{"id":"87c834e0-3b7a-4eaf-b33f-f07bc2660fd8","arxiv_id":"2608.10840","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Monodromy defects in ABJM theory are integrable boundary states, and their leading one-point functions are given by a closed overlap formula.","lead":"Researchers show that special line-shaped defects in the three-dimensional quantum field theory ABJM can be described by a single elegant boundary formula, making many calculations about the defects easy. The formula links these defects to known solvable spin-chain problems and reproduces several earlier results as special cases.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (3.43) is singular for any Bethe state with an unpaired zero root at node 2 (Q2(0)=0); the lone L=3 check evades this by an ad hoc zero-root omission, so the claimed closed form is not established for that sector.","rationale":"The strongest claim is eq. (3.43)/(3.45). I looked for the least secure condition needed for that claim to hold for every Bethe eigenstate. The KT derivation itself is credible: the K-matrix (3.17) solves the uncrossed KT relation, and for states without a zero root at node 2 the b1-independence argument and the G/F construction have independent support from [22,23,30] and from the L=2 field-theory match. The fragile point is the zero-root sector, which the paper itself flags in Appendix A, so this is not an artifact of my reading. Because Q2(0) sits in the denominator, any state with an unpaired zero root at the middle node makes the literal formula divergent; the only state checked is handled by deleting the zero root and fixing gamma by comparison with a direct computation. One fitted example is not enough to support the claim that the formula encompasses all such one-point functions, and the text explicitly says the method of [22] gives no recipe for these roots. This is a correctness/scope risk, not a disagreement with consensus, and it is testable by exact diagonalization for a longer chain. Since the issue is a gap that a precise prescription could close, the appropriate verdict remains the reader's CONDITIONAL rather than a full ACCEPT or a REJECT.","tokens_in":23415,"tokens_out":5227,"duration_ms":870023,"concrete_test":"Construct a second Bethe eigenstate with an unpaired zero root at node 2 and L >= 4 (e.g. solve the Bethe equations (3.8) with r1=r2=r3 and an odd r2 containing a zero root), compute its overlap with the MPS (3.12) by exact diagonalization of the alternating SU(4) spin chain, and compare with (3.43) evaluated with the zero root omitted from Q2. If the ratio is not 1, the zero-root omission is not a general rule and the claimed closed formula fails on that sector.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central overlap formula (3.43) contains Q2(0) in the denominator. The selection rules (3.18) and (3.42) do not forbid an unpaired zero root at the middle node, since u=0 is self-paired under u -> -u and r2 can be odd. For any such state, Q2(0)=0 and the literal formula diverges. The only check in this sector is the L=3 operator in Appendix A, whose roots (A.9) include 0 at node 2. The authors do not evaluate (3.43) literally; they state that the unpaired zero root 'requires special treatment', introduce a free parameter gamma, fix it 'proportional to b1' with the proportionality constant set to one, and omit the zero root from the Baxter polynomial. This prescription is not derived from the KT relation and is calibrated to reproduce the field-theory result. Since the paper claims a closed-form expression encompassing all one-point functions, the validity of the formula on the Q2(0)=0 sector is load-bearing; it currently rests on one fitted example rather than on the KT derivation.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript studies co-dimension-two supersymmetric monodromy defects (``vortex loops'') in ABJM theory. Its central result is a claimed closed-form expression for the leading-order one-point functions of non-protected scalar operators, obtained by representing the defect as a matrix product state, solving the uncrossed KT relation for the reflection matrix (3.17), and applying Gombor's nested-K-matrix formalism with a deformation (3.31)-(3.32) that regularizes a vanishing intermediate entry. The resulting overlap (3.43), combined with the prefactor (3.11), involves only the Baxter polynomials Q1(i/2), Q2(0), Q2(i/2) and the Gaudin superdeterminant, subject to the selection rules r1=r3, r1=r2, and L=r1 for beta4=0. The paper shows that this formula reduces to the previously known giant-graviton three-point-function overlap (3.49) and to certain Wilson-loop one-point-function overlaps (3.51), and Appendix A checks the formula against direct field theory for an L=2 and an L=3 operator. Section 