Pith. sign in

REVIEW 1 major objections 5 minor 1 cited by

Hamiltonian renormalisation VIII. P(Phi,2) quantum field theory

T0 review · 1 major / 5 minor · reviewed 2026-08-15 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The Hamiltonian renormalisation flow reproduces the continuum $\mathrm{P}(\Phi)_2$ model on the circle as its fixed point.

desk verdict The real news is the finite-step convergence of a naive local coupling to the quasi-local fixed point in §4.3; the projected-flow part is a consistency check, and the paper is mostly honest about that, even if the abstract overstates it. read the letter →

arxiv 2505.13030 v2 pith:BENHDCKL submitted 2025-05-19 hep-th gr-qc

M. Rodriguez Zarate , T. Thiemann This is my paper · ORCID
classification hep-thgr-qc MSC 81T0881T1681T17 PACS 11.10.Gh11.10.Kk
keywords HamiltonianrenormalisationP(Phi)2modelDirichletkernelfixedpointconstructivequantumfieldtheoryFockrepresentationcoarsegrainingquasi-localcoupling
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper extends a Hamiltonian renormalisation scheme, previously tested on free fields, to an interacting quantum field theory: the self-interacting scalar field with polynomial potential in two spacetime dimensions, $\mathrm{P}(\Phi)_2$, on a finite circle. Its central claim is that the renormalisation flow — a blocking procedure that removes high-resolution degrees of freedom — has the known continuum $\mathrm{P}(\Phi)_2$ theory as a fixed point. Because this theory is one of the few interacting QFTs with a rigorous non-perturbative construction, it gives the scheme a benchmark it did not have for interacting fields. The paper shows the natural Fock family is already the one blocked from the continuum, and that a separate discrete blocking calculation drives a naive local starting coupling to the correct quasi-local fixed point in a finite number of steps.

What carries the argument

The central object is the Dirichlet kernel $P_M(x,y)=\sum_{n\in Z_M}e_n(x-y)$, an orthogonal projection onto the $M$-mode subspace $L_M$ spanned by Fourier modes with $|n|\le (M-1)/2$; it acts as a smoothed replacement for the delta distribution. The load-bearing identity is $\omega\,P_M = P_M\,\omega_M$ with $\omega=(p^2-\partial^2)^{1/2}$ and $\omega_M=(p^2-\partial_M^2)^{1/2}$, which holds because the spatial derivative $\partial$ preserves $L_M$. In the discretised version, the same property makes the coarse-graining map $I_M$ and its threefold interpolation $I_M^{3M}$ compatible with the derivative, so the flow equation (4.21) contracts the momentum-conservation delta function $\delta_{n_1+\cdots+n_k,0\,(\mathrm{mod}\,3^r M)}$ until the modulo constraint drops out and the quasi-local coupling (4.15) is reached.

What would settle it

Compute the free-theory matrix elements that define blocking from the continuum using a non-Dirichlet kernel, such as the Schwarz kernel, with the same initial Fock family; the paper predicts a mismatch, so a match would refute the claimed necessity of the Dirichlet kernel's intertwining property. The same comparison at finite resolution $M$ for the interacting theory would show whether the fixed point is an artefact of this kernel choice.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that the Hamiltonian renormalisation flow has the continuum $\mathrm{P}(\Phi)_2$ model as a fixed point, in both formulations used. In the projected-field picture the flow is fixed immediately: because the Dirichlet-kernel subspace $L_M$ is invariant under the spatial derivative, the covariance satisfies $\omega\,P_M = P_M\,\omega_M$, which makes the natural Fock family of Section 3 identical to the family obtained by blocking the continuum theory of Section 2. In the discretised picture the statement is stronger: starting from the ultra-local coupling $g^{(0)}_{M;m_1,\dots,m_k} = M^{k-1}\prod_{s=1}^{k-1}\delta_{m_s,m_k}$, the blocking equation drives the coupling after at most $r_k = 1 + \lfloor \ln(k/2)/\ln 3\rfloor$ steps to the quasi-local fixed point $g_{M;m_1,\dots,m_k}$ of eq. (4.15). The authors read this as the first demonstration that their Hamiltonian renormalisation scheme, previously applied only to free fields, works for an interacting QFT.

