{"id":"3d53cece-4376-4ac0-ab18-3e332e0c5e77","arxiv_id":"2501.18077","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":3,"one_line_summary":"The first PBC continuum extrapolation of the ΔI=1/2 K→ππ amplitude gives Re(ε'/ε)=17.5(6.8)(4.9)(5.0)×10^{-4}, consistent with the experimental 16.6(2.3).","lead":"Lattice QCD calculations of kaon decays into two pions were performed on two lattice spacings with periodic boundary conditions, and the continuum limit was attempted for the first time. The preliminary direct CP violation measure ε'/ε agrees with experiment, but the continuum extrapolation uses only two coarse lattices and is not yet a final prediction.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The 1.4 GeV PBC point—the new endpoint of the continuum fit—carries an unquantified on-shell interpolation systematic (Eππ ≈ 1.11 mK) that is postponed to the full paper; it can bias the continuum ε'/ε value.","rationale":"The reader correctly identifies the two-point continuum extrapolation as the main weak area, and the authors themselves describe the scaling-violation estimate as uncertain. However, the more specific and more load-bearing sub-assumption is that the new 1.4 GeV PBC point—the finer endpoint of the fit—has controlled on-shell kinematics. The interpolation gap is 11% of mK on this ensemble, and the paper explicitly postpones estimating this systematic. Since A0 and ε' are extracted from matrix elements interpolated to the on-shell point, an error here biases the 1.4 GeV data point and propagates directly into the continuum limit. The reported 'tension' in Q7/Q8 with the GPBC calculation is an additional indicator that this systematic may not be negligible. My recommendation is therefore unchanged: the paper is a transparent preliminary proceedings and the result is plausibly correct, but the continuum value should be treated as conditional until the on-shell interpolation systematic on the finer ensemble is quantified. This is consistent with the reader's CONDITIONAL verdict, though the precise concern is narrower than the general two-point-fit worry.","tokens_in":9082,"tokens_out":7489,"duration_ms":82991,"concrete_test":"Estimate the on-shell interpolation systematic on the 323 (a⁻¹≈1.4 GeV) ensemble by repeating the extraction with alternative interpolation models—e.g., a quadratic fit that also includes the n=2 finite-volume state, or a parameterized energy dependence constrained by the n=0 and n=1 points—and propagate the resulting shift in the Q7/Q8 (and all other operator) matrix elements through renormalization and the Figure 4 continuum fit. If the shift in Re(ε'/ε) is comparable to or larger than the quoted total uncertainty (≈10×10⁻⁴), the central continuum claim is not yet supported; if it is negligible, the postponed systematic can be safely deferred.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central result Re(ε'/ε)=17.5(6.8)(4.9)(5.0)×10⁻⁴ depends on the continuum extrapolation of A0 in Figure 4, in which the newly added 1.4 GeV PBC ensemble is the finer data point. On this ensemble the n=1 two-pion energy is about 11% above mK, and the matrix elements are linearly interpolated in Eππ to the on-shell point (Section 3, Figure 3). The paper explicitly postpones estimating the associated systematic: 'On the 323 lattice, the n=1 two-pion energy is about 11% larger than the kaon mass and this could enhance the systematic error, but we postpone the estimation of this effect until the full paper.' This systematic is therefore absent from the errors on the 1.4 GeV data point and from the continuum limit, even though the interpolation gap is roughly twice as large as on the 1.0 GeV ensemble (6%), where the effect was argued to be small. The concern is sharpened by the observation that the electroweak penguin operators Q7 and Q8, which are important for Im A0 and hence ε', show 'some tension' with the GPBC results at this same lattice spacing. If this tension reflects on-shell interpolation or excited-state contamination, the new finer point is biased and the continuum extrapolated ε' shifts by an amount not included in the quoted errors. The first PBC continuum-extrapolated value is therefore conditional on an unquantified systematic that directly affects the new data point.