4 initiates quantization around the defect in the unit-matrix-VEV case, computing a renormalized tadpole (B.12) and the first quantum correction (4.30); Section 5 speculatively extends the formula to the all-loop osp(4|6) Dynkin diagram via fermionic dualities.","tokens_in":23687,"tokens_out":38110,"duration_ms":371058,"significance":"If fully established, the main result is a genuine advance: it unifies the one-point functions of non-protected scalar operators for the three classes of monodromy defects (1/2-, 1/3-, 1/6-BPS) into the single closed formula (3.43) with the block generalization (3.45), and it recovers the previously known giant-graviton overlap [26] and Wilson-loop overlaps [27] as special cases. The derivation has real substance: it proceeds from the KT relation with an explicit regularized K-matrix (3.31)-(3.32) and derives the selection rules r1=r3, r1=r2, L=r1 rather than assuming them, and it is not circular, since nothing in the final formula was defined to reproduce [26] or [27]. The Appendix A checks are genuine field-theory computations for two short operators, and the formula (3.43) is falsifiable, for instance by Bethe-ansatz numerics for longer states. Section 4's first quantum correction, though restricted to the unit-matrix-VEV case, is a concrete and reproducible computation with a transparent scheme, and Section 5 is appropriately hedged as a speculation.","major_comments":[{"comment":"The claimed closed form is not established for Bethe states with an unpaired zero root at the middle node, and the resulting gap is larger than a corner case. The denominator of (3.43) contains Q2(0), which vanishes whenever one of the u^{(2)} roots is zero; such states satisfy the selection rules, since u=0 is self-paired under u -> -u and (3.18)/(3.42) do not forbid odd r2. Indeed the derivation itself presupposes even r2, since (3.40) contains a product over r2/2. For the beta4=0 cases the rules (3.42) and (3.44) imply r1=r2=L, so every odd-L operator belongs to the Q2(0)=0 sector. Appendix A treats the L=3 state (A.9) by declaring that the unpaired zero root 'requires special treatment': a free parameter gamma is introduced, set proportional to b1 with unit proportionality constant, the zero root is omitted from the Baxter polynomial, and an unexplained overall sign is then dropped in arriving at (A.12). This prescription is calibrated to the single field-theory result (A.8), not derived from the KT relation. The abstract's claim of a closed-form expression for all one-point functions is therefore supported only on the even-r2, Q2(0) != 0 sector; the odd-r2 sector rests on one fitted example. The authors should either extend the KT derivation to cover the zero-root sector or explicitly restrict the claim and present the zero-root rule as an empirically supported conjecture.","section":"Section 3, Eq. (3.43); Appendix A, Eqs. (A.9)-(A.12)"}],"minor_comments":[{"comment":"'Subsquently' (p. 2) should be 'Subsequently'; similar typos appear later, notably 'auxillary' after Eq. (3.17), 'mondromy' and 'the the interplay' in Section 6, and 'read of from' in Appendix A.","section":"Section 1 and throughout"},{"comment":"The overlap formula is split across two displayed equations with a dangling multiplication sign; a single display would prevent misreading of the power structure.","section":"Eqs. (3.40)-(3.41)"},{"comment":"The comparison with [27] is worded confusingly: 'which up to a phase which agrees with the result of [27]' should be rewritten as a single clear statement of the phase agreement.","section":"Section 3, Eq. (3.51)"},{"comment":"The overall sign (-1) removed before (A.12) is stated without explanation; please clarify whether this is a phase convention of the Bethe-state normalization or part of the zero-root prescription, since the L=2 check did not require it.","section":"Appendix A, Eqs. (A.11)-(A.12)"},{"comment":"The half-integer monodromy assigned to the quantum fluctuation of Y^1 is an input to the tadpole computation (B.12); the single-valuedness argument for the Lagrangian is plausible, but the text should state explicitly whether this monodromy is a definition of the quantized defect theory or a derived consistency condition, and the subtraction in (B.4) should be identified as a renormalization-scheme choice.","section":"Section 4, Eq. (4.23) and Appendix B"},{"comment":"The combinatorial origin of the factor 2L in the tadpole correction is not spelled out; a sentence on the counting of contraction channels would make the result checkable.","section":"Section 4, Eq. (4.30)"},{"comment":"The schematic Dynkin-diagram figure must be redrawn with clear node numbering and unambiguous numerator/denominator markings; the current ASCII version cannot be parsed reliably.","section":"Section 5, Figure 2"},{"comment":"Reference [8] cites an unpublished talk; please replace it with a published source if one exists.