Load-bearing premise

The result depends on the coarse-graining kernel being the Dirichlet kernel, whose projection leaves the derivative-invariant mode subspace intact; if that property fails, the natural starting family is not the blocked continuum family even for free theories, and only the separate discrete-blocking calculation supports the fixed-point claim.

Editorial extensions

If this is right

  • At every finite resolution $M$, the fixed-point family of the flow equals the family obtained by blocking the continuum $\mathrm{P}(\Phi)_2$ theory, so the scheme is consistent with the known rigorous construction.
  • A naive ultra-local starting coupling is not preserved by blocking: after one step it becomes quasi-local, and after finitely many steps it has reached the fixed-point coupling $g_{M;m_1,\dots,m_k}$.
  • For a polynomial of degree $k$, the number of blocking steps needed is at most $1+\lfloor\ln(k/2)/\ln 3\rfloor$, so convergence is logarithmically fast in the polynomial degree and independent of the resolution $M$.
  • The Dirichlet kernel's smoothness and derivative-invariance are responsible; with position-more-local kernels such as the Schwarz kernel, even free scalar theories fail to have the natural family as the blocked family.
  • In higher dimensions with a Fock representation adapted to the free Hamiltonian, the same flow would find at best a quadratic form as its fixed point rather than an operator, so additional dressing transformations would be needed to promote it to an operator.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the fixed-point claim is correct, the known $\mathrm{P}(\Phi)_2$ answer becomes a calibration target: for theories without a known continuum solution, convergence of the same flow to a quasi-local fixed point could serve as evidence that a continuum Hamiltonian exists, while non-convergence would flag a problem.
  • The finite-step convergence with ratio 3 suggests a testable scaling law: for a coarse-graining ratio $q$, the required number of steps may grow like $\ln(k)/\ln q$; running the flow for $q\neq 3$ would show whether the bound $1+\lfloor\ln(k/2)/\ln q\rfloor$ is generic or special to this choice.
  • The paper leaves open non-polynomial and unbounded potentials; an editorially suggested extension is to run the same discretised flow for a potential with infinitely many terms, where finite-step convergence in the degree is no longer automatic.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

1 major / 5 minor

Summary. This paper applies the Hamiltonian renormalisation scheme developed in earlier works, with Dirichlet-kernel coarse graining, to the P(Φ)_2 model on a circle in finite volume. The authors define a natural discretised family in the Fock representation selected by the free Hamiltonian, with one-particle kernel ω_M^(0) = (p^2 − Δ_M)^(1/2). They show that this family coincides with the family obtained by blocking the continuum theory, and that the projected-field renormalisation flow is a fixed point because the Dirichlet projection P_M intertwines ω and ω_M (Eqs. (4.11)–(4.13)). To obtain a non-trivial check, they introduce a naive, perfectly local interaction coupling g_M^(0) in the χ-basis and derive the discrete blocking flow for the coupling constants (Eq. (4.21)). They prove by induction that after r_k = 1 + [ln(k/2)/ln 3] steps the coupling reaches the quasi-local fixed point g_M (Eq. (4.15)). The paper also gives an elementary proof in Appendix A that the normal-ordered interaction is a densely defined symmetric operator in the free Fock representation. The stated conclusion is that the scheme reproduces the known continuum P(Φ)_2 Hamiltonian at finite volume.

