{"id":"6582e0f5-df05-40aa-b6b3-6d2482874b6a","arxiv_id":"2608.03314","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For cyclic quiver representations and certain products of SL2 representations, the q-character of the nilpotent cone is computed by a Kostant-type partition formula, and a new class of Hesselink-type representations is introduced with a conjecture.","lead":"This paper proves a formula that counts the characters of functions on the nilpotent cone for several families of Lie algebra representations, extending a classical result of Hesselink. It also defines a new class of representations, called Hesselink-type, and conjectures a broader theory.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.2 as stated includes l-cycles for l>2, which are only conjectural; the proven acyclic cases do not cover them, so the headline theorem overclaims.","rationale":"The reader's weakest_assumption focuses on the combinatorial claim in Lemmas 3.2, 3.3, and 3.6 that every weight λ_I is either on a reflection hyperplane or in the W-orbit of ρ. I checked the small cases of the 3-hyperedge cancellation and the two-vertex case d1=d2+1, and the acyclic argument is internally coherent: the bounds on coordinate components, the cancellation of hyperplane weights by anti-invariance, and the induction in Lemma 3.3 all appear consistent, and the q^2 and q^4 terms for the 3-hyperedge cancel as claimed. So I do not have a concrete objection to the acyclic central claim itself. The genuine load-bearing problem is the gap between Theorem 1.2 as stated and what is proved: the theorem's case 4 is not restricted to acyclic extended quivers, but the paper explicitly leaves l-cycles with l>2 as a conjecture. This is not merely a wording issue, because the mechanism used for the proved cases—freeness of C[V^{ρ∨}] over C[V^{ρ∨}]^h—fails for l-cycles, and the only cycle case treated, l=2, needs an ad hoc syzygy resolution. The paper's own Remark 3.4 shows that nearby quiver representations can violate the core cancellation property, so the missing cases are not a formality. The reader's rationale already flags this mismatch, but their formal weakest_assumption identifies a different, more local combinatorial step; hence partial agreement. Since the reader's verdict is CONDITIONAL and this concern supports rather than overturns that verdict, I recommend no change to the verdict.","tokens_in":22533,"tokens_out":51105,"duration_ms":426839,"concrete_test":"Compute X_V symbolically for the l=3 extended quiver cycle, i.e. g=sl2^3 and V=⊕_{k=1}^{3} (V1)_k ⊠ (V1)_{k+1}, expanding through q-degree 6 in the basis of irreducible characters of sl2^3. The conjectural value forced by equation (2) is (1-q^6)/(1-q^3)=1+q^3. If the expansion contains any nontrivial irreducible character, such as χ_{V2}⊠trivial⊠trivial, then Theorem 1.2 as stated is false. If X_V=1+q^3 exactly, the claimed formula is at least consistent for this case, though a proof for l>2 would still be missing.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing gap is in the scope of Theorem 1.2, not in the acyclic combinatorial lemmas. In Section 3.4 the authors define extended quiver representations of trivial type recursively, and the basic extended quivers include an l-cycle for every l. Theorem 1.2, case 4, then asserts the q-character formula for all representations described in Section 3.4, which includes these l-cycles. The proof, however, establishes the formula only for acyclic extended quiver representations of trivial type in Proposition 3.7 and for the 2-cycle in Proposition 3.8. Immediately after Proposition 3.8 the text says that the authors 'expect' Theorem 1.2 to hold for all extended quiver representations of trivial type, 'including ones with l-cycles,' and notes that C[V^{ρ∨}] is not free over C[V^{ρ∨}]^h in those cases. Thus the l>2 cycles are explicitly unproved, and the missing part requires a genuinely different argument, not a routine extension of the lemmas used for the acyclic cases. Remark 3.4 strengthens the concern: for the cyclic quiver with 4 vertices and dimension vector (3,2,3,2), the key property that every λ_I lies on a reflection hyperplane or in Wρ fails, and (2) fails as a result. So the unproven l>2 cycles are not an innocuous technicality; the theorem as literally stated is strictly stronger than what is demonstrated.