{"id":"033dda66-cd85-4766-bc16-d02b7fe565df","arxiv_id":"hep-lat/0506013","paper_version":3,"verdict":"CONDITIONAL","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The I=2 pion-pion scattering length is computed on 2+1-flavor dynamical lattices at pion masses below 500 MeV and extrapolated to the physical point, giving m_pi a_2 = -0.0426 with a chiral-truncation error of 0.0018.","lead":"Lattice QCD with 2+1 dynamical quark flavors yields the I=2 pion-pion scattering length at three pion masses (294, 348, 484 MeV), extrapolated to m_pi a_2 = -0.0426(6)(3)(18), in agreement with experiment. It is the first fully-dynamical calculation in the chiral regime and provides a lattice determination of a next-to-leading-order chiral constant.","discovery_kind":"new_application","skeptic_critique":{"model":"selfhosted/deepseek-v4.1-flash","headline":"The extrapolated mπa2 rests on a long chiral fit whose truncation error is likely underestimated: the paper's own footnote-8 alternative gives -0.0462, twice the quoted ±0.0018 systematic away from the headline -0.0426.","rationale":"The paper's direct lattice measurements of ΔE0 and a2 at mπ = 294, 348, 484 MeV are the solid core, and the finite-volume method (Lüscher, eqs. 1-4) is applied correctly; the <1% difference between eq. (1) and eq. (4) for a2 is stated and credible. The scale setting via fπ is cross-checked against MILC's r0-based scale and is consistent. The paper is candid about its limitations: single lattice spacing, mixed-action partial quenching not treated with the mixed-action χPT of Ref. [43], heaviest mass at the edge of the chiral regime, and the unresolved chiral curvature. The reader's weakest_assumption correctly identifies the chiral extrapolation as the load-bearing soft spot. My sharpening is that the paper's own footnote-8 alternative fit already demonstrates a spread (0.0036) twice the quoted truncation error (0.0018), so the error budget is internally inconsistent. This does not undermine the central value — which is close to the leading-order χPT prediction and agrees with experiment — but it does mean the quoted precision is overstated. CONDITIONAL remains the right verdict: the result is a genuine advance, but the extrapolated precision should be enlarged and the mixed-action χPT treatment (or a demonstration that its neglect is subdominant) would strengthen the claim. I do not see a soundness or circularity failure; the concern is a quantitative error-budget issue, not a flaw in the method.","tokens_in":35206,"tokens_out":39014,"duration_ms":722382,"concrete_test":"Recompute the truncation systematic as the spread over the paper's own extrapolations — one-loop eq. (15) (-0.0426), LO×A footnote 8 (-0.0462), and the simplified two-loop fits — rather than the single two-loop comparison used. If the spread exceeds ±0.0018 (it appears to be ≥0.0036), the headline error bar should be enlarged to at least ±0.0036. A stronger check: add a fourth, lighter pion mass (mπ ~ 250 MeV) and verify that the one-loop fit still describes the data and that the extrapolated value is stable against the fit form.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is the extrapolated value mπa2 = -0.0426 ± 0.0018 (truncation), obtained by fitting the one-loop χPT formula (eq. 15) to three points at mπ/fπ = 2.03, 2.34, 3.07 and running to mπ/fπ ≈ 1.04. The extrapolation is long (factor 2-3 in mπ) and the data show essentially no curvature: the 'curvature' function C of eq. (18) is 0.29, 0.02, 0.02, consistent with zero at the two lighter masses. The paper acknowledges (§IV.B) that 'there is an almost perfect cancellation between the counterterm and the logarithm in eq. (15),' so the fit form, not the data, largely fixes the extrapolated value. The quoted ±0.0018 truncation error is estimated by comparing the one-loop fit to simplified two-loop fits (l(2)=l(3)=0, then l(3)=0). But the paper's own footnote 8 presents an alternative LO×A fit giving mπa2 = -0.0462 ± 0.0026 ± 0.0014 ± 0.0018, which differs from the headline by 0.0036 — twice the quoted truncation error. Thus the paper's own family of reasonable extrapolations spans at least ±0.0036, not ±0.0018. This is an internal inconsistency in the error budget, not a disagreement with consensus. The central value still agrees with the K(e4) result (-0.0454 ± 0.0031), so the claim survives, but its precision is overstated.","agreement_with_reader":"agree"},"referee_report":{"model":"selfhosted/deepseek-v4.1-flash","summary":"The manuscript reports a fully-dynamical lattice QCD calculation of the I=2 ππ scattering length using Lüscher's finite-volume method. Domain-wall valence propagators