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The paper is within JHEP's scope and the physics is interesting; the KT-relation derivation, the special-case reductions, and the short-operator checks are solid. My main concern is the zero-root sector described in the report: the headline formula (3.43) is literally singular there, and the appendix's resolution involves a fitted parameter and a hand-removed sign. This is fixable within the manuscript's scope either by a derivation of the zero-root prescription or by an honest restatement of the domain of validity of (3.43). I would not reject over this, since the even-r2 sector and the relations to [26] and [27] are sound. For the novelty assessment: the K-matrix (3.17) appeared independently in [30] and the framework is from [22,23]; the added value is the concrete ABJM application, the selection rules, and the quantization section, which I judge sufficient for publication after revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe paper does what it claims, up to one real gap. It constructs integrable matrix product states for the supersymmetric monodromy defects in ABJM, solves the KT-relation, and derives a closed-form overlap, eq. (3.43), for leading one-point functions of non-protected operators. That formula unifies previous results for giant graviton overlaps and certain Wilson loops, and the recovery of those special cases is a genuine check. The K-matrix had already appeared in [30] and the general overlap machinery is Gombor's [22,23], but the identification of monodromy defects with this boundary state, the F-functions, and the selection rules are new. The derivation is coherent and the two short-operator checks in Appendix A match direct field theory.\n\nThe soft spot is in the L=3 check. That state has an unpaired zero root at the middle node, so Q2(0)=0 and the literal formula (3.43) is singular. The authors say the zero root requires special treatment, introduce a free parameter gamma, fix it proportional to b1 with constant set to one, and omit the root from the Baxter polynomial. This prescription is not derived from the KT relation; it is fitted to the field-theory result. So the claim of a closed form encompassing all one-point functions is not established for the Q2(0)=0 sector. For states without unpaired zero roots the derivation is convincing, and the singularity is arguably measure-zero in the space of Bethe states, but it is a real gap in the advertised result, not a cosmetic detail.\n\nThe quantum correction section is preliminary: the tadpole calculation for the simplest unit-matrix background is fine, and the resulting -2Lλ/(π|β|2) correction is plausible, but the general case is left as a bookkeeping program. The fermionic-duality extension in Section 5 is more speculative, though it makes a sharp prediction worth testing.\n\nThe citation pattern is honest: the authors credit [30] for the K-matrix and [22,23] for the machinery, and there is no sign of fitting the central claim to data.\n\nI would send this to a serious referee. The referee should press on the zero-root sector: either derive the prescription from the KT relation or state that the closed formula applies only to states with even r2 and no zero roots. For everyone working on defect one-point functions in ABJM, this is a useful and mostly careful calculation, and I would cite it.","headline":"A solid integrability calculation extending the overlap machinery to ABJM monodromy defects, but the claimed closed form has a genuine gap in the unpaired-zero-root sector.","tokens_in":24195,"tokens_out":2943,"would_cite":true,"duration_ms":28842,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"Supersymmetric monodromy defects in ABJM theory, known as vortex loops, are integrable boundary states of the alternating SU(4) spin chain, and the leading one-point functions of non-protected scalar operators are given by a single closed…","keywords":["ABJM theory","monodromy defects","vortex loops","one-point functions","integrable boundary states","KT relation","alternating SU(4) spin chain","Baxter polynomials"],"falsifier":"Compute the leading-order one-point function of a length-$L=4$ non-protected scalar operator in the vortex-loop background directly in ABJM field theory and compare it with the closed formula; any discrepancy would show that the regularization or zero-root prescription misses certain sectors. A second, cheaper check: find a Bethe state that violates $r_1=r_2$ but has a non-vanishing direct overlap with the matrix product state, which would falsify the selection rule on which the formula