves.","section":"Section 5.6"},{"comment":"The abstract promises that 'the state spaces of simple webs are identified with standard representations of the Witt algebra on polynomials,' but the body does not contain an explicit statement or theorem to this effect. The computations for the circle and theta webs in Section 7 partially support this, but the identification should be stated and proved or the abstract should be adjusted.","section":"Abstract and Section 7"},{"comment":"There are several typos and formatting infelicities, including 'For :C(A#H)→C (A)' in Section 2.1, 'projbigr' in Corollary 6.6, and inconsistent use of math mode for gl_N in the introduction. These do not affect the mathematics but should be corrected.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The paper is mathematically serious and the central construction is coherent, but the authors should be asked to make the dependence on [QRSW24, Theorem 4.4] explicit and to fill the omitted equivariance checks that support the Reidemeister invariance. The concerns raised by the stress-test are legitimate: the correctness of the whole paper is concentrated in the descent of L_n to state spaces, and the manuscript currently relies on an imported theorem without re-deriving the key compatibility with dot migration. This is a fixable issue within the scope of the paper, not a fatal one. There is no evidence of circularity: the W∞ action is imported from prior work, and the torus-link computations are confirmatory rather than parameter-fitted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The real news here is the red-dot twist formalism and the functorial statement: framed oriented links carry a W-infinity_{-1}-module structure that survives Reidemeister moves in the relative homotopy category, and the construction is made explicit enough to compute the Hopf link and T(2,m) torus links. That is a substantial enrichment of Khovanov–Rozansky homology, and the paper does the honest work of writing down bases and actions. The section on the Witt structure of KRW2(∅) is genuinely illuminating, and the appendix on the Verdier quotient is a useful service for readers who want the relative homotopy category spelled out.\n\nThe soft spot is exactly where the stress-test note points. Theorem 3.7, imported from QRSW24, is the whole foundation: the operators L_n descend to the state space FN(Γ) modulo all foam-evaluation relations, including dot migration. The paper does not re-derive that compatibility, and the Reidemeister and functoriality arguments all lean on it. I would not call this a fatal flaw — the cited theorem is published and the authors seem to know it well — but it is a concentrated risk, and a referee should be asked to check whether the compatibility with [RW20a, (11)] really was established in QRSW24 or only asserted there too. Relatedly, several equivariance lemmas in Section 4 are left to the reader (4.14–4.17, and the splitting checks in 5.2 and 5.8). For a paper whose invariance proof depends on those checks, this is more than a cosmetic omission; it is the part that needs verification.\n\nThe abstract overreaches. It promises “some simple relations to Lee homology and genus bounds,” but the body's structural results are about sl2 and Witt decompositions, highest weight vectors, and irreducibility — I do not find a genus-bound argument or an actual Lee-homology statement anywhere. That should either be added or cut from the abstract.\n\nOn circularity: I side with the reader's low burden. The W∞ action is imported from QRSW24, not derived from the target theorem, and the torus-link computations confirm known sl2 structures rather than fit parameters. The self-citation [Roz23] is only used for a prior T(2,k) computation, which is fine.\n\nWho is this for? Knot theorists working on link homology symmetries, and people who want to use a Witt action to decompose skein lasagna modules. It deserves a serious referee, and I would recommend sending it out with instructions to verify the imported compatibility theorem and the delegated equivariance checks, plus a request to fix the abstract. If those checks pass, this is a solid paper and I would cite it.","headline":"A plausible and well-written enrichment of gl_N link homology whose main risk is a single load-bearing citation to QRSW24; worth a serious referee, with the abstract's Lee-homology/genus-bound claims needing attention.","tokens_in":29427,"tokens_out":2163,"would_cite":true,"duration_ms":25970,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["57K18","57K16","17B10","18N25","18G35"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper constructs an action of the positive part of the Witt algebra on