generated by LHPC are contracted on the publicly available asqtad-improved MILC 2+1-flavor staggered configurations at three light-quark masses corresponding to mπ ≈ 294, 348 and 484 MeV, at a single lattice spacing b ≈ 0.127 fm and L ≈ 2.5 fm. The ground-state energy shift ΔE0 relative to two non-interacting pions is extracted from the ratio Gππ = Cππ/Cπ² of Eq. (10) and converted to a2 through Eq. (4); the first excited level at p = 2π/L yields a phase shift at the heaviest mass. The three values of mπa2 are fitted with the one-loop χPT expression of Eq. (15) with a single low-energy constant lππ, giving lππ = 3.3 ± 0.6 ± 0.3 and an extrapolated mπa2 = −0.0426 ± 0.0006 ± 0.0003 ± 0.0018 at the physical point, together with δ = −43 ± 10 ± 5° at |p| ≈ 544 MeV for mπ ≈ 484 MeV. A two-loop expression and the associated low-energy constants are collected in Appendix A but not fitted.","tokens_in":17959,"tokens_out":7377,"duration_ms":135296,"significance":"If the result holds, this is the first fully-dynamical 2+1-flavor lattice determination of the I=2 ππ scattering length in the chiral regime, and it supplies a lattice determination of the next-to-leading-order constant lππ that is otherwise only estimated. The mixed-action strategy (domain-wall valence on staggered sea) is itself of methodological interest for the community. The paper is commendably transparent about its assumptions: the single lattice spacing and the absence of a continuum extrapolation are stated explicitly in §IV, the neglect of exponentially suppressed finite-volume corrections is justified through mπL ≫ 1, and the error budget is quoted as three separate contributions (statistical, fitting, chiral truncation). The cross-checks against the CP-PACS dynamical calculation [22] and against the K(e4) determination [57] are appropriate, and the comparison of the one-loop fit to simplified two-loop fits is a genuine, if heuristic, attempt to quantify truncation. The agreement of the extrapolated value with experiment is a nontrivial consistency check of the whole program.","major_comments":[{"comment":"The third error in the headline result, ±0.0018, is described as an estimate of the chiral-truncation systematic, obtained by comparing the one-loop fit of Eq. (15) to two simplified two-loop fits (l(2)=l(3)=0, then l(3)=0). Footnote 8, however, reports a leading-order fit with an arbitrary coefficient A, which gives mπa2 = −0.0462 ± 0.0026 ± 0.0014 ± 0.0018, a shift of 0.0036 from the headline value — twice the quoted truncation error. As written, the reader cannot tell why the LO×A variation is excluded from the truncation estimate while the two-loop variations are included. Either the paper should justify why the LO×A ansatz is not a member of the family of reasonable extrapolations, or the truncation error should be widened to encompass the observed spread. This is load-bearing for the precision claim in the abstract, even though the central value remains consistent with the K(e4) result.","section":"§IV.B and footnote 8","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The calculation is performed at a single lattice spacing with a mixed action, and the paper states that m_res ≠ 0 implies O(b) corrections that are 'exponentially suppressed with the extent of the fifth dimension and are therefore safely ignored given our current level of precision.' No quantitative bound accompanies this assertion, and the valence-sea matching is performed only to a few-per-cent precision in the pion mass (Table I). Given that the target precision on mπa2 is roughly 4%, an unquantified mixed-action systematic of this order would be comparable to the quoted truncation error. A quantitative estimate — for instance from the m_res values and L5 = 16 in Table I, or from a comparison of the domain-wall and staggered pion masses at the same parameters — would substantially strengthen the error budget. The paper acknowledges the absence of a continuum extrapolation, but the specific size of the residual mixed-action effect is not bounded.","section":"§III.A and §IV","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The extrapolation runs from mπ/fπ = 3.07 down to ≈ 1.04, and the data show essentially no curvature: the function C of Eq. (18) is 0.29 ± 0.16, 0.021 ± 0.067 and 0.017 ± 0.082, consistent with zero at the two lighter masses. The paper itself notes 'an almost perfect cancellation between the counterterm and the logarithm in eq. (15) in the range of mπ considered' and that 'a more refined lattice QCD calculation is required in order to detect the predicted chiral curvature.' With a single fitted parameter and no observed curvature, the