Significance. If the result stands, this is the first application of this particular Hamiltonian renormalisation scheme to an interacting quantum field theory and a non-trivial demonstration that the scheme can reproduce a known constructive-QFT fixed point. The paper's strengths include an explicit finite-step convergence bound independent of the resolution scale M; a closed-form expression for the quasi-local fixed-point coupling with a detailed analysis of its locality properties; a clear identification of the intertwining condition (4.11)/(4.13) that makes the natural initial family a fixed point; and a self-contained proof of the dense definability of the interaction in the free Fock representation. The authors are also candid about the kernel dependence of the construction, noting that the intertwining property is special to the Dirichlet kernel and fails for kernels such as the Schwarz kernel even in the free case. No internal inconsistency or mathematical error was found in the central flow computation, the finite-step bound, or the fixed-point verification.

major comments (1)
  1. [Abstract; §4.2; §4.3] The abstract's unconditional claim that the Hamiltonian renormalisation flow 'finds this theory indeed as a fixed point' is broader than what is demonstrated. In §4.2 the projected-field flow is a fixed point by construction: the initial Fock kernel (3.8) is the compression of the continuum kernel, and the intertwining identity (4.13) forces the natural family to be both the blocked-continuum family and the fixed point. The genuinely dynamical result is the convergence of the interaction coupling in §4.3, starting from the naive local coupling (4.18), and even there the Fock state and the free part are held fixed at the natural choice. The authors should state this scope explicitly in the abstract and conclusions, distinguishing the consistency check of §4.2 from the non-trivial flow of §4.3. This is a presentation issue, but it affects the central advertised claim and should be corrected before publication.
minor comments (5)
  1. [§4.3, text before Eq. (4.20)] The displayed intertwining identity in the text has a type mismatch: it should read ω^{-1}_{3M} I_{M3M} = I_{M3M} ω^{-1}_M, not I_{M3M} ω^{-1}_{3M}. The subsequent equations use the correct version, so this is a typo, but it should be fixed for clarity.
  2. [§4.3, bound after Eq. (4.26)] The stated number r_k = 1 + [ln(k/2)/ln 3] is an upper bound; when k/2 is an exact power of 3 the expression overestimates the minimal sufficient number of steps by one. The exact sufficient value is ceil(log_3(k/2)), with at least 1. Since the text says 'at most r_k steps', the claim is not wrong, but the formula could be made tighter.
  3. [§4, introductory paragraph] The sentence 'the former is in the spirit of renormalisation schemes outside a lattice context while the former emphasises the traditional real space block spin interpretation' should read 'while the latter emphasises' in the second clause.
  4. [Title and Abstract] The notation P(Φ)2 is easily misread; P(Φ)_2 would be clearer and consistent with the body of the paper.
  5. [Figures 1–4] The figures illustrating the quasi-local coupling would be easier to read with explicit axis labels and a statement of the colour scale; as presented, quantitative values are hard to extract from the interpolated surfaces.

Circularity Check

1 steps flagged · score 3.0 of 10

Projected-field fixed point is constructed from the initial data; the discretised flow provides independent support.

  1. self definitional [Section 3, eq. (3.8); Section 4.2, eqs. (4.12)-(4.13)]
    "The specific form of h0,M suggests a Fock quantisation with annihilators ... [ωM]2 = p2 1M − ΔM ... Going through literally the same steps as in section 4.1, we find that the flow is trivial: ω(n)F,M = ω(0)F,M = ω∗F,M = ωF,M and H(n)M = H(0)M = H∗M = HM ... This is again due to (for M′>M) ωM′·PM = PM·ωM."