Khovanov-Rozansky gl_N link homology and proves it is invariant under Reidemeister moves and functorial for link cobordisms.","keywords":["Witt algebra action","Khovanov-Rozansky homology","gl_N link homology","foam evaluation","relative homotopy category","link cobordisms","torus links","symmetric polynomials"],"falsifier":"Choose a small web, such as the theta web with thicknesses (1,1,2), and compute L_n on both sides of the dot-migration relation from [RW20a, eq. (11)]; if the two results differ for any n, the state-space action does not exist and Theorems 5.1 and 6.4 lose their foundation. Alternatively, run the explicit Reidemeister II complex comparison of Proposition 5.9 for small N and check whether the two complexes are genuinely isomorphic in the relative homotopy category rather than merely homotopy equivalent after forgetting the module structure.","tokens_in":28293,"feed_emoji":"🪢","tokens_out":10527,"duration_ms":101288,"temperature":0.7,"pith_summary":"The paper aims to show that Khovanov-Rozansky gl_N link homology carries a substantial symmetry: an action of the subalgebra of the Witt algebra generated by the operators L_n for n at least -1. This action is built locally on foams, and the authors prove that the resulting chain complex for any framed oriented link is unchanged by Reidemeister moves, up to isomorphism in a relative homotopy category. They also prove that link cobordisms induce equivariant maps between twisted homologies, once the source and target links are decorated with bookkeeping dots. If correct, every framed link acquires a module structure over an infinite-dimensional Lie algebra, refining previously known sl2 symmetries and giving explicit presentations for Hopf and (2,m) torus links. State spaces of simple webs are identified with the standard Witt-algebra action on symmetric polynomials, which is what makes the torus-link computations possible.","feed_headline":"Witt algebra acts on Khovanov-Rozansky link homology","feed_subtitle":"Every framed link acquires a Witt-algebra module structure that respects Reidemeister moves and cobordisms.","key_machinery":"The load-bearing object is the local W_{\\infty}^{-1}-action on foams: for each generator L_n, formulas (9)-(17) specify its effect on polynomial decorations and on each basic foam, with parameters s, $\\lambda$, and mu, and the action extends by the Leibniz rule to foams in good position. A second mechanism is the relative homotopy category, the Verdier quotient by complexes that become null-homotopic once the W_{\\infty}^{-1}-module structure is forgotten; this absorbs the non-equivariant homotopies that appear in Reidemeister-move proofs. A third mechanism is twisting by red dots: Maurer-Cartan elements attached to edges deform the module structure, and dot-migration lemmas allow these dots to slide, so that cobordism maps can be made equivariant. The state spaces themselves are quotients by foam-evaluation relations, and the cited theorem that the local L_n respect those relations is what converts local formulas into a global link invariant.","core_discovery":"The central discovery is that explicit local operators L_n, defined by equations (9)-(17) on basic foams by acting on polynomial decorations through L_n(Q) = -sum_i $x_i^{{n+1}}$ partial Q / partial x_i and on cups, caps, saddles, zips, unzips, digons, and associativity foams, descend to a well-defined action of W_{\\infty}^{-1} on the gl_N state space of every web. The action can be twisted by Maurer-Cartan elements encoded as red dots on edges, and these twists make the crossing braiding complexes equivariant. The main theorems state that the complex KRWN(L) is a framed link invariant in the relative homotopy category and that KRWN is a functor from the category of framed links with red-dot labels to the homotopy category of W_{\\infty}^{-1}-modules whose cobordism images are equivariant. Concretely, the homology of the positive Hopf link and of T(2,m) torus links is given with explicit bases on which every L_n acts by explicit formulas, and for N=2 the resulting module structure is analyzed completely, including cases where it is irreducible.","pith_inferences":["A natural extension the paper does not pursue is to transfer the W_{\\infty}^{-1} action to Skein lasagna modules; if the functoriality behaves as stated, the infinite-dimensional state spaces of these 4-manifold invariants would decompose into weight spaces, potentially making them more computable.","The parameters s, lambda, and mu suggest a family of module structures rather than a single one; varying them may interpolate between known link-homology symmetries, and whether the resulting invariants are independent of these choices is