extrapolated value is largely fixed by the fit ansatz rather than by the data. This is honestly acknowledged, but it should be reflected both in the error budget (see the preceding comment) and in the language: the word 'prediction' in §IV.B and §V overstates what is, in effect, an ansatz-dependent extrapolation of a one-parameter fit. A concrete remedy would be to quote the spread of the extrapolated value over an explicit family of ansätze (one-loop, the two two-loop variants, and LO×A) as the truncation error, and to describe the result as an extrapolated determination rather than a prediction.","section":"§IV.B, Eq. (15), and Fig. 2","authors_acknowledge":null,"authors_acknowledge_p":null}],"minor_comments":[{"comment":"The abstract quotes b ≈ 0.125 fm while §V quotes b ≈ 0.127 fm, and Table II gives 12.72–12.78 × 10⁻² fm from the fπ scale setting. These should be harmonized, and the abstract should reflect the value the paper actually determines.","section":"Abstract and §V","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The right panel's vertical axis is labelled 'Log(|Gππ(1,t)|)' while the caption refers to Gππ(2π/L, t). The label should match the caption, and the meaning of the '(G.S. subtracted)' curve should be stated in the caption rather than only in the text.","section":"Fig. 1","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"No χ²/d.o.f. is reported for the chiral fits to Eq. (15) or Eq. (18), although one is reported for the ΔE0 extraction. With three points and one parameter there are two degrees of freedom, and the fit quality is directly relevant to the truncation discussion. The paper should also state explicitly whether these fits are weighted by statistical errors only or by combined statistical and systematic errors.","section":"Table II and Figs. 2–3","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The systematic uncertainty on mπ/fπ for ml = 0.007 is quoted as '±0.03' while the other two entries carry three significant digits. The entries should be given to a consistent precision.","section":"Table II","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The pion decay constant is computed from Eq. (11) using m_dwf and m_res, and the m_res values in Table I carry roughly 2% uncertainties. The paper should state whether these are propagated into the quoted fπ errors.","section":"Eq. (11)","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"ΔE1 and |p| are tabulated for all three masses, but §IV.C explains that the phase shift can be extracted only at the heaviest mass because the first excited state lies near the four-pion inelastic threshold at the two lighter masses. The meaning of the ΔE1 entries for those two masses should be clarified, since they do not correspond to a clean ππ scattering state.","section":"§IV.C and Table II","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The extracted δ = −43 ± 10 ± 5° at |p| ≈ 544 MeV would be more informative if compared with the one-loop χPT prediction for the phase shift at that momentum, as a further cross-check of the same low-energy constants used in the scattering-length fit.","section":"§IV.C","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The statement that 'the chiral logarithm is resolved by our data at this order in the chiral expansion' is stronger than the fit quality supports; the alternative fit without the logarithm gives parameters inconsistent with l4^phy, which is suggestive but not decisive. The wording should be softened or supported with the fit quality.","section":"§IV.A","authors_acknowledge":null,"authors_acknowledge_p":null},{"comment":"The grant number for K.O. is printed as 'DF-FC02-94ER40818'; this appears to be a typographical error for 'DE-FC02-94ER40818'.","section":"Acknowledgments","authors_acknowledge":null,"authors_acknowledge_p":null}],"recommendation":"minor_revision"},"author_rebuttal":null,"desk_editor":{"model":"selfhosted/deepseek-v4.1-flash","letter":"Worth your time if you care about lattice determinations of χPT low-energy constants, but read it for the three direct lattice measurements, not the headline number. This is the first fully-dynamical 2+1-flavor calculation of the I=2 ππ scattering length down in the chiral regime — mπ = 294, 348, 484 MeV — using Lüscher's finite-volume method with domain-wall valence quarks on MILC asqtad configurations. That is a real step past CP-PACS's two-flavor work, and the machinery (jackknife, covariance-matrix fits, fit-range variation for systematics) is done properly. The phase shift at |p| ≈ 544 MeV, δ = −43 ± 10 ± 5°, is a genuine addition, though it is one point at one mass and nothing more can be said