{"id":"3462c815-8130-462b-b455-cbb6792d59a6","arxiv_id":"2411.14399","paper_version":1,"verdict":"REJECT","confidence":"MODERATE","novelty_score":4.0,"correctness_risk":"high","formal_verification":"none","parameter_count":3,"one_line_summary":"DiscoTEX is extended to 12th-order discontinuous Hermite time integration, but validation is incomplete and the printed higher-order formulas contain apparent errors.","lead":"This paper extends the DiscoTEX numerical method to time-integration orders up to 12 and provides the detailed formulas. The author reports that accuracy stagnates beyond sixth order while computational cost rises, and the full wave-equation convergence tests cover only up to sixth order.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Equation (47) for JH12 is internally inconsistent: the J1 term is missing its J1 factor and J5 appears twice, so the printed 12th-order recipe cannot support the claimed convergence.","rationale":"I read the paper as claiming a usable, fully specified order-12 extension. The condition for that claim is that the printed JH12 formula be correct. It is not, on its face: the missing J1 factor is a dimensional inconsistency, and the duplicated J5 label contradicts the uniqueness of the jump contribution in the integration rule. This is a stronger problem than the absence of H8-H12 PDE convergence experiments, because even with unlimited computing time the formula as printed cannot yield the claimed order. The symbolic check would settle the issue directly. I agree with the reader's weakest_assumption. The other concerns, such as Table 2 anomalies and the absence of code or data, are real but secondary; they would matter even with a correct formula, whereas the JH12 inconsistency is a direct internal contradiction of the central algorithmic recipe. If the symbolic check somehow confirmed Eq. (47) despite appearances, the paper would still need explicit H8-H12 PDE convergence studies before the 12th-order claim could be accepted; my recommendation is therefore to keep the current rejection verdict unchanged.","tokens_in":19959,"tokens_out":5741,"duration_ms":56664,"concrete_test":"Solve the 24×24 linear system (22)-(45) symbolically for the polynomial coefficients in (19)-(21), with tν=0, tν+1=Δt, t×=Δt×, then integrate p−(t) on [0,Δt×] and p+(t) on [Δt×,Δt], and compare the resulting expression with Eq. (46)-(47). If the coefficient of J1 is absent or the two J5-labeled terms do not merge into one, Eq. (47) is wrong as printed. A second, independent check: run the H12 integrator on the Legendre test (48) with jumps (49)-(51); with nonzero J1, a correct formula must show a 12th-order error slope.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central new object of the paper is the order-12 discontinuous Hermite rule. As printed, Eq. (47) cannot be the correct jump correction. The second term, 1/44(5Δt^2−22ΔtΔt×+22Δt×^2), has dimensions of Δt^2 but is added to an integral; it should be multiplied by J1, and no J1 term appears anywhere else in JH12. In addition, two different polynomials are both labelled J5 (denominators 665280 and 30240), while each jump condition (34)-(45) should contribute exactly one term to the integrated rule. Since J1=15/8 is nonzero in the paper's own Legendre test (Eq. (49)), using Eq. (47) literally would either introduce an inconsistent dimensional term or lose the J1 contribution, destroying the claimed 12th-order rate. This is not a matter of convention: the formula is internally inconsistent. The absence of PDE convergence tests above H6 and the stagnant Table 2 errors mean this inconsistency is never checked. The central claim therefore rests on an unverified, plausibly incorrect printed formula.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript extends the DiscoTEX algorithm to time-integration orders 8, 10, and 12. The method uses piecewise Hermite interpolants in time, corrected by jump conditions, and combines them with the discontinuous collocation spatial discretization introduced in the companion paper [1]. The numerical evidence consists of a discontinuous Legendre integral test (Figure 1) and weak-form solutions of a distributionally sourced wave equation compared against exact solutions (Figures 2-5, Table 2). The paper also reports wall-clock times and concludes that orders beyond six are not cost-effective for the tested configuration.","tokens_in":20194,"tokens_out":8157,"duration_ms":74564,"significance":"If the printed formulas are correct, the manuscript would provide a systematic route to high-order weak-form solvers for distributionally sourced PDEs, with potential applications in extreme-mass-ratio inspiral modelling and black-hole spectroscopy. The numerical comparison against independent exact solutions, with no fitted constants, is a genuine strength, as is the explicit reporting of computational cost. However, the central new objects -- the order-8, order-10, and order-12 discontinuous Hermite rules -- are not verified at the PDE level, and the printed 12th-order jump correction contains apparent transcription errors. The significance of the claimed extension therefore cannot be assessed from the manuscript in its present form.","major_comments":[{"comment":"The printed JH12 expression is internally