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a q-analog of the Kostant partition function and proposes a q-character formula for the ring of functions on the nilpotent cone of certain Lie algebra representations. The main theorem is proved for the adjoint representation, cyclic quiver representations with equal vertex dimensions, cyclic quiver representations with two vertices, and acyclic extended quiver representations of trivial type for products of SL2's. The proof reduces the formula to identity (2), which is verified case by case through combinatorial lemmas. The paper also defines a new class of Hesselink-type representations, gives a geometric interpretation via Borel-Weil-Bott, and states a conjecture relating Hesselink-type representations to cofree representations.","tokens_in":22787,"tokens_out":6219,"duration_ms":60555,"significance":"If the theorem is read with the limitation that the l>2 cycle cases are unproved, the paper still makes a solid contribution: Proposition 2.1 gives a clean reduction of the q-character formula to a combinatorial identity, and Lemmas 3.2, 3.3, and 3.6 provide explicit, case-by-case verifications for the acyclic families. The Hesselink-type condition and its geometric interpretation are useful new concepts. The paper is also commendably honest about the remaining conjectural l-cycle cases and about the counterexample in Remark 3.4. However, the main theorem as stated is broader than what is proved, and this scope mismatch is load-bearing for the central claim.","major_comments":[{"comment":"Theorem 1.2, case 4, states the formula for every representation described in Section 3.4. Section 3.4 defines extended quiver representations of trivial type recursively, and the basic cases include an l-cycle for every l. Proposition 3.7 proves the theorem only for acyclic extended quiver representations, and Proposition 3.8 establishes the case of the 2-cycle. The paragraph immediately after Proposition 3.8 states that the authors 'expect' Theorem 1.2 to hold for all extended quiver representations of trivial type, 'including ones with l-cycles,' and notes that C[V^{ρ∨}] is not free over C[V^{ρ∨}]^h in those cases. Thus the proof does not cover l-cycles for l>2, and the theorem as stated is strictly stronger than what is demonstrated. The statement should be restricted to the proven cases unless a proof for l>2 is supplied.","section":"1 (Theorem 1.2, case 4)"},{"comment":"Remark 3.4 shows that for the cyclic quiver with four vertices and dimension vector (3,2,3,2), the combinatorial property used in Lemma 3.2 fails: there is an I with λ_I neither on a reflection hyperplane nor in Wρ, and (2) fails. This is not by itself a counterexample to the sl2 l-cycle case, but it demonstrates that the absence of the combinatorial property is a genuine failure mode rather than a harmless technicality. Since the l>2 cycles are left unproved, the reader cannot infer from the methods of Section 3.4 that the theorem extends to them, and the authors' expectation is not a substitute for a proof.","section":"3.4 (after Proposition 3.8; Remark 3.4)"},{"comment":"In the proof of Lemma 3.6(1), the 3-hyperedge case is verified by the statement 'A short computation shows that the 2 element subsets cancel out in X_V.' This cancellation is a base case of the acyclic trivial-type family and is not obvious from the displayed weights; the computation should be written out or the relevant cancellation identity should be stated explicitly. As written, this basic case is asserted rather than verified.","section":"3.4 (Lemma 3.6(1), 3-hyperedge case)"}],"minor_comments":[{"comment":"Figure 2 is referenced in Section 3.4, but no image appears in the manuscript; please include the figure or remove the reference.","section":"3.4 (Figure 2)"},{"comment":"The sentence 'Through computer calculations, we expect Theorem 1.2 to hold for all extended quiver representations of trivial type, including ones with l-cycles' should be clearly marked as a conjecture, and the statement of Theorem 1.2 should be aligned with the proven cases.","section":"3.4 (after Proposition 3.7)"},{"comment":"In the l-cycle case of Lemma 3.5(1), the cofreeness and the structure of