about it. The extraction of lππ = 3.3 ± 0.6 ± 0.3 is the useful byproduct for people doing χPT phenomenology. The soft spot is the extrapolation, and the stress-test note has it right. The data show essentially no curvature — C = 0.29, 0.02, 0.02, consistent with zero at the two lighter masses — and the authors say so plainly in §IV.B: there is 'an almost perfect cancellation between the counterterm and the logarithm.' So the fit form, not the data, is doing most of the work in getting from mπ/fπ ≈ 2–3 down to 1.04. The quoted ±0.0018 truncation error comes from comparing one-loop to simplified two-loop fits, but footnote 8's own LO×A alternative gives −0.0462, which is 0.0036 from the headline — twice the quoted truncation error. That is an internal inconsistency in the error budget, not a wrong central value. The result still sits comfortably on the K(e4) number, −0.0454 ± 0.0031, so the physics survives; the precision claim does not. Calling the extrapolated value a 'prediction' is a mild overreach, but a minor one. The other acknowledged limitations — single lattice spacing, mixed-action partial quenching not treated with the mixed-action χPT they cite in Ref. [43], heaviest mass at the chiral edge — are all stated and all fair. This deserves a serious referee, mainly to press on the error budget rather than the method. I would cite it for the lattice points and lππ.","headline":"A solid first 2+1-flavor dynamical lattice calculation of the I=2 ππ scattering length; the direct lattice points are the real result, and the extrapolated precision is overstated by roughly a factor of two.","tokens_in":18653,"tokens_out":943,"duration_ms":17278},"pith_extraction":{"msc":[],"pacs":["12.38.Gc","13.75.Lb","12.39.Fe"],"model":"selfhosted/deepseek-v4.1-flash","headline":"By measuring how two pions shift each other's energy levels inside a finite lattice box, and fitting the result with one-loop chiral perturbation theory, the authors extract the I=2 pion-pion scattering length at the physical pion mass and ","keywords":["lattice QCD","pion-pion scattering","scattering length","chiral perturbation theory","Lüscher finite-volume method","domain-wall fermions","mixed-action lattice QCD","2+1 flavor QCD"],"falsifier":"A lattice calculation at a substantially lighter pion mass, or with enough new mass points to fit the two-loop curvature rather than assume it, would settle the extrapolation: if the resulting $m_2 a_2$ at the physical point moves outside the quoted truncation band around $-0.0426$, the one-loop extrapolation is inadequate.","tokens_in":18946,"feed_emoji":"⚛️","tokens_out":2933,"duration_ms":53982,"temperature":0.7,"pith_summary":"The authors want to know whether lattice QCD, run with realistic dynamical quarks, can reproduce the strength of pion-pion scattering in the channel where the two pions carry isospin 2. That channel is the cleanest testing ground for the chiral symmetry of QCD: leading-order chiral perturbation theory makes a parameter-free prediction for the scattering length, and the corrections to it are the thing lattice data are supposed to pin down. Because a Euclidean lattice cannot give S-matrix elements directly, the authors use Lüscher's finite-volume method, in which the small shift in the energy of two pions in a box encodes their scattering length. Running at pion masses of roughly 294, 348 and 484 MeV on fully dynamical 2+1-flavor configurations, then extrapolating with the one-loop chiral formula, they obtain $m_2 a_2 = -0.0426$ with a dominant truncation uncertainty of $0.0018$, close to the measured value. If this holds, lattice QCD has for the first time delivered a controlled handle on the next-to-leading-order constant $l_{2 2}$ that governs this channel.","feed_headline":"Pion-pion scattering length computed from full lattice QCD","feed_subtitle":"Three pion masses, one chiral fit, and a result that lands on the measured value within the quoted truncation error.","key_machinery":"The load-bearing device is Lüscher's finite-volume relation, $p coth 2(p) = (1/(2L)) S(p^2 L^2 / 4 2^2)$, which converts the energy of two interacting pions in a periodic box into their scattering phase shift; at small momentum it reduces to $E_0 = -4 a/(m L^3)[1 + c_1 a/L + c_2 (a/L)^2]$. The energy difference itself is isolated by forming the ratio $G_{22}(p,t) = C_{22}(p,t)/C_2(t)^2$, which cancels correlated fluctuations and behaves as a sum of exponentials in the energy shifts. The bridge from lattice masses to the physical point is the one-loop chiral expression $m_2 a_2 = -(m_2^2/8 f_2^2)[1 + (3 m_2^2/16 2^2 f_2^2)( (m_2^2/2^2) + l_{2 2})]$, with a single fitted constant $l_{2 