    The initial Fock kernel ω(0)M is defined in (3.8) by projecting the continuum free Hamiltonian to L_M, i.e. ωM^2 = p^2 − Δ_M with Δ_M = ∂_M^2 = (P_M ∂ P_M)^2. The blocking-from-the-continuum computation in §4.1 produces exactly the same projected operator because the Dirichlet kernel preserves L_M, so ∂ P_M = P_M ∂ and hence (4.11)/(4.13) hold. Therefore the projected-field flow in §4.2 is at its fixed point by construction: the 'fixed point family' is the initial family, and no independent discovery is made in that part. The authors acknowledge this at the start of §4.3, where they write that one should use a deviating initial family 'to test ... less trivially'.

full rationale

The paper's central fixed-point claim splits into two parts. In §4.1–4.2 the fixed point is a consistency check: the initial family (3.3)/(3.8) is defined by compressing the continuum operator onto L_M, and the Dirichlet-kernel identity ω·P_M = P_M·ω_M makes the blocking of the continuum state coincide with that initial family, so the projected flow is trivial. This part is circular in the narrow self-definitional sense, and the authors themselves flag it by introducing §4.3 'to test ... less trivially'. The discretised flow of §4.3 is genuinely independent: starting from the perfectly local coupling (4.18), the exact flow equation (4.21) drives the coupling to the quasi-local fixed point (4.15) in at most r_k = 1 + floor(ln(k/2)/ln 3) steps, with the finite-step computation carried out explicitly in (4.23)–(4.26). No coupling constant is fitted to any target, no external benchmark is used to adjust parameters, and the convergence proof is self-contained. The cited earlier work supplies the Dirichlet-kernel projection scheme, but the fixed-point verification here does not reduce to a self-citation chain: the decisive equations are derived in the paper. Kernel-dependence is a real scope limitation, not a circularity, since the authors state that other kernels such as the Schwarz kernel fail even for free theories. Overall, the projected-field result is constructed, but the discretised result gives the headline claim independent content, so the circularity score is moderate.

Assumptions & free parameters 0 free parameters · 4 assumptions · 0 invented entities

The central claim rests mostly on the special intertwining property of the Dirichlet kernel and on the choice of initial Fock family; once these are granted, the derivations in sections 4.2 and 4.3 are explicit. No free parameters are fitted and no new entities are postulated.

assumptions (4)
  • domain assumption The spatial circle has finite circumference and p^2 > 0, so the inverse covariance omega^{-1} is integrable enough for the interaction V to be densely defined on Fock vectors with compact momentum support.
    Invoked in section 2.1 (p^2 > 0 to avoid zero modes) and proved in Appendix A; without compactness and positive mass the density argument in (A.8)-(A.10) fails.
  • ad hoc to paper The Dirichlet kernel projection P_M commutes with the spatial derivative (equivalently, the subspace L_M is invariant under partial_x), and the coarse graining map M' = 3M is chosen.
    Identity (4.13) is the reason the natural initial family is already the fixed point in the projected-field framework; the paper states this fails for e.g. the Schwarz kernel (end of section 4.1). The finite-step convergence count depends on the ratio 3.
  • ad hoc to paper The initial family is quantised in the Fock representation selected by the free Hamiltonian, with normal ordering and kernel omega_M^(0) = sqrt(p^2 - Delta_M).
    This 'natural' initial condition is the input whose fixed-point property is verified in section 4.2; section 4.3 notes that a different initial family would make the test nontrivial.
  • standard math Standard facts of Fock space and Weyl algebra GNS construction are used.
    Background assumed throughout sections 2-3 and Appendix B; these are textbook results (e.g., Bratteli-Robinson).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Hamiltonian renormalisation VIII. P(Phi,2) quantum field theory." pith.science (2026). https://pith.science/paper/BENHDCKL

@misc{pith2026250513030,
  author       = {Pith},
  title        = {Pith review of: Hamiltonian renormalisation VIII. P(Phi,2) quantum field theory},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BENHDCKL}},
  note         = {Machine review of arXiv:2505.13030}
}
read the original abstract

In previous works in this series we focussed on Hamiltonian renormalisation of free field theories in all spacetime dimensions. In this paper we address the Hamiltonian renormalisation of the self-interacting scalar field in two spacetime dimensions with polynomial potential, called P(Phi,2). We consider only the finite volume case. The P(Phi,2) theory is one of the few interacting QFT's that can be rigorously constructed non-perturbatively. We find that our Hamiltonian renormalisation flow finds this theory indeed as a fixed point.