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This proceedings paper reports preliminary lattice QCD results for the ΔI = 1/2 K→ππ matrix elements and the direct CP-violation parameter ε′/ε, computed with periodic boundary conditions (PBC) on two ensembles with inverse lattice spacings a⁻¹ ≈ 1.0 GeV and 1.4 GeV. The coarser ensemble updates an earlier PBC calculation with roughly doubled statistics, while the finer ensemble is new. The authors perform a GEVP analysis to isolate the first-excited two-pion state, apply Lellouch–Lüscher factors, interpolate bare matrix elements to the on-shell energy, renormalize in RI/SMOM with step scaling, and combine the renormalized amplitudes with external Wilson coefficients. They present the first continuum extrapolation of the ΔI = 1/2 amplitude and of ε′/ε from PBC data, obtaining Re(ε′/ε) = 17.5(6.8)(4.9)(5.0)×10⁻⁴, consistent with experiment and with the earlier GPBC result, and estimate an O(a²) scaling-violation systematic by comparing with an O(a⁴) fit. The paper is explicitly preliminary, with several systematic effects deferred to a later full publication.","tokens_in":9415,"tokens_out":4736,"duration_ms":48363,"significance":"If correct, this work provides the first continuum-extrapolated value of ε′/ε from lattice QCD with periodic boundary conditions, a milestone that could eventually reduce the dominant finite-spacing systematic in this quantity. The analysis follows well-established procedures from the RBC/UKQCD program: the GEVP state isolation, A2A/AMA methods, RI/SMOM renormalization with step scaling, and Lellouch–Lüscher factor treatment are all standard and internally consistent. The paper also honestly acknowledges the limitations of a two-point continuum extrapolation on coarse lattices and the preliminary nature of the result. The main risk is that the central claim — the continuum-extrapolated ε′/ε — depends on a new finer ensemble whose on-shell interpolation systematic is unquantified and explicitly postponed to the full paper. That omission directly affects the headline number and the conclusions drawn from it.","major_comments":[{"comment":"The on-shell interpolation systematic for the new finer ensemble is left unquantified. The paper states: 'On the 32³ lattice, the n = 1 two-pion energy is about 11% larger than the kaon mass and this could enhance the systematic error, but we postpone the estimation of this effect until the full paper.' This ensemble is the finer point in the continuum extrapolation (Figure 4), so the central result Re(ε′/ε) = 17.5(6.8)(4.9)(5.0)×10⁻⁴ does not include this systematic, despite the interpolation gap being roughly twice as large as on the 1.0 GeV lattice (6%). The omission is load-bearing: if the interpolation bias at 1.4 GeV is not negligible, the continuum extrapolation and thus the quoted ε′/ε would shift by an amount not covered by the quoted errors. This concern is reinforced by the 'some tension' observed for Q₇ and Q₈, which are important for Im A₀, at this same lattice spacing. The authors should either provide an estimate of this systematic (e.g., from a comparison of linear and quadratic interpolation, or from the GPBC data) or explicitly exclude it from the quoted central value and error budget, with a clear statement that the presented continuum result is conditional on this systematic being small.","section":"Section 3, Figure 3 and text on the 32³ lattice"},{"comment":"The continuum extrapolation uses only two lattice spacings with an assumed O(a²) form, and the scaling-violation systematic is estimated by comparing with an O(a⁴) fit. With only two data points, an O(a⁴) alternative is not a test of the extrapolation; it simply shifts the curve by a fixed amount. The paper acknowledges this ('somewhat uncertain since we use only two lattices with relatively large spacings'), but the quoted 11% scaling-violation error for Im A₀ should be understood as a qualitative measure rather than a robust estimate. The authors should make this limitation more prominent in the abstract and summary, and ideally present the O(a⁴) central value explicitly so that readers can see the sensitivity. This issue is separate from the interpolation systematic of the previous comment but compounds its impact on the final ε′/ε.","section":"Section 3, Figure 4 and continuum limit discussion"}],"minor_comments":[{"comment":"The phrase 'the244 lattice' appears in the discussion of the coarser