{"id":"f6282962-b72a-4c6c-82a9-99350596fe41","arxiv_id":"2607.03617","paper_version":1,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Elliptic genera and lower-dimensional BPS indices equal discrete sums over molecule boundaries in crystals defined by Jeffrey-Kirwan residues, generalizing Nekrasov Young-diagram formulas.","lead":"Exact formulas count BPS states in supersymmetric quiver theories as sums over crystal molecules, with each weight fixed by the molecule's boundary atoms. This generalizes the Nekrasov instanton sum and shows how Calabi-Yau geometry can emerge from discrete crystal data in a thermodynamic limit.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified beyond the already-flagged no-overlap hypothesis.","rationale":"The Reader correctly isolates the no-overlap condition as the single weakest assumption supporting the strongest claim. The derivation of the residual boundary factors (cancellation mechanisms in §§3.2 and 4.2) is otherwise careful and recovers the known Nekrasov shell formula and the one-dimensional Jordan-quiver result. The thermodynamic-limit discussion is provisional by the authors’ own admission and does not underwrite the exact formulae. Because the paper already states the hypothesis, restricts to cyclic chambers, and checks the formulae on concrete examples that satisfy no-overlap, the concern does not move the verdict. A direct low-rank verification on Q^{1,1,1} would settle residual doubt without requiring a general proof.","tokens_in":32283,"tokens_out":475,"duration_ms":4969,"concrete_test":"Take the N=2 Q^{1,1,1} quiver of §5.3. Enumerate all admissible JK poles for total rank N=3 by direct residue computation (or by computer algebra). Verify that every pole yields three distinct atoms and that the resulting boundary product (4.26) exactly reproduces the three explicit Z_3 expressions given in the text. If any pole produces a repeated atom or a mismatched numerical value, the no-overlap hypothesis fails for this toric example.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (grand-canonical Z equals sum over molecules of boundary zeta products, eqs. (3.18)–(3.19) and (4.21)–(4.26)) rests on the no-overlap condition of §2.1: every admissible JK pole produces |A(u*)|=rank distinct atoms. The paper states this as a working hypothesis rather than a theorem for arbitrary quivers. If two distinct hyperplanes force the same linear combination of fugacities, the atom set collapses, the melting rule becomes ill-defined, and the residual boundary factors after cancellation are no longer correctly enumerated. The authors already flag the restriction; the examples (Jordan quiver, Young-diagram quiver, Q^{1,1,1}) all satisfy no-overlap, so the formulae are verified where they are claimed. No deeper internal inconsistency appears.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper derives exact formulae for the elliptic genera of 2d N≥2 quiver gauge theories (and their 1d Witten indices and 0d matrix-model partition functions) by evaluating Jeffrey-Kirwan residues and organising the non-vanishing poles into crystals and molecules. The grand-canonical partition function is expressed as a sum over molecules M of a weight that, after systematic cancellation of one-loop factors, depends only on boundary data of M (eqs. (3.18)–(3.19) for N=4 and (4.21)–(4.26) for N=2). The formulae recover the Nekrasov partition function for Young diagrams and are illustrated on the Jordan quiver and Q^{1,1,1}. A thermodynamic-limit analysis for toric Calabi-Yau fourfolds proposes that the limit shape of the molten crystal is governed by the Ronkin function of the Newton polynomial of the associated brane brick model.","tokens_in":32512,"tokens_out":1098,"duration_ms":18967,"significance":"If the boundary formulae hold, the work supplies a broad combinatorial generalisation of the Nekrasov instanton sum and of earlier crystal-melting results for toric Calabi-Yau three-folds, extending them to a large class of N=2 and N=4 quivers (including those dual to toric CY4s). The careful cancellation analysis that reduces the JK integrand to boundary zeta factors, together with explicit recovery of known special cases, constitutes a concrete computational advance. The thermodynamic discussion, while more provisional, offers a concrete proposal for how a projection of the Calabi-Yau geometry can emerge from discrete crystal data, potentially linking BPS counting to amoebae and Ronkin functions in higher dimensions.","major_comments":[{"comment":"Section 2.1 introduces the no-overlap condition as a