2}$.","core_discovery":"The paper reports a fully dynamical lattice determination of the $I=2$ pion-pion scattering length in the chiral regime, and from it a value for the combination of next-to-leading-order chiral constants that enters that channel. The three lattice points give $m_2 a_2 = -0.212$, $-0.222$ and $-0.38$ at pion masses of about 294, 348 and 484 MeV; fitting the one-loop chiral expression to them yields $l_{2 2} = 3.3$ and, extrapolated to the physical pion mass, $m_2 a_2 = -0.0426$ with statistical, fitting and truncation errors of $0.0006$, $0.0003$ and $0.0018$. The authors also extract, at the heaviest pion mass only, the phase shift at nonzero momentum, $|p|$ around 544 MeV, finding $-43$ degrees with errors of 10 and 5 degrees. They note that the fitted counterterm and the chiral logarithm nearly cancel over the mass range they cover, so their data do not by themselves reveal the curvature the one-loop formula predicts.","pith_inferences":["The quoted truncation error of $0.0018$ comes from comparing the one-loop fit against simplified two-loop fits with the higher constants set to zero; a genuine three-parameter two-loop fit, once enough mass points exist, could move the central value by more than that estimate rather than less.","Adding a fourth lattice ensemble near 200 MeV would do more for this channel than adding statistics at the existing masses, because it is the curvature, not the error bars, that the fit currently cannot see.","The near-cancellation the authors report between counterterm and logarithm means the same data are consistent with a family of chiral parameter sets; a determination of $l_{2 2}$ from an independent process would break the degeneracy.","The phase-shift extraction at only the heaviest mass suggests that pushing to lighter pions will require larger volumes or a dedicated multi-state variational analysis, since the first excited two-pion level crowds the four-pion threshold."],"forward_implications":["The $I=2$ channel acquires a lattice-anchored value of the next-to-leading-order chiral constant $l_{2 2}$, which can be inserted into chiral computations of more complex processes that are too costly to simulate directly.","The same technology transfers to the $I=0$ channel and to $K  o 2 2$ amplitudes, where the relevant low-energy constants are otherwise poorly constrained.","The extracted phase shift at $|p|$ around 544 MeV gives a first nonzero-momentum check of the finite-volume machinery at these quark masses.","Because both $m_2 a_2$ and $m_2/f_2$ are dimensionless, the extrapolated scattering length carries no scale-setting uncertainty, making it a clean benchmark for future lattice calculations.","The agreement with the measured value and with the two-loop chiral prediction gives an independent consistency check on the chiral expansion of the pion-pion amplitude."],"supporting_citations":[{"why":"Supplies Lüscher's finite-volume formalism relating two-particle energy levels in a box to the scattering phase shift, the method the whole calculation rests on.","marker":"[2]"},{"why":"Source of the ratio-of-correlators construction $G_{22} = C_{22}/C_2^2$ used to isolate the interaction energy shift.","marker":"[4]"},{"why":"Provides the asqtad-improved MILC 2+1-flavor gauge configurations on which the calculation is performed.","marker":"[36]"},{"why":"Describes the smeared-source domain-wall propagators whose overlap properties the one- and two-pion correlators depend on.","marker":"[31]"},{"why":"Gasser and Leutwyler's one-loop chiral perturbation theory supplies the amplitude, the $f_2$ chiral expansion, and the constants $l_i$ used in the extrapolation.","marker":"[52]"},{"why":"Provides the two-loop chiral prediction for $m_2 a_2$ against which the extrapolated result is compared.","marker":"[53]"},{"why":"Supplies the explicit two-loop expression for the scattering length used to estimate the truncation error.","marker":"[56]"},{"why":"CP-PACS dynamical two-flavor result at a comparable lattice spacing, the external lattice benchmark the new data are plotted against.","marker":"[22]"},{"why":"The $K(e4)$ measurement giving the experimental value of the $I=2$ scattering length that the extrapolated result is compared with.","marker":"[57]"},{"why":"MILC's own scale-setting and pion-mass determination on the same ensembles, used to cross-check the lattice spacing and masses obtained here.","marker":"[49]"}],"fun_headline_variants":["Lattice QCD computes I=2 pion-pion scattering length","Three pion masses give I=2 scattering length in lattice QCD","Chiral fit to lattice data agrees with measured pion scattering","Lattice data do not