ensemble; this is likely a typo for 'the 24³×64 lattice' or 'the 24³ lattice'. Please clarify.","section":"Section 2"},{"comment":"In the displayed equation for C³ᵖᵗₙ,ᵢ, the angle brackets are rendered as 'D' and 'E' (apparently a LaTeX/encoding issue). The expectation value notation should be fixed in the final version.","section":"Section 2, Eq. (6)"},{"comment":"The word 'Möbuis' in the sentence about existing ensembles is a typo for 'Möbius'.","section":"Section 4"},{"comment":"Reference [9] is a PoS proceedings contribution by C. Kelly; it would be helpful to include the full author list and arXiv identifier if available, to improve traceability of the ensemble parameters.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"This is a LATTICE2024 proceedings contribution, and the authors are appropriately cautious in calling the results preliminary. However, the manuscript presents a central numerical value for Re(ε′/ε) that is not yet supported by a fully quantified error budget, because the on-shell interpolation systematic at the finer lattice spacing — the very point that drives the continuum limit — is explicitly deferred. In a proceedings article this might be acceptable if the claim were framed as conditional, but as written the headline result and its 'good agreement' with experiment appear in the abstract and conclusion without this caveat. I would request that the authors add an explicit caveat and, ideally, a crude estimate of the interpolation systematic before publication. The two-point continuum extrapolation is also a known limitation that the paper acknowledges; for a proceedings this is acceptable if clearly stated, but the version of record should not overstate the reliability of the 11% scaling-violation error."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First continuum-extrapolated PBC result for ΔI=1/2 K→ππ and ε' — a genuine step for the field, but the new 1.4 GeV data point carries an on-shell interpolation systematic that the paper postpones, so read the central ε' as conditional.\n\nWhat's new: a new 32³ PBC ensemble at a⁻¹≈1.4 GeV, doubled statistics on the old 1.0 GeV ensemble, and a re-basing GEVP tweak that gives ~20% statistical improvement. The first PBC continuum extrapolation of Re A0, Im A0 and ε' follows. The quoted ε'/ε = 17.5(6.8)(4.9)(5.0)×10⁻⁴ sits nicely between experiment and the GPBC value. The analysis is careful and standard: RI/SMOM renormalization with step scaling, Lellouch-Lüscher factors, linear interpolation to on-shell kinematics. No circularity — Wilson coefficients and Z factors come from external sources, and the GPBC comparison is a consistency check.\n\nThe soft spots are real but the paper mostly owns them. The continuum extrapolation is a two-point fit in a² at 1.0 and 1.4 GeV; the O(a⁴) alternative is only a partial handle. The paper admits this is 'somewhat uncertain.' The sharper issue, which the stress-test note gets right, is the on-shell interpolation on the new 32³ ensemble: E_ππ for n=1 is 11% above m_K, roughly double the gap on the 1.0 GeV ensemble, and the systematic is explicitly postponed to the full paper. That is exactly the point that anchors the continuum fit, so a bias there moves ε' by an amount not in the error bars. The mild tension in Q7 and Q8 against GPBC at the same lattice spacing is a hint that the interpolation or excited-state handling is not perfectly clean. I wouldn't call it a fatal flaw — the paper is transparent and this is a proceedings — but it does mean the headline number is not yet a final lattice prediction.\n\nWho should read this: lattice practitioners and phenomenologists tracking the ε' status. It is a useful, honest progress report. I would send it to a referee — it deserves careful review as a proceedings contribution — with the expectation that the full paper must quantify the on-shell interpolation and add a third, finer lattice before the continuum result carries field weight. I'd cite it as the current PBC status, not as a final number.","headline":"First PBC continuum extrapolation of ε' is a real step, but the new 1.4 GeV point has a postponed on-shell interpolation systematic that makes the central value conditional.","tokens_in":9995,"tokens_out":2960,"would_cite":true,"duration_ms":30445,"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":"The paper reports the first