left open.","The explicit torus-link bases make it possible to test whether the advertised relations to Lee homology and genus bounds are W_{\\infty}^{-1}-equivariant; this would turn those structural claims into concrete spectral-sequence or inequality statements.","Because invariance is proved only in the relative homotopy category, a natural open question is whether the W_{\\infty}^{-1} module structure itself, rather than merely the underlying complex, is independent of the red-dot bookkeeping choices."],"forward_implications":["The complex KRWN(L) is a framed link invariant in the relative homotopy category, and taking homology yields an unframed invariant together with a genuine W_{\\infty}^{-1} # R_N-module structure on Khovanov-Rozansky homology.","Connected link concordances induce W_{\\infty}^{-1}-equivariant maps, because their Euler characteristic is zero and no red-dot bookkeeping is required.","The action restricts to the sl2 action generated by L_{-1}, 2L_0, and -L_1, so the previously studied sl2 symmetries of gl_N-homology become special cases of the Witt action.","State spaces of circle and theta webs carry the standard polynomial representation of W_{\\infty}^{-1}, yielding explicit formulas for the action on T(2,m) torus links and a full description of the N=2 case.","The degree-raising property deg_N(L_n F) = deg_N(F) + 2n provides a new grading filtration on the homology that can be used for structural analysis."],"supporting_citations":[{"why":"Supplies the W_{\\infty}^{-1}-action on foams and the theorem, reproduced here as Theorem 3.7, that the local operators L_n descend to a well-defined action on state spaces.","marker":"[QRSW24]"},{"why":"Constructs the sl2 action, the red-dot twist formalism, and the equivariant morphism lemmas that this paper extends to W_{\\infty}^{-1}.","marker":"[QRSW23]"},{"why":"Provides the foam-evaluation and dot-migration relations that define the state spaces and are used throughout the equivariance checks.","marker":"[RW20a]"},{"why":"Establishes functoriality of colored Khovanov-Rozansky homology, the basis for the functor from rLinks to the homotopy category of modules.","marker":"[ETW18]"},{"why":"Origin of the positive Witt-algebra action on triply graded HOMFLY-PT homology and of the sliding-lemma argument reused in Lemma 5.2.","marker":"[KR16]"},{"why":"Shows every foam is isotopic to a foam in good position, which the Leibniz-rule extension of the local operators requires.","marker":"[QW24]"},{"why":"Provides the earlier sl2 action on T(2,k) torus links that the torus-link computations here extend.","marker":"[Roz23]"},{"why":"Supplies the Schur-polynomial identities used to express the action of L_n on the explicit torus-link bases.","marker":"[Mac95]"}],"fun_headline_variants":["Khovanov-Rozansky homology is a Witt module","Explicit Witt action on link homology","Link homology gains Witt algebra structure","Twisted Witt action respects link cobordisms","Khovanov-Rozansky homology carries a Witt action"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The whole construction depends on the cited result that the local operators L_n are compatible with every foam-evaluation relation used to build the state spaces; this compatibility is not re-derived here, so if any such relation is not preserved, the Reidemeister-invariance and functoriality arguments collapse.","fun_headline_variants_meta":{"raw":{"variants":["Khovanov-Rozansky homology is a Witt module","Explicit Witt action on link homology","Link homology gains Witt algebra structure","Twisted Witt action respects link cobordisms","Khovanov-Rozansky homology carries a Witt action"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000735,"raw_usage":{"total_tokens":3235,"prompt_tokens":845,"completion_tokens":2390,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":461,"completion_tokens_details":{"reasoning_tokens":2320}},"tokens_in":461,"tokens_out":2390,"duration_ms":19463,"temperature":1.0,"reasoning_tokens":2320,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-09T21:20:11.222944+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Choose a small web, such as the theta web with thicknesses (1,1,2), and compute L_n on both sides of the dot-migration relation from [RW20a, eq. (11)]; if the two results differ for any n, the state-space action does not exist and Theorems 5.1 and 6.4 lose their foundation. Alternatively, run the explicit Reidemeister II complex comparison of Proposition 5.9 for small N and check whether the two complexes are genuinely isomorphic in the relative homotopy category rather than merely homotopy equivalent after forgetting the module structure.","supporting_citations":[],"review_version":1}