depends.","tokens_in":23171,"feed_emoji":"🌀","tokens_out":16866,"duration_ms":142229,"temperature":0.7,"pith_summary":"The paper establishes that the supersymmetric monodromy defects of ABJM theory—line defects around which scalar fields wind, also called vortex loops—are integrable boundary states of the planar alternating SU(4) spin chain. Solving the KT relation of the algebraic Bethe ansatz for the matrix product state representing the defect, the authors obtain one closed formula for the leading-order one-point function of every non-protected single-trace scalar operator in these backgrounds. The normalized overlap is a product of Baxter polynomials evaluated at half-integer arguments, powers of the defect parameters, and the Gaudin superdeterminant, together with selection rules that decide which operators couple to the defect. This matters because one-point functions of non-protected operators in defect theories are normally accessible only state by state; here the whole class collapses to a single expression, with the giant-graviton three-point overlap and a class of Wilson-loop one-point functions emerging as special cases. The paper also initiates quantization in the defect background and computes the first quantum correction in the simplest setting.","feed_headline":"Vortex-loop defects in ABJM are integrable boundary states","feed_subtitle":"One closed formula covers every non-protected scalar one-point function at these defects.","key_machinery":"The engine is the KT relation, the consistency equation between a boundary reflection matrix and the spin-chain monodromy matrix that makes a matrix product state an integrable boundary state. It is applied to the two-site building block (one fundamental and one anti-fundamental site) of the alternating SU(4) chain. The defect data enter through the $K$-matrix solution $K_{ij}(u)=\\delta_{ij}+\\frac{2u}{1-u}\\frac{\\beta_i^*\\beta_j}{|\\beta|^2}$, from which nested $G$- and $F$-functions are constructed; their products reorganize into the Baxter polynomials $Q_1(i/2)$, $Q_2(0)$, and $Q_2(i/2)$ in the final overlap. Because a nested component of the $K$-matrix vanishes, a deformation with a free parameter $b_1$ is needed, and demanding independence from this regularization enforces the extra selection rule $r_1=r_2$. The square root of the Gaudin superdeterminant, $\\sqrt{\\mathrm{Sdet}G}$, is the universal normalization factor independent of the defect. The same machinery supports the fermionic-duality extension to the all-loop $\\mathfrak{osp}(4|6)$ Dynkin diagram.","core_discovery":"The central discovery is that a supersymmetric monodromy defect in ABJM theory corresponds to an integrable matrix product state of the alternating SU(4) spin chain, and the leading semi-classical one-point function of any non-protected scalar operator follows from one closed expression. For a Bethe eigenstate $|u\\rangle$ with root numbers $r_1,r_2,r_3$ satisfying the selection rules $r_1=r_2=r_3$, and additionally $L=r_1$ when $\\beta_4=0$, the normalized overlap reads $$\\frac{\\langle \\mathrm{MPS}|u\\rangle}{\\sqrt{\\langle u|u\\rangle}} = (\\beta_1\\beta_4^*)^{L-r_1}\\,|\\$\\beta$|^{2r_1}\\,$2^{{-r_1}}$\\,(-i)^{r_1}\\,\\frac{Q_1(i/2)}{\\sqrt{Q_2(0)\\,Q_2(i/2)}}\\,\\sqrt{\\mathrm{Sdet}\\,G},$$ with $Q_a(u)=\\prod_j(u-u^{(a)}_j)$ the Baxter polynomial of node $a$. The one-point function is this overlap multiplied by the prefactor $1/(|z|^L\\lambda^L L^{1/2})$. The same formula reproduces the overlap for two maximal giant gravitons and a tiny graviton and, up to a phase, the scalar one-point functions for a class of supersymmetric Wilson loops. In the simplest quantum background the first correction multiplies every one-point function by the universal factor $1-2L\\lambda/(\\pi|\\beta|^2)+\\cdots$.","pith_inferences":["A sharp testable consequence: for the 1/2- and 1/3-BPS defects ($\\beta_4=0$), the selection rule $L=r_1$ restricts non-vanishing one-point functions to a specific class of Bethe states, and a direct field-theory computation of a length-four operator would probe this restriction.","The same $K$-matrix has also appeared in a systematic search for integrable boundary states of the ABJM spin chain, so the vortex-loop overlaps are likely members of a larger family of boundary states, some of which may correspond to as-yet-unidentified defects or Wilson-loop-like observables.","Because the first quantum correction comes only from a tadpole, one may conjecture that the one-point functions exponentiate to a closed all-order factor in the simple background; a two-loop computation would either confirm the pattern or reveal where it breaks."],"forward_implications":["All leading-order one-point functions of non-protected scalar operators in the 1/2-, 1/3-, and 1/6-BPS monodromy backgrounds are governed