Figures

Figures reproduced from arXiv: 2505.13030 by the authors.

Figure 1
Figure 1. Left: Coupling for M = 11 and m3 = 5. Right: Coupling for M = 21 and m3 = 10. 11 [PITH_FULL_IMAGE:figures/full_fig_p011_1.png] view at source ↗
Figure 2
Figure 2. Left: Coupling for M = 51 and m3 = 25. Right: Coupling for M = 71 and m3 = 35. (a) [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗
Figure 3
Figure 3. Coupling for M = 111 and m3 = 55. Left: Full range. Right: Vicinity of maximum. In figure 4 we zoom into the √ M vicinity of the maximum at m1 = m2 = m3 = 55 for M = 111. We cut off the maximum peak at a convenient value in order not to suppress the values of the coupling in its vicinity. 12 [PITH_FULL_IMAGE:figures/full_fig_p012_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: High resolution in the vicinity of the maximum for coupling for [PITH_FULL_IMAGE:figures/full_fig_p013_4.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. You Had One Job: Per-Task Quantization Using LLMs' Hidden Representations

    cs.CL 2025-11 reject novelty 6.0 of 10

    TAQ estimates per-layer importance from hidden representations and output sensitivity on task calibration data to allocate mixed precision in a training-free PTQ setting, outperforming task-agnostic baselines on accur...

Reference graph

Works this paper leans on

23 extracted references · 3 canonical work pages · cited by 1 Pith paper

  1. [1]

    Local Quantum Physics

    R. Haag, “Local Quantum Physics”, Springer Verlag, Berlin, 1984

    1984

  2. [2]

    Quantum Physics

    J. Glimm and A. Jaffe, “Quantum Physics”, Springer Verlag, New York, 1987

    1987

  3. [3]

    K. G. Wilson. The renormalization group: Critical phenomena and the Kondo problem. Rev. Mod. Phys. 47 (1975) 773

    1975

  4. [4]

    Thiemann

    T. Thiemann. Canonical quantum gravity, constructive QFT and renormalisation. Front. in Phys. 8 (2020) 548232. arXiv:2003.13622 [gr-qc]

    arXiv 2020

  5. [5]

    T. Lang, K. Liegener, T. Thiemann. Hamiltonian Renormalisation I. Derivation from Osterwalder-Schrader Reconstruction. Class. Quant. Grav. 35 (2018) 245011. [arXiv:1711.05685]

    arXiv 2018

  6. [6]

    T. Lang, K. Liegener, T. Thiemann. Hamiltonian Renormalisation II. Renormalisation Flow of 1+1 dimen- sional free, scalar fields: Derivation. Class. Quant. Grav. 35 (2018) 245012. [arXiv:1711.06727]

    arXiv 2018

  7. [7]

    T. Lang, K. Liegener, T. Thiemann. Hamiltonian Renormalisation III. Renormalisation Flow of 1+1 dimen- sional free, scalar fields: Properties. Class. Quant. Grav. 35 (2018) 245013. [arXiv:1711.05688]

    arXiv 2018

  8. [8]

    T. Lang, K. Liegener, T. Thiemann. Hamiltonian Renormalisation IV. Renormalisation Flow of D+1 dimen- sional free scalar fields and Rotation Invariance. Class. Quant. Grav. 35 (2018) 245014, [arXiv:1711.05695]

    arXiv 2018

Show all 23 references
  1. [9]

    Liegener, T

    K. Liegener, T. Thiemann. Hamiltonian Renormalisation V. Free Vector Bosons. Front. Astron. Space Sci. 7 (2021) 547550. e-Print: 2003.13059 [gr-qc]

    2021 arXiv

  2. [10]

    Thiemann

    T. Thiemann. Hamiltonian Renormalisation VII. Free Fermions and doubler free kernels. Phys. Rev. D108 (2023) 12, 125007. e-Print: 2207.08291 [hep-th]