periodic-boundary-condition lattice-QCD continuum extrapolation of ε'/ε, finding Re(ε'/ε)=17.5(6.8)(4.9)(5.0)×10^-4, in agreement with experiment.","keywords":["lattice QCD","direct CP violation","kaon decay","epsilon-prime","Delta I = 1/2 amplitude","continuum extrapolation","periodic boundary conditions","domain wall fermions"],"falsifier":"Compute the same $\\Delta I = 1/2$ amplitude and $\\mathrm{Re}(\\varepsilon'/\\varepsilon)$ on an ensemble with $a^{-1}\\approx 2.7$ GeV using the identical periodic-boundary-condition analysis; if the new point lies off the $O(a^2)$ line by more than the quoted scaling-violation error, the continuum extrapolation and its uncertainty would be wrong.","tokens_in":8847,"feed_emoji":"⚛️","tokens_out":11238,"duration_ms":99483,"temperature":0.7,"pith_summary":"The paper reports the first attempt to take the continuum limit of $\\varepsilon'$, the measure of direct CP violation in $K\\to\\pi\\pi$ decay, using periodic boundary conditions on the lattice. It computes the $\\Delta I = 1/2$ amplitude $A_0$ on two ensembles with inverse lattice spacings $a^{-1}\\approx 1.0$ GeV and $1.4$ GeV, the finer ensemble new and the coarser one updated with roughly doubled statistics. Extrapolating $A_0$ linearly in $a^2$, the paper obtains $\\mathrm{Re}(\\varepsilon'/\\varepsilon) = 17.5(6.8)(4.9)(5.0)\\times 10^{-4}$, in good agreement with the experimental value $16.6(2.3)\\times 10^{-4}$ and with the earlier G-parity calculation. The result is preliminary, and the paper explicitly notes that the $O(a^2)$ scaling-violation estimate is uncertain because only two relatively coarse lattice spacings are used.","feed_headline":"Direct CP violation in kaon decay now continuum-extrapolated: 17.5e-4","feed_subtitle":"Preliminary lattice result 17.5(6.8)(4.9)(5.0)×10^-4 agrees with experiment 16.6(2.3) and earlier G-parity values.","key_machinery":"The load-bearing objects are the $\\Delta I = 1/2$ amplitude $A_0$ and the ratio $\\varepsilon'/\\varepsilon$, built from renormalized matrix elements $M_i^{\\overline{\\mathrm{MS}}}(\\mu)$ and three-flavor Wilson coefficients $z_i(\\mu)$ and $y_i(\\mu)$. The two-pion final state is isolated with a variational GEVP over momentum and $\\sigma$ operators, the Lellouch-Lüscher factor converts finite-volume states to infinite-volume matrix elements, linear interpolation in $E_{\\pi\\pi}/m_K$ reaches the on-shell point, and RI/SMOM step scaling to $\\mu = 4.0$ GeV supplies the nonperturbative renormalization. The continuum limit is taken by a two-point linear fit of $A_0$ in $a^2$, with the scaling-violation error estimated from the shift to an $O(a^4)$ fit.","core_discovery":"On its own terms, this paper establishes that the $\\Delta I = 1/2$ $K\\to\\pi\\pi$ amplitude and $\\varepsilon'$ can be computed with periodic boundary conditions at two lattice spacings and extrapolated to the continuum. The central number is $\\mathrm{Re}(\\varepsilon'/\\varepsilon) = 17.5(6.8)(4.9)(5.0)\\times 10^{-4}$, which agrees with both the experimental value $16.6(2.3)\\times 10^{-4}$ and the earlier G-parity lattice result $21.7(2.6)(6.2)(5.0)\\times 10^{-4}$. The paper also reports that the bare matrix elements of the electroweak penguin operators $Q_7$ and $Q_8$ show some tension between the periodic and G-parity calculations at the shared inverse spacing of about 1.4 GeV, while the other operators are consistent. All of these results are labeled preliminary.","pith_inferences":["If the linear $a^2$ extrapolation holds on finer lattices, the periodic-boundary-condition route could reach the experimental precision for $\\varepsilon'$ using already-generated domain-wall ensembles, avoiding the overhead of G-parity boundary conditions.","The reported tension in the $Q_7$ and $Q_8$ bare matrix elements between the two boundary-condition treatments suggests that finite-volume or boundary-condition effects on the left-right penguin operators deserve a dedicated study before the final error budget is closed.","A direct test of the claimed order-independence of interpolation and renormalization would be to repeat the earlier procedure, renormalizing before the energy interpolation, on the new 1.4 GeV ensemble and compare the on-shell matrix elements.","Once a third spacing is included, the current $O(a^2)$ scaling-violation estimate, based on only two relatively coarse points, is