C[V]^G are justified by 'A similar computation to the previous case shows...'; please provide the details, since this statement is used in Lemma 3.6(1).","section":"3.4 (Lemma 3.5(1))"},{"comment":"In the proof of Proposition 3.8, the exactness of the displayed free resolution is justified by a brief sentence; a short explanation of the induction for the higher syzygies would improve readability.","section":"3.4 (Proposition 3.8)"},{"comment":"There are several formatting issues with spacing in mathematical expressions, such as 'theq-character' in the abstract and the opening sentence of the introduction; these should be corrected in the final version.","section":"Abstract"}],"recommendation":"major_revision","confidential_remarks":"The scope mismatch between Theorem 1.2 and the proofs is the main obstacle. It is, however, likely fixable by precise rewording: the abstract itself says 'acyclic extended quiver representation of trivial type,' while Theorem 1.2 case 4 does not include that qualifier. The authors should either prove the l>2 cycle cases or state the theorem with the acyclic hypothesis and present the l-cycle case as a conjecture. I would not recommend rejection, because the acyclic cases are substantial and the gap is clearly identified in the text; a major revision addressing the theorem statement and the missing details in the 3-hyperedge computation would be appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague —\n\nRead Krylov-Wang. The genuine contribution: they prove Hesselink-type q-character formulas for nilpotent cones of cyclic quiver reps with equal vertex dimensions, two-vertex cyclic quivers, and acyclic extended quiver reps of trivial type for products of SL2, and they introduce Hesselink-type representations with a conjecture and a Borel-Weil-Bott interpretation. Recovering Hesselink's classical formula in the adjoint case is a sanity check, not the point. The acyclic machinery is the point.\n\nThe paper is honestly written. No fitted parameters, no circular reasoning. P_q is defined from the weight multiset, M_q is an alternating sum, and the proof reduces the theorem to a combinatorial identity via the cofree reduction in Prop 2.1. The lemmas in 3.2 and 3.3 are detailed: Lemma 3.2 carefully walks through the cancellation and the W-orbit structure. The leaf-attachment and disjoint-union arguments in Lemma 3.6 are clean. The paper also flags the cyclic quiver counterexample in Remark 3.4, which is the right kind of intellectual honesty.\n\nThe soft spot is real and load-bearing in scope, not in the acyclic core. Theorem 1.2, case 4, claims the formula for all representations described in Section 3.4, and Section 3.4 includes l-cycles for every l. What is actually proved for that family is the acyclic cases (Prop 3.7) and the 2-cycle (Prop 3.8). After Prop 3.8 the authors say they \"expect\" the theorem to hold for l-cycles and note that C[V^{ρ∨}] is not free over C[V^{ρ∨}]^h there. The theorem statement therefore overclaims. This is not a routine gap: Remark 3.4 gives a concrete 4-vertex quiver with dimension vector (3,2,3,2) where the key combinatorial property fails. So the l>2 cycles are a genuine open part.\n\nOne more flag, minor: the syzygy resolution in Prop 3.8 is credited to ChatGPT. It is checkable and the idea is sensible, but independent verification would be appropriate before relying on it.\n\nVerdict: conditional acceptance territory. The acyclic cases are proven. The introduction should be revised so that Theorem 1.2 states exactly the proved cases and the l>2 cycles are moved to a conjecture. The Hesselink-type framework and Conjecture 1.6 are worth airing.