inconsistent. The second term, (1/44)(5 Delta t^2 - 22 Delta t Delta t_x + 22 Delta t_x^2), is not multiplied by J1, and no other term in JH12 carries J1; since J1 = 15/8 in the paper's own Legendre test (Eq. (49)), this either drops a nonzero jump contribution or introduces a dimensionally inconsistent term. In addition, two different polynomials are labelled J5, with denominators 665280 and 30240, although each jump condition (34)-(45) should contribute exactly one term. As printed, Eq. (47) cannot be the correct 12th-order jump correction, so the claimed 12th-order convergence in Figure 1 and Table 2 is not supported by the manuscript.","section":"Sec. 2.3, Eq. (47)"},{"comment":"The JH10 expression has the same missing-factor pattern: the term (1/18)(Delta t - 3 Delta t_x)(2 Delta t - 3 Delta t_x) appears without a J1 factor, while all neighbouring terms carry the expected jumps. This affects the printed 10th-order recipe and, together with the JH12 issue, indicates that the higher-order jump formulas need a systematic check rather than isolated typographical fixes.","section":"Sec. A.1, Eq. (A.10)"},{"comment":"PDE-level convergence is demonstrated only for orders 2, 4, and 6. Figure 5 contains DH2, DH4, and DH6 only, and Table 2 shows stagnating errors (7.7e-11 for H4, 7.2e-11 for H6, 7.2e-9 for H8, 7.2e-11 for H10 and H12) with no convergence-order study. The abstract's claim of an extension to twelve orders therefore rests on the Legendre integral test alone, which uses the very formulas questioned above. A revision must either add H8-H12 convergence studies for the wave equation or restrict the claims.","section":"Sec. 3.2, Fig. 5 and Table 2"},{"comment":"The higher-order time-jump formulas are asserted without derivation. The text states that the remaining terms are determined by 'an analogous trivial process which we omit for brevity,' but these jumps are the central new object of the paper. Without a derivation, a symbolic notebook, or a fully specified recurrence, the reader cannot verify the printed formulas; this gap is especially serious in view of the apparent errors in Eqs. (47) and (A.10).","section":"Appendix A.4, after Eq. (A.55)"}],"minor_comments":[{"comment":"The JH4 term printed as 'Delta t_x^2/12 (Delta t^2 - 6 Delta t Delta t_x + 6 Delta t_x^2) J1' appears to have the wrong prefactor; dimensionally it should be Delta t^2/12 (Delta t^2 - 6 Delta t Delta t_x + 6 Delta t_x^2) J1, so please check.","section":"Eq. (A.8)"},{"comment":"The second component of the second time-derivative source term is printed as ~s^{(2)}_{\\Psi,(2)} twice; one occurrence should presumably be ~s^{(2)}_{\\Psi,(1)}.","section":"Eq. (73)"},{"comment":"The caption says 'all the 12th order discontinuous Hermite interpolation schemes' but the table lists orders 2 through 12; the wording should be corrected.","section":"Table 1 caption"},{"comment":"The exact-solution formulas are cited to an unpublished work; for reproducibility they should either be included in an appendix or linked to a public preprint.","section":"Reference [5]"},{"comment":"The notation f(t)_{DH12} for the quadrature result is not defined and could be confused with the integrand f(t); a different symbol for the integrated quantity would improve clarity.","section":"Eq. (46)"}],"recommendation":"major_revision","confidential_remarks":"The decisive question is whether the author's implemented code differs from the printed formulas. If the code that produced Figure 1 uses a correct JH12, the manuscript can be repaired by replacing the defective formulas and adding derivations and H8-H12 PDE tests. If the code matches the printed Eq. (47), the 12th-order claim is unsupported and rejection would be appropriate."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"This paper extends your DiscoTEX method to 12th-order time integration. The new material—explicit JH8, JH10, JH12 jump-corrected Hermite rules and a Legendre time-integration convergence study—is genuine, and the author is refreshingly honest that past 6th order accuracy stagnates while runtime grows, confirming what [1] already hinted. If the formulas are corrected, the approach could matter for EMRI waveform modeling.\n\nBut as printed, the central new formula (47) is not internally consistent. The second term, (1/44)(5Δt^2−22ΔtΔt×+22Δt×^2), should be multiplied by J1; no J1 term appears anywhere in JH12. Meanwhile, two different polynomials are both labeled J5 (denominators 665280 and 30240), and each jump condition (34)–(45) should contribute exactly one term. JH10 in (A.10) has the same missing-J1 problem. These are not cosmetic typos: the formula as written cannot be correct. Since the paper's stated aim is to present the extension to twelve orders, this is a load-bearing flaw.\n\nThe other gap is that full PDE convergence is only demonstrated for H2, H4, and H6 (Figure 5), despite the abstract claiming extension to twelve orders. Table 2 gives headline errors for H8–H12 but no convergence scans, and the H8 entry (7.2×10^−9) sits oddly between H6 and H10 (both 7.2×10^−11), suggesting a possible implementation issue. No code or data are provided to check. The exact solutions also come from an unpublished note [5], which is a lesser concern.