reveal the chiral curvature the fit assumes","Lattice scattering length agrees with experiment within truncation error"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The one-loop chiral formula with a single fitted constant correctly carries the scattering length from pion masses of 294 to 484 MeV down to 138 MeV, even though the authors themselves observe that the counterterm and the logarithm nearly cancel over the fitted range and that their data cannot resolve the predicted curvature.","fun_headline_variants_meta":{"raw":{"variants":["Lattice QCD computes I=2 pion-pion scattering length","Three pion masses give I=2 scattering length in lattice QCD","Chiral fit to lattice data agrees with measured pion scattering","Lattice data do not reveal the chiral curvature the fit assumes","Lattice scattering length agrees with experiment within truncation error"]},"model":"selfhosted/deepseek-v4.1-flash","effort":"low","attempts":1,"cost_usd":0.0,"raw_usage":{"total_tokens":2731,"prompt_tokens":952,"completion_tokens":1779,"prompt_tokens_details":null,"completion_tokens_details":{"reasoning_tokens":1706}},"tokens_in":952,"tokens_out":1779,"duration_ms":30200,"image_pages":[],"temperature":1.0,"reasoning_budget":50,"reasoning_tokens":1706,"transport_provider":"selfhosted","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-13T08:09:50.356223+00:00","model_set":{"reader":"selfhosted/deepseek-v4.1-flash"},"pipeline_version":"pith-pipeline@v1.5.0-archive-selfhost","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"A lattice calculation at a substantially lighter pion mass, or with enough new mass points to fit the two-loop curvature rather than assume it, would settle the extrapolation: if the resulting $m_2 a_2$ at the physical point moves outside the quoted truncation band around $-0.0426$, the one-loop extrapolation is inadequate.","supporting_citations":[{"cited_title":"Phys. Rev. 105, 767 (1957); H. W. Hamber, E. Marinari, G. Parisi and C. Rebbi, Nucl. Phys. B 225, 475 (1983); M. L¨ uscher, Commun. Math. Phys. 105, 153 (1986); M. L¨ uscher, Nucl. Phys. B 354, 531","cited_arxiv_id":null,"evidence_quote":"Supplies Lüscher's finite-volume formalism relating two-particle energy levels in a box to the scattering phase shift, the method the whole calculation rests on."},{"cited_title":"$I=2$ pion scattering amplitude with Wilson fermions","cited_arxiv_id":"hep-lat/9301016","evidence_quote":"Source of the ratio-of-correlators construction $G_{22} = C_{22}/C_2^2$ used to isolate the interaction energy shift."},{"cited_title":"The QCD spectrum with three quark flavors","cited_arxiv_id":"hep-lat/0104002","evidence_quote":"Provides the asqtad-improved MILC 2+1-flavor gauge configurations on which the calculation is performed."},{"cited_title":"Lattice Computations of the Pion Form Factor","cited_arxiv_id":"hep-lat/0411028","evidence_quote":"Describes the smeared-source domain-wall propagators whose overlap properties the one- and two-pion correlators depend on."},{"cited_title":"Annals Phys. 158, 142","cited_arxiv_id":null,"evidence_quote":"Gasser and Leutwyler's one-loop chiral perturbation theory supplies the amplitude, the $f_2$ chiral expansion, and the constants $l_i$ used in the extrapolation."},{"cited_title":"\\pi\\pi scattering","cited_arxiv_id":"hep-ph/0103088","evidence_quote":"Provides the two-loop chiral prediction for $m_2 a_2$ against which the extrapolated result is compared."},{"cited_title":"Pion-pion scattering at low energy","cited_arxiv_id":"hep-ph/9707291","evidence_quote":"Supplies the explicit two-loop expression for the scattering length used to estimate the truncation error."},{"cited_title":"I=2 $\\pi\\pi$ Scattering Phase Shift with two Flavors of $O(a)$ Improved Dynamical Quarks","cited_arxiv_id":"hep-lat/0402025","evidence_quote":"CP-PACS dynamical two-flavor result at a comparable lattice spacing, the external lattice benchmark the new data are plotted against."},{"cited_title":"High statistics measurement of Ke4 decay properties","cited_arxiv_id":"hep-ex/0301040","evidence_quote":"The $K(e4)$ measurement giving the experimental value of the $I=2$ scattering length that the extrapolated result is compared with."},{"cited_title":"Light pseudoscalar decay constants, quark masses, and low energy constants from three-flavor lattice QCD","cited_arxiv_id":"hep-lat/0407028","evidence_quote":"MILC's own scale-setting and pion-mass determination on the same ensembles, used to cross-check the lattice spacing and masses obtained here."}],"lane_accuracy":{"status":"not yet measured","lane":{"model":"selfhosted/deepseek-v4.1-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}