by one formula, so overlaps no longer need to be computed operator by operator.","The known giant-graviton three-point overlap and the scalar one-point functions for a class of supersymmetric Wilson loops are recovered as special cases of the same expression.","The fermionic-duality argument predicts a factorized overlap on the all-loop $\\mathfrak{osp}(4|6)$ Dynkin diagram, which is a first indication of higher-loop integrability for these one-point functions.","In the simplest quantum background the first correction is the universal factor $1-2L\\lambda/(\\pi|\\beta|^2)+\\cdots$, independent of the detailed structure of the operator at this order.","The spectrum of conformal dimensions of gauge-invariant bulk operators is unchanged by the defect, and gauge-invariant correlation functions remain single-valued despite the monodromy of the underlying fields."],"supporting_citations":[{"why":"Defines the vortex-loop (monodromy) defects in ABJM theory and gives the classical BPS solutions that the matrix product state is built from.","marker":"[17]"},{"why":"Introduces the alternating SU(4) spin chain and its Bethe ansatz, the language in which operators become Bethe eigenstates.","marker":"[19]"},{"why":"Supplies the general KT-relation overlap formula for integrable matrix product states that the paper solves in this setting.","marker":"[22]"},{"why":"Provides the recursive derivation and the deformed-K-matrix regularization used to extract the overlap and the r1=r2 selection rule.","marker":"[23]"},{"why":"Gives the giant-graviton three-point overlap that the paper's formula reproduces as a special case.","marker":"[26]"},{"why":"Gives the Wilson-loop one-point function overlaps that are recovered, up to phase, from the formula.","marker":"[27]"},{"why":"Identifies non-protected single-trace operators with Bethe eigenstates of the two-loop dilatation operator, the correspondence underlying the computation.","marker":"[31]"},{"why":"Provides the fermionic-duality transformation rule for Gaudin superdeterminants used to extend the overlap to the all-loop Dynkin diagram.","marker":"[36]"},{"why":"Supplies the short operators and Bethe roots used in Appendix A to verify the overlap formula for L=2 and L=3.","marker":"[24]"}],"fun_headline_variants":["ABJM vortex loops are integrable boundary states","Closed formula for ABJM defect one-point functions","Monodromy defects yield integrable spin-chain states","Defect one-point functions solved in ABJM theory","Integrable boundary states from ABJM defects"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the KT-relation recipe, including the regularized $K$-matrix and the treatment of unpaired zero roots, correctly computes the overlap for every Bethe state; the paper verifies the formula explicitly only for operators of length $L=2$ and $L=3$, and in the $L=3$ case with a zero root a free parameter is fixed by matching a field-theory result rather than derived.","fun_headline_variants_meta":{"raw":{"variants":["ABJM vortex loops are integrable boundary states","Closed formula for ABJM defect one-point functions","Monodromy defects yield integrable spin-chain states","Defect one-point functions solved in ABJM theory","Integrable boundary states from ABJM defects"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000138,"raw_usage":{"total_tokens":1175,"prompt_tokens":989,"completion_tokens":186,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":605,"completion_tokens_details":{"reasoning_tokens":113}},"tokens_in":605,"tokens_out":186,"duration_ms":2883,"temperature":1.0,"reasoning_tokens":113,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T16:02:10.046973+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the leading-order one-point function of a length-$L=4$ non-protected scalar operator in the vortex-loop background directly in ABJM field theory and compare it with the closed formula; any discrepancy would show that the regularization or zero-root prescription misses certain sectors. A second, cheaper check: find a Bethe state that violates $r_1=r_2$ but has a non-vanishing direct overlap with the matrix product state, which would falsify the selection rule on which the formula depends.","supporting_citations":[{"cited_title":"Vortex Loop Operators, M2-branes and Holography","cited_arxiv_id":"0810.4344","evidence_quote":"Defines the vortex-loop (monodromy) defects in ABJM theory and gives the classical BPS solutions that the matrix product state is built from."},{"cited_title":"Overlaps and Fermionic Dualities for Integrable Super Spin Chains","cited_arxiv_id":"2011.12192","evidence_quote":"Provides the fermionic-duality transformation rule for Gaudin superdeterminants used to extend the overlap to the all-loop Dynkin diagram."}],"review_version":1}