working hypothesis: every admissible JK pole u* must produce exactly rank(G) distinct atoms. All subsequent identifications of poles with atoms, the melting rule, and the residual boundary factors after cancellation (Sections 3.2 and 4.2) rely on this condition. The paper never proves that the condition holds for a general quiver; it only verifies it for the examples treated later. The central claims should therefore be explicitly restricted to quivers satisfying no-overlap, or a proof (or a clear sufficient criterion) should be supplied.","section":"Section 2.1"},{"comment":"In Section 6 the true molecule weights w(M;q,ϵ) that appear in the exact formulae are replaced by ordinary brick-matching (dimer) weights. The authors note that this changes the partition function in general and leave the equivariant correction σ_equiv unspecified (eq. (6.29)). Consequently the claim that “a projection of the Calabi-Yau geometry emerges” is established only for the modified statistical model, not for the BPS indices derived in Sections 3–4. The status of the thermodynamic statements relative to the exact formulae should be clarified.","section":"Section 6"}],"minor_comments":[{"comment":"Numerous typographical errors appear throughout (e.g. “cyrstal” in §5.3, “conditinons”, “pricesely”, “occassionally”, “unqiue”, “non-overlap conditinons”). A careful proof-reading pass is needed.","section":null},{"comment":"The notation E_{ij}(x) is introduced in (3.17) and then used with slightly varying index conventions; a single consistent convention (or a short glossary) would improve readability.","section":"Section 3.3"},{"comment":"Figures 4.1 and 6.1–6.2 are referenced but lack captions that fully explain the colour coding and the dashed circles; expanding the captions would help the reader follow the cancellation arguments.","section":"Sections 4 and 6"},{"comment":"In the N=2 rewriting (4.23)–(4.26) the outer-boundary sets ∂_a M and ∂_{-a} M are defined; a short remark confirming that these sets are independent of the arbitrary choice of which chiral realises a given uncancelled factor would remove a potential ambiguity.","section":"Section 4.3"}],"recommendation":"minor_revision","confidential_remarks":"The manuscript leans heavily on the crystal construction of the authors’ previous work [BY25]; the genuine novelty lies in the boundary-weight formulae and the thermodynamic proposal. The paper is a natural fit for a hep-th journal that publishes exact results and combinatorial methods in supersymmetric gauge theory."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"The new content is the cancellation analysis that turns every one-loop factor into an explicit product over boundary atoms of a molecule. Equations (3.18)–(3.19) for N=4 and (4.21)–(4.26) for N=2 are the real payload: the grand-canonical partition function is a sum over molecules of q^N times zeta (or sine or linear) factors evaluated only on those boundaries. That is a genuine advance over residue-by-residue bookkeeping and cleanly generalizes the arm-leg formula for Young diagrams.\n\nThe derivation is careful. Sections 3.2 and 4.2 track which numerators and denominators survive, define the various boundary sets, and handle the fractional powers that appear for N=2 by rewriting them as ordinary products over outer boundaries. The three worked examples (Jordan quiver, the two-dimensional Young-diagram quiver, and Q^{1,1,1}) all check out and recover known answers. Self-citations to the earlier crystal paper supply the combinatorial language but do not smuggle in the weight formulae; the algebra starts from the standard JK integrand.\n\nThe soft spots are already flagged by the authors and are not fatal. The no-overlap hypothesis is a working assumption rather than a theorem for arbitrary quivers; if two hyperplanes force the same linear combination of fugacities the atom set collapses and the boundary bookkeeping fails. All the examples satisfy it, so the formulae are verified where claimed. The thermodynamic-limit discussion is more provisional: they replace the true BPS weights by dimer weights and note that equivariant corrections remain to be included. That section is exploratory, not load-bearing for the main claims. Restriction to cyclic chambers is likewise explicit.