likely to be revised; the fine ensembles already generated provide the data for that test."],"forward_implications":["The value Re(ε'/ε)=17.5(6.8)(4.9)(5.0)×10^-4 is compatible with the experimental world average, so the Standard Model remains consistent with direct CP violation in kaon decays at current precision.","The consistency of most bare matrix elements between the periodic and G-parity calculations at the shared 1.4 GeV spacing supports periodic boundary conditions as a viable route for this process.","The continuum extrapolation, though based on two spacings, gives the first estimate of ε' in which a discretization error is assigned rather than quoted as a single-spacing uncertainty.","The paper's stated outlook is that adding finer periodic-boundary-condition ensembles with inverse spacings up to 2.7 GeV should reduce the finite-lattice-spacing error, and that the dominant remaining within-isospin error is the ~12% perturbative truncation of the Wilson coefficients."],"supporting_citations":[{"why":"Supplies the original periodic-boundary-condition calculation on the 1.0 GeV ensemble and the methodology this work extends and updates with doubled statistics.","marker":"[1]"},{"why":"Gives the previous G-parity calculation at 1.4 GeV whose results are compared with the new periodic-boundary-condition matrix elements and whose 21.7×10^-4 value is the benchmark for ε'.","marker":"[6]"},{"why":"Provides the Im A2 value and the ΔI=3/2 amplitude machinery used in the ε'/ε formula.","marker":"[4]"},{"why":"The first ab initio lattice calculation of the ΔI=1/2 process, establishing the physical-kinematics framework this work inherits.","marker":"[5]"},{"why":"Supplies the Lellouch-Lüscher factor that converts finite-volume two-pion matrix elements to infinite-volume ones.","marker":"[19]"},{"why":"Provides the step-scaling procedure used for the nonperturbative RI/SMOM scale evolution to 4 GeV.","marker":"[20]"},{"why":"Supplies the three-flavor Wilson coefficients z_i and y_i used in constructing A0.","marker":"[21]"},{"why":"Companion reference for the three-flavor effective Hamiltonian and Wilson coefficients in the same scheme.","marker":"[22]"},{"why":"Describes the re-basing prescription and the AMA-with-all-to-all implementation that improves the statistical precision of the GEVP matrix elements.","marker":"[14]"}],"fun_headline_variants":["First attempt at continuum limit for kaon CP violation","Direct CP violation in kaon decay matches experiment","Continuum-extrapolated kaon epsilon-prime agrees with data","Tension in penguin operators for lattice kaon CP violation","Two-lattice continuum limit of epsilon-prime matches experiment"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The continuum numbers rest on the assumption that discretization errors behave as the square of the lattice spacing across the two spacings used here, so adding a third, finer lattice could move the central value.","fun_headline_variants_meta":{"raw":{"variants":["First attempt at continuum limit for kaon CP violation","Direct CP violation in kaon decay matches experiment","Continuum-extrapolated kaon epsilon-prime agrees with data","Tension in penguin operators for lattice kaon CP violation","Two-lattice continuum limit of epsilon-prime matches experiment"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000744,"raw_usage":{"total_tokens":3302,"prompt_tokens":914,"completion_tokens":2388,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":530,"completion_tokens_details":{"reasoning_tokens":2306}},"tokens_in":530,"tokens_out":2388,"duration_ms":18018,"temperature":1.0,"reasoning_tokens":2306,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T00:44:44.461633+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the same $\\Delta I = 1/2$ amplitude and $\\mathrm{Re}(\\varepsilon'/\\varepsilon)$ on an ensemble with $a^{-1}\\approx 2.7$ GeV using the identical periodic-boundary-condition analysis; if the new point lies off the $O(a^2)$ line by more than the quoted scaling-violation error, the continuum extrapolation and its uncertainty would be wrong.","supporting_citations":[{"cited_title":"Step Scaling with off-shell renormalisation","cited_arxiv_id":"1006.0422","evidence_quote":"Provides the step-scaling procedure used for the nonperturbative RI/SMOM scale evolution to 4 GeV."}],"review_version":1}