    2023 arXiv

  3. [11]

    Zwicknagel

    E.-A. Zwicknagel. Hamiltonian renormalization. VI. Parametrized field theory on the cylinder. Phys. Rev. D108 (2023) 12, 125006. e-Print: 2207.08290 [gr-qc]

    2023 arXiv

  4. [12]

    The P( ϕ)2 Euclidean (Quantum) Field Theory

    B. Simon, “The P( ϕ)2 Euclidean (Quantum) Field Theory”, Princeton Unviersity Press, 1974

    1974

  5. [13]

    Daubechies

    I. Daubechies. Ten lectures of wavelets. Springer Verlag, Berlin, 1993. A. Cohen, I. Daubechies, P.Vial. Wavelets on the interval and fast wavelet transforms. Appl. and Comp. Harm. Anlysis, Elsevier, 1993. [hal-01311753]

    1993

  6. [14]

    Thiemann

    T. Thiemann. Renormalization, wavelets, and the Dirichlet-Shannon kernels. Phys. Rev. D 108 (2023) 12, 125008. e-Print: 2207.08294 [hep-th]

    2023 arXiv

  7. [15]

    Operator Algebras and Quantum Statistical Mechanics

    O. Bratteli, D. W. Robinson, “Operator Algebras and Quantum Statistical Mechanics”, vol. 1,2, Springer Verlag, Berlin, 1997

    1997

  8. [16]

    Boson Fields with the :ϕ4 : Interaction in Three Dimensions

    J. Glimm, “Boson Fields with the :ϕ4 : Interaction in Three Dimensions”, Comm. Math. Phys. 10 (1968) 1-47. J. Glimm, A. Jaffe, Positivity of the ϕ4 3 Hamiltonian”, Fortschr. Phys. 21 (1973) 327–376

    1968

  9. [17]

    L. Smolin. The G(Newton) → 0 limit of Euclidean quantum gravity. Class. Quant. Grav. 9 (1992) 883-894. e-Print: hep-th/9202076 [hep-th]

    1992 arXiv

  10. [18]

    Exact quantisation of U(1) 3 quantum gravity via exponentiation of the hypersurface deformation algebroid. Class. Quant. Grav. 40 (2023) 24, 245003. e-Print: 2207.08302 [gr-qc]

    2023 arXiv

  11. [19]

    Narnhofer, W.E

    H. Narnhofer, W.E. Thirring. Covariant QED without indefinite metric. Rev. Math. Phys. 4 (1992) spec01, 197-211

    1992

  12. [20]

    Thiemann

    T. Thiemann. Nonperturbative quantum gravity in Fock representations. Phys. Rev. D 110 (2024) 12, 124023. e-Print: 2405.01212 [gr-qc]

    2024 arXiv

  13. [21]

    Quantum Gravity

    C. Rovelli, “Quantum Gravity”, Cambridge University Press, Cambridge, 2004. T. Thiemann, “Modern Canonical Quantum General Relativity”, Cambridge University Press, Cambridge, 2007. J. Pullin, R. Gambini, “A first course in Loop Quantum Gravity”, Oxford University Press, New Yo...

    2011 arXiv

  14. [22]

    Osterwalder, R

    K. Osterwalder, R. Schrader. Axioms for Euclidean Green’s functions. Communications in Mathematical Physics 31(2):83–112, 6 1973

    1973

  15. [23]

    Glimm and A

    J. Glimm and A. Jaffe, ”A λϕ4 quantum field theory without cutoffs. I”, Phys. Rev. 176 (1968), 1945–1951. DOI: 10.1103/PhysRev.176.1945. J. Glimm and A. Jaffe, ”The λ(ϕ4)2 quantum field theory without cutoffs: III. The physical vacuum”, Acta Math. 125 (1970), 203–267. DOI: 10....

    1968 doi

Pith tools

Reviewed August 15, 2026 · model on record in the stance chip above.