\n\nSo the paper is not acceptable in its current form. But it deserves a serious referee: the method is potentially useful, and the errors look fixable. I would recommend major revision, not desk rejection. My own verdict: reject for now, with encouragement to resubmit after a careful rewrite that corrects the jump formulas and adds convergence evidence for H8, H10, H12 on the full PDE.","headline":"Useful extension in intent, but the printed 12th-order formula is internally inconsistent and the PDE convergence evidence stops at 6th order.","tokens_in":20708,"tokens_out":4423,"would_cite":false,"duration_ms":40363,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["65M70","65L06"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that DiscoTEX's jump-corrected Hermite time integration extends to order twelve, with each order converging at its nominal rate on distributionally sourced wave equations.","keywords":["DiscoTEX","distributionally sourced wave equation","Dirac delta source","discontinuous collocation","Hermite time integration","higher-order convergence","jump conditions","hyperboloidal coordinates"],"falsifier":"Recompute the 12th-order jump correction $J_{H12}$ symbolically from the 24 collocation-and-jump conditions stated in Section 2.3 and verify that Eq. (47) is reproduced exactly, then run the DH12 scheme on the Legendre test integral (48): if the error does not decay at the 12th-order rate, the central claim is falsified.","tokens_in":19760,"feed_emoji":"🌊","tokens_out":12196,"duration_ms":100037,"temperature":0.7,"pith_summary":"This paper extends the DiscoTEX algorithm—a discontinuous collocation plus implicit-turned-explicit (IMTEX) Hermite time-integration method—to solve wave equations whose sources contain Dirac deltas and their derivatives, up to twelfth order in time. The author's aim is to supply the full higher-order jump conditions that correct the time integrator when a point-particle trajectory crosses a grid point, and to verify the scheme by computing numerical weak-form solutions to the distributionally sourced wave equation and comparing them with exact solutions. The reported convergence tests show that orders 2 through 12 achieve their nominal convergence rates, and that the 12th-order scheme reaches the same error accuracy as lower orders for the tested settings. The practical conclusion is that accuracy stagnates past sixth order while computational cost rises steeply, so the fourth-order variant is the best cost-accuracy compromise under the current control factors. If correct, this gives a concrete recipe for long evolutions, such as extreme-mass-ratio inspirals, that do not require a second interpolation step.","feed_headline":"Delta-source wave solver extended to 12th-order accuracy","feed_subtitle":"Discontinuous Hermite rules now carry jumps up to J11, and order-12 runs converge at the expected rate.","key_machinery":"The mechanism is the discontinuous Hermite time integrator: a smooth Hermite rule whose kernel is replaced by two polynomials $p_-$ (before the discontinuity) and $p_+$ (after), meeting at $t_\\times$ with prescribed jumps in value and derivatives. The jump-corrected rule at order 12 is written as the smooth trapezoidal/Hermite combination plus the correction $J_{H12}(\\Delta t_\\times, \\Delta t)$, which packages all $J_0,\\dots,J_{11}$ terms produced by the 24-condition linear solve. In the full DiscoTEX scheme the same philosophy is applied in space via a discontinuous Lagrange basis, and the time jumps are combined with differential-operator corrections through the $g$-vectors and the $\\Upsilon$ source term; the paper explicitly gives the $g^{(3)},g^{(4)},g^{(5)}$ time derivatives needed by the H8, H10, and H12 variants.","core_discovery":"The central claim is that the DiscoTEX construction generalizes from the second- and fourth-order schemes of the companion paper to a twelfth-order scheme. On each time step the integrand is represented as a pair of degree-11 polynomials matched at the crossing time $t_\\times$; the 24 unknowns are fixed by 12 endpoint collocation conditions and 12 jump conditions $J_m = p_+^{(m)}(t_\\times)-p_-^{(m)}(t_\\times)$, $m=0,\\dots,11$. Integrating the piecewise polynomial yields the corrected rule with jump term $J_{H12}(\\Delta t_\\times, \\Delta t)$; the paper also lists the full set of time jumps for the wave-equation application in a hyperboloidal chart. Applied to $\\Box\\Psi = F(\\tau)\\delta'(\\sigma-\\xi_p)+G(\\tau)\\delta(\\sigma-\\xi_p)$, the scheme reproduces the exact weak-form solution and the reported convergence rates match the nominal order of each integrator. The paper's own numerical summary is that, for equal control factors, higher-order integrators past sixth order do not improve accuracy but increase wall-clock time substantially.","pith_inferences":["Beyond the paper, the same 24-condition construction could be checked automatically by symbolic computation, which would settle whether the apparent typographical inconsistencies in Eq. (47) are typographical or substantive.","If the jump machinery is correct, the procedure should extend to sources with higher-order delta derivatives (for example, $\\delta''$), where exact solutions are available and the 12th-order convergence could be tested directly.","The accuracy stagnation observed past sixth