\n\nThis is for people who already work with crystal melting, quiver BPS algebras, or JK localization of elliptic genera. The math is solid, the citations are appropriate, and the central identities look right. I would send it to referees; the caveats are clear enough that a competent referee can handle them. Worth reading if you care about the combinatorics of these indices.","headline":"Solid combinatorial reduction of JK residues to molecule-boundary products; the main formulae look correct and recover Nekrasov as a special case.","tokens_in":33107,"tokens_out":518,"would_cite":true,"duration_ms":5729,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"Elliptic genera and Witten indices equal a sum over crystal molecules of weights fixed only by the molecules' boundaries.","keywords":["crystal melting","elliptic genus","Jeffrey-Kirwan residue","quiver gauge theory","Nekrasov partition function","brane brick model","Ronkin function","BPS index"],"falsifier":"Compute the elliptic genus of any concrete N=2 or N=4 quiver by direct Jeffrey-Kirwan residue and by the crystal-boundary formula; any mismatch for a single dimension vector falsifies the claim.","tokens_in":33169,"feed_emoji":"💎","tokens_out":609,"duration_ms":5910,"temperature":0.7,"pith_summary":"The paper claims that the elliptic genus of a two-dimensional supersymmetric gauge theory (and its one- and zero-dimensional analogues) can be rewritten exactly as a sum over finite pieces of an infinite crystal. Each piece, called a molecule, contributes a weight that depends only on the atoms sitting on its boundary. The same language that once produced the Nekrasov instanton sum over Young diagrams now works for far more general quivers, including those that come from toric Calabi-Yau fourfolds. In the thermodynamic limit the molten crystal settles into a smooth profile whose shape is the Ronkin function of the associated Newton polynomial, so a projection of the original Calabi-Yau geometry reappears from discrete combinatorics. The result therefore turns an abstract Jeffrey-Kirwan residue into a concrete crystal-melting rule and simultaneously generalises the classic combinatorics of Young diagrams.","feed_headline":"BPS indices reduce to crystal-boundary weights","feed_subtitle":"Elliptic genera equal sums over molecules whose only data sit on the edge","key_machinery":"Boundary-only cancellation of one-loop determinants inside the Jeffrey-Kirwan residue: after all internal numerators and denominators cancel, the surviving factors are completely determined by two (or more) combinatorial boundaries of the molecule, written as products of the functions E_ij(x).","core_discovery":"For a broad class of quiver gauge theories the grand-canonical partition function equals a discrete sum over all molecules M inside a crystal C, where the weight of each molecule is a product of elementary zeta (or sine or linear) factors evaluated solely on the boundary atoms of M. Explicit formulae are given for both N=4 and N=2 theories; when the quiver is the Jordan quiver the sum reduces to the ordinary Nekrasov partition function.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Quiver BPS indices equal sums over crystal-molecule boundaries","Elliptic genera reduce to boundary zeta factors on molecules","Crystal boundary atoms alone fix the BPS partition function","Nekrasov function generalizes via molecule-edge weights","Exact quiver indices from discrete sums of crystal profiles"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"Atoms that arise from different gauge nodes or different colour indices are never allowed to sit at the same point in flavour space; if two atoms ever coincide the whole identification of poles with crystal sites collapses.","fun_headline_variants_meta":{"raw":{"variants":["Quiver BPS indices equal sums over crystal-molecule boundaries","Elliptic genera reduce to boundary zeta factors on molecules","Crystal boundary atoms alone fix the BPS partition function","Nekrasov function generalizes via molecule-edge weights","Exact quiver indices from discrete sums of crystal profiles"]},"model":"grok-4.5","effort":"low","cost_usd":0.005128,"raw_usage":{"total_tokens":1384,"prompt_tokens":698,"num_sources_used":0,"completion_tokens":79,"cost_in_usd_ticks":51280000,"prompt_tokens_details":{"text_tokens":698,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":607,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":698,"tokens_out":79,"duration_ms":4734,"temperature":1.0,"reasoning_tokens":607,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-12T01:09:10.622105+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Compute the elliptic genus of any concrete N=2 or N=4 quiver by direct Jeffrey-Kirwan residue and by the crystal-boundary formula; any mismatch for a single dimension vector falsifies the claim.","supporting_citations":[],"review_version":1}