\n\nWho benefits: anyone working on graded Lie algebras, quiver varieties, and q-analogs of Kostant partition functions. It deserves a serious referee, and I would like to see the revised version.","headline":"Solid, honest q-character paper with one real scope problem: the headline theorem claims l-cycles that only the 2-cycle is proved.","tokens_in":23370,"tokens_out":1762,"would_cite":true,"duration_ms":15882,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["17B10","16G20","05A30"],"pacs":[],"model":"deepseek-v4-flash","headline":"For four families of Lie algebra representations, the nilpotent cone's q-character is a Weyl-alternating sum of a q-analog of the Kostant partition function.","keywords":["q-character","nilpotent cone","cyclic quiver","extended quiver representation","Kostant partition function","Hesselink-type representation","cofree representation","Weyl character formula"],"falsifier":"Expand the numerator of $X_V$ for a covered representation not worked out in the paper, say the two-vertex cyclic quiver with dimensions $(d_1,d_2)=(5,3)$, and antisymmetrize over the Weyl group: if any surviving coefficient sits at a weight that is neither fixed by a reflection nor in the Weyl orbit of $\\rho$, then identity (2) fails and Theorem 1.2 would not follow. The paper's own counterexample for the 4-vertex cyclic quiver with dimension vector $(3,2,3,2)$ is the template, since that expansion produces exactly such uncancelled exceptional weights.","tokens_in":22268,"feed_emoji":"🧮","tokens_out":14984,"duration_ms":120522,"temperature":0.7,"pith_summary":"The paper proves a q-character formula for functions on the nilpotent cone of a Lie algebra representation in four families: the adjoint representation, cyclic quiver representations with equal vertex dimensions, cyclic quiver representations with two vertices, and acyclic extended-quiver representations of trivial type built from copies of sl2. In every case the q-character is $\\sum_{\\lambda\\in\\Lambda^+} M_q(\\lambda)\\chi_\\lambda$, where $M_q(\\lambda)$ is the Weyl alternating sum of a q-analog of the Kostant partition function that counts certain weakly negative integer-valued functions on the weight multiset. The common engine is a single identity, labeled (2), asserting that a normalized antisymmetrized product of weight factors equals the Hilbert-series ratio $\\dim_q C[V^{\\rho^\\vee}]^h / \\dim_q C[V]^G$. That identity also motivates a new class of representations the authors call Hesselink-type, for which they conjecture a connection to cofree representations and give a geometric interpretation via Borel-Weil-Bott. If the main formula is right, it shows that several seemingly different representation classes obey one uniform character-theoretic law.","feed_headline":"One identity yields nilpotent-cone q-characters for four families","feed_subtitle":"The formula extends the classical nilpotent-cone character formula to cyclic quivers and product-SL2 representations.","key_machinery":"The load-bearing mechanism is identity (2), $X_V = \\dim_q C[V^{\\rho^\\vee}]^h/\\dim_q C[V]^G$, where $$X_V := \\frac{\\sum_{w\\in W}(-1)^{\\ell(w)}$t^{{w\\rho}}$\\prod_{\\$\\alpha$\\in S_+}(1-$qt^{{-w\\alpha}}$)}{\\sum_{w\\in W}(-1)^{\\ell(w)}$t^{{w\\rho}}$}$$ and $S_+$ is the multiset of weights of $V$ pairing positively with $\\rho^\\vee$. For a cofree representation, Proposition 2.1 converts this identity into the q-character formula of Theorem 1.2; the definition of a Hesselink-type representation is exactly that $X_V$ is a $\\mathbb{Z}[q]$-multiple of the trivial character. The proof of (2) in each family is carried by a combinatorial claim: in the expansion of the numerator of $X_V$, every weight $\\lambda_I = \\rho - \\sum_{\\alpha\\in I}\\alpha$ with $I\\subset S_+$ is either fixed by a reflection of the Weyl group (so its coefficient cancels under antisymmetrization) or lies in the Weyl orbit of $\\rho$, and the surviving coefficients are controlled by the Coxeter-group Poincar\\'e series $\\sum_{w\\in W}q^{\\ell(w)}$. For the product-of-$\\mathfrak{sl}_2$ case the same identity is propagated structurally through extended quivers, whose basic pieces are framed vertices, edges, self-loops, and 3-hyperedges, with leaf attachment and disjoint union as composition operations.","core_discovery":"On the paper's own terms, the discovery is Theorem 1.2: for the adjoint representation, cyclic quiver representations with equal vertex dimensions, cyclic quiver representations with two vertices, and acyclic extended-quiver representations of trivial type of a product of copies of $\\mathfrak{sl}_2$, the q-character of functions on the nilpotent cone $\\mathcal{N}_V$ has the form $\\operatorname{ch}_q C[\\mathcal{N}_V] = \\sum_{\\lambda\\in\\Lambda^+} M_q(\\lambda)\\chi_\\lambda$, with $M_q(\\lambda) = \\sum_{w\\in W}(-1)^{\\ell(w)}P_q(w(\\lambda+\\rho)-\\rho)$ and $P_q$ the q-analog of the