order may be an artifact of the time-dependent trajectory jumps dominating the error; for fixed-position sources, where only the time-stepper order matters, higher-order variants might show a clearer benefit.","A natural testable extension is to apply the H12 scheme to the same wave equation with a self-consistently coupled trajectory, where no exact solution exists, and compare against lower-order variants at fixed accuracy."],"forward_implications":["Orders 2, 4, 6, 8, 10, and 12 of the jump-corrected Hermite integrator converge at their nominal rates on the distributionally sourced wave equation, verified against exact solutions.","The explicit jump formulas in Appendix A.4 give a template for applying DiscoTEX to other time-dependent point-particle trajectories, including eccentric or generic orbits.","For the tested setup, running past sixth order does not lower the error but multiplies the wall-clock time, so fourth order is the recommended setting among the available variants.","Because the higher-order schemes reach a given accuracy with larger timesteps, they are candidates for long evolutions where a second interpolation step would be undesirable."],"supporting_citations":[{"why":"Defines the base DiscoTEX algorithm, its control factors, and the lower-order discontinuous Hermite rules that this paper extends to order 12.","marker":"[1]"},{"why":"Supplies the discontinuous Galerkin weak-form framework for the distributionally sourced wave equation that DiscoTEX is benchmarked against.","marker":"[2]"},{"why":"Gives the exact solution to the distributionally sourced wave equation used as the reference solution in the convergence tests.","marker":"[3]"},{"why":"Extends the numerical treatment to linear wave equations involving derivatives of the Dirac delta, the class of sources considered in this paper.","marker":"[4]"},{"why":"Provides exact solutions to distributionally sourced wave equations with higher-order deltas, used as test problems.","marker":"[5]"},{"why":"Introduces the hyperboloidal discontinuous time-symmetric algorithm with higher-order jumps whose notation and jump machinery DiscoTEX builds on.","marker":"[7]"},{"why":"Supplies the scri-fixing hyperboloidal foliation used to map the wave equation into the numerical chart.","marker":"[9]"},{"why":"Provides the hyperboloidal initial-value formulation at null infinity underlying the numerical coordinate setup.","marker":"[10]"}],"fun_headline_variants":["DiscoTEX extends to 12th-order accuracy","Delta-source wave equations solved at order 12","Higher-order DiscoTEX: accuracy plateaus after 6th","IMTEX method reaches 12th order for distributional PDEs"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The extension stands on the correctness of the printed high-order jump formulas, especially the 12th-order correction term and the time jumps in the appendix, which the paper does not fully derive and one of which appears to contain typographical inconsistencies; if those formulas are wrong, the claimed orders of convergence do not follow.","fun_headline_variants_meta":{"raw":{"variants":["DiscoTEX extends to 12th-order accuracy","Delta-source wave equations solved at order 12","Higher-order DiscoTEX: accuracy plateaus after 6th","IMTEX method reaches 12th order for distributional PDEs"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000964,"raw_usage":{"total_tokens":4077,"prompt_tokens":892,"completion_tokens":3185,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":508,"completion_tokens_details":{"reasoning_tokens":3114}},"tokens_in":508,"tokens_out":3185,"duration_ms":21970,"temperature":1.0,"reasoning_tokens":3114,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T15:13:16.111674+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Recompute the 12th-order jump correction $J_{H12}$ symbolically from the 24 collocation-and-jump conditions stated in Section 2.3 and verify that Eq. (47) is reproduced exactly, then run the DH12 scheme on the Legendre test integral (48): if the error does not decay at the 12th-order rate, the central claim is falsified.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the discontinuous Galerkin weak-form framework for the distributionally sourced wave equation that DiscoTEX is benchmarked against."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the exact solution to the distributionally sourced wave equation used as the reference solution in the convergence tests."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Extends the numerical treatment to linear wave equations involving derivatives of the Dirac delta, the class of sources considered in this paper."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides exact solutions to distributionally sourced wave equations with higher-order deltas, used as test problems."},{"cited_title":"Zengino ˘glu, Hyperboloidal foliations and scri-fixing, Classical and Quantum Gravity 25 (14) (2008) 145002","cited_arxiv_id":null,"evidence_quote":"Supplies the scri-fixing hyperboloidal foliation used to map the wave equation into the numerical chart."},{"cited_title":"Zengino ˘glu, D","cited_arxiv_id":null,"evidence_quote":"Provides the hyperboloidal initial-value formulation at null infinity underlying the numerical coordinate setup."}],"review_version":1}