Kostant partition function defined by counting weakly negative functions on the weight multiset $S$. This reproduces the classical nilpotent-cone formula when $V$ is the adjoint representation. The central technical statement behind it is identity (2), $X_V = \\dim_q C[V^{\\rho^\\vee}]^h/\\dim_q C[V]^G$, where $X_V$ is the normalized Weyl-antisymmetrized product over the positive weights of $V$; for a cofree $V$, Proposition 2.1 shows this identity is equivalent to the character formula. The paper further defines a representation to be Hesselink-type exactly when $X_V$ is a $\\mathbb{Z}[q]$-multiple of the trivial character, proves all four families have this property, and conjectures that for semisimple $G$ the Hesselink-type condition implies cofreeness and the same q-character formula.","pith_inferences":["Our inference: the 'every $\\lambda_I$ is either on a wall or in $W\\rho$' condition is a natural combinatorial invariant to test on any new representation; it is directly checkable by expansion, and the paper's $(3,2,3,2)$ cyclic quiver shows the failure mode appears in small examples.","Our inference: if Conjecture 1.6 holds, the Hesselink-type condition would provide a character-theoretic characterization of a large class of cofree representations, with the Borel-Weil-Bott description turning the formula into a statement about vanishing of higher cohomology for exterior-algebra bundles on the flag variety.","Our inference: Proposition 3.8's syzygy computation for the 2-cycle suggests that identity (2) can survive even when $C[V^{\\rho^\\vee}]$ is not free over its $h$-invariants; applying the same free-resolution technique to other $l$-cycles would be a direct way to test Conjecture 3.11.","Our inference: the parity dependence seen in the two-vertex case suggests a general theorem for cyclic quivers would need a dimension-vector condition finer than equality of dimensions, possibly expressible in terms of differences $d_k - d_{k+1}$."],"forward_implications":["For cyclic quivers with equal vertex dimensions, the q-character of the nilpotent cone is computable from the explicit Hilbert-series ratio $\\prod_{i=1}^N(1-q^{il})/(1-q^l)$ times the Weyl denominator, giving a closed form in terms of the dimension $N$ and the number of vertices $l$.","For two-vertex cyclic quivers, the formula depends on the two dimensions only through the smaller dimension and the parity of the larger one, so all such quivers fall under Theorem 1.2.","For acyclic extended quiver representations of trivial type of a product of $\\mathfrak{sl}_2$'s, Theorem 1.2 holds, and the proof gives a recursive recipe for computing the q-character from the diagram's leaf structure.","All representations appearing in Theorem 1.2 are Hesselink-type; if Conjecture 1.6 is correct, Hesselink-type representations of semisimple groups are cofree and satisfy the same q-character formula, making the class the natural home of the formula.","For $\\mathfrak{g}=\\mathfrak{sl}_2$, the only Hesselink-type representations are direct sums of extended quiver representations of trivial type and copies of the trivial representation, closing the classification in that case."],"supporting_citations":[{"why":"It supplies the classical q-character formula for the ordinary nilpotent cone that Theorem 1.2 generalizes.","marker":"[4]"},{"why":"It introduces the q-analog of the Kostant partition function whose Weyl alternating sum defines $M_q(\\lambda)$.","marker":"[9]"},{"why":"It computes the invariants and exponents for cyclic-quiver and $\\mathbb{Z}/l\\mathbb{Z}$-graded representations, giving $\\dim_q C[V]^G$ in the quiver cases.","marker":"[18]"},{"why":"It provides the background on cofree representations used in Proposition 2.1 and Lemma 2.3.","marker":"[16]"},{"why":"It supplies the Coxeter-group Poincar\\'e-series identity used to identify the coefficient of $t^\\rho$ in the adjoint and equal-dimension cyclic quiver cases.","marker":"[12]"},{"why":"It gives the standard identity $\\sum_{w\\in W}q^{\\ell(w)}=\\prod(1-q^{m_i+1})/(1-q)$ that closes the coefficient computation in the adjoint case.","marker":"[5]"},{"why":"It establishes cofreeness for the 3-hyperedge and other basic extended-quiver pieces used in Section 3.4.","marker":"[7]"},{"why":"It performs the Chevalley restriction computation for cyclic quivers that yields the invariant Hilbert series used in Section 3.2.","marker":"[2]"}],"fun_headline_variants":["q-character formula extends to cyclic quiver families","Nilpotent cone q-characters via identity (2)","Hesselink-type reps: q-character proof and conjecture","Four rep families share nilpotent-cone q-character formula","Extending Hesselink's formula to quiver and sl2 reps"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof of the q-character formula rests on a combinatorial claim about the weight multiset of $V$: in the expansion of the numerator of $X_V$, every weight $\\lambda_I = \\rho$ minus a subset of the positive weights must be either fixed by a reflection (so its coefficient cancels) or equivalent to $\\rho$ under the Weyl group, and the surviving terms must assemble into the known Poincar\\'e series; if any covered family produced an exceptional weight like the one the paper exhibits for the cyclic quiver with dimensions $(3,2,3,2)$, identity (2) and the formula would not follow from the reduction.","fun_headline_variants_meta":{"raw":{"variants":["q-character formula extends to cyclic quiver families","Nilpotent cone q-characters via identity (2)","Hesselink-type reps: q-character proof and conjecture","Four rep families share nilpotent-cone q-character formula","Extending Hesselink's formula to quiver and sl2 reps"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000263,"raw_usage":{"total_tokens":1642,"prompt_tokens":1030,"completion_tokens":612,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":646,"completion_tokens_details":{"reasoning_tokens":528}},"tokens_in":646,"tokens_out":612,"duration_ms":5271,"temperature":1.0,"reasoning_tokens":528,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T14:52:53.946636+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Expand the numerator of $X_V$ for a covered representation not worked out in the paper, say the two-vertex cyclic quiver with dimensions $(d_1,d_2)=(5,3)$, and antisymmetrize over the Weyl group: if any surviving coefficient sits at a weight that is neither fixed by a reflection nor in the Weyl orbit of $\\rho$, then identity (2) fails and Theorem 1.2 would not follow. The paper's own counterexample for the 4-vertex cyclic quiver with dimension vector $(3,2,3,2)$ is the template, since that expansion produces exactly such uncancelled exceptional weights.","supporting_citations":[{"cited_title":"Characters of the nullcone","cited_arxiv_id":null,"evidence_quote":"It supplies the classical q-character formula for the ordinary nilpotent cone that Theorem 1.2 generalizes."},{"cited_title":"Singularities, character formulas, and aq-analog of weight multiplicities","cited_arxiv_id":null,"evidence_quote":"It introduces the q-analog of the Kostant partition function whose Weyl alternating sum defines $M_q(\\lambda)$."},{"cited_title":"The Weyl group of a graded Lie algebra","cited_arxiv_id":null,"evidence_quote":"It computes the invariants and exponents for cyclic-quiver and $\\mathbb{Z}/l\\mathbb{Z}$-graded representations, giving $\\dim_q C[V]^G$ in the quiver cases."},{"cited_title":"Invariant theory","cited_arxiv_id":null,"evidence_quote":"It provides the background on cofree representations used in Proposition 2.1 and Lemma 2.3."},{"cited_title":"The Poincar´ e series of a Coxeter group","cited_arxiv_id":null,"evidence_quote":"It supplies the Coxeter-group Poincar\\'e-series identity used to identify the coefficient of $t^\\rho$ in the adjoint and equal-dimension cyclic quiver cases."},{"cited_title":"Cambridge Studies in Advanced Mathematics","cited_arxiv_id":null,"evidence_quote":"It gives the standard identity $\\sum_{w\\in W}q^{\\ell(w)}=\\prod(1-q^{m_i+1})/(1-q)$ that closes the coefficient computation in the adjoint case."},{"cited_title":"Koregul¨ are und ¨ aquidimensionale Darstellungen","cited_arxiv_id":null,"evidence_quote":"It establishes cofreeness for the 3-hyperedge and other basic extended-quiver pieces used in Section 3.4."},{"cited_title":"Chevalley restriction theorem for the cyclic quiver","cited_arxiv_id":null,"evidence_quote":"It performs the Chevalley restriction computation for cyclic quivers that yields the invariant Hilbert series used in Section 3.2."}],"review_version":2}