Renormalization in quantum field theory and the Riemann-Hilbert problem II: the $\beta$-function, diffeomorphisms and the renormalization group

Alain Connes; Dirk Kreimer

REVIEW

Renormalization in quantum field theory and the Riemann-Hilbert problem II: the $\beta$-function, diffeomorphisms and the renormalization group

Alain Connes, Dirk Kreimer

High Energy Physics - Theory · Submitted 21 Mar 2000 · 9 references

Pith's verdict: A landmark paper in the Hopf-algebraic renormalization program, with one acknowledged gap: Theorem 10 is proved only for m = 0, and the massive extension is asserted rather than derived.

T0 · AI review

How this review was made: An AI referee (selfhosted/deepseek-v4.1-flash) read the paper on November 11, 2026 and answered a fixed list of questions about it. T0 is the starting level every paper gets. The levels describe how deeply Pith has checked the paper, never the authors.

T0 AI review An AI read the paper and answered a fixed list of questions.
T1 Audit Automatic checks that the references and files the paper relies on exist and match.
T2 Compute The paper's code or calculations re-run with the same results.
T3 Formal statement The claims written out in a formal math language a computer can read.
T4 Proof Those claims proved, and the proof checked by computer.

How Pith checks papers →

· 1 major objection · 7 minor · Ask about this paper · The compiled record · Read the full review →

Challenge this review Comment (0) Read on arXiv ↗

Cite ▾

arXiv IDhep-th/0003188 (version 1)

Pith IDXDAPVJDJ, Pith's permanent record of this paper and its review

CitationAlain Connes, Dirk Kreimer. "Renormalization in quantum field theory and the Riemann-Hilbert problem II: the $\beta$-function, diffeomorphisms and the renormalization group." arXiv:hep-th/0003188 (2000). https://pith.science/paper/XDAPVJDJ

Cite this reviewBibTeX for Pith's review of this paper

Claim paper

Why Pith says this

Highlight any text to comment on it.

Editor's note

read the letter

Two things here are worth your time: the scattering formula (Theorem 2) and the Hopf algebra homomorphism (Theorem 10). The first is a clean general result about graded Hopf algebras — the negative part of the Birkhoff decomposition is determined by the residue via γ_-(ε) = lim_{t→∞} e^{-t(β/ε + Z_0)} e^{t Z_0}. That is a genuine closed form, and it delivers the higher pole structure of MS counterterms from the residue alone. The second is the physical punchline: g_0 = g Z_1 Z_3^{-3/2}, read as a formal power series, defines a map from the Hopf algebra of formal diffeomorphisms to H_c. The proof (Lemma 9, then the bilinear bookkeeping in (37)–(52)) is complete for the massless case, and the appendix checks it to three loops. The bundle interpretation in section 5 is the kind of reframing that changes how you think about the ε parameter. The citation pattern is clean — part I, Collins, Araki, Connes–Moscovici, all published — and the adjoined generator Z_0 is a standard semi-direct product device, not a fudge. The soft spot is the one the authors flag themselves: Theorem 10 is proved for m = 0. Section 4 says the massive case follows from m-independence of counterterms, but that is a statement about the values Φ(a_n), not about the coproduct Δ_c that the morphism property actually tests. In the massive case the coproduct carries crosses and (0)/(1)^k decorations; the massless simplification drops them because (1)^k contributes k^2 and cancels the extra propagator, whereas in the massive case it contributes k^2 + m^2 and the cancellation fails. So the massive version of Theorem 10, and hence Theorem 14 (which uses ρ as a group anti-homomorphism), is asserted rather than derived. It may well be true — the α_n are m-independent, so the extra terms might vanish on them — but the paper does not show it. That said, the core mathematical results (Theorem 2, the scattering formula) do not depend on this, and the conceptual claims about the RG flow and the bundle are robust to it. The paper is dense and for experts; the appendix is a check, not a substitute for the general proof. Who should read it: anyone working on Hopf-algebraic renormalization, Birkhoff decomposition, or the geometry of the RG. It deserves a serious referee — the massless gap is exactly the kind of thing a careful referee should pin down, and the rest of the paper is solid enough to warrant the effort.

Referee report

Major objections (1)

[§4, Theorem 10 and the paragraph following Eq. (3)] Theorem 10 is stated without mass qualification, but the proof (Lemma 9, Eqs. (37)–(52)) is carried out only after the massless simplification, as the authors acknowledge in the sentence 'This result is only valid if we perform on Hc the simplification that pertains to the massless case, m = 0'. The extension to m ≠ 0 is asserted via Lemma 4 (m-independence of counterterms, cited to Collins [4]), but Lemma 4 concerns the values γ_−, equivalently the α_n = Φ(a_n), whereas Theorem 10 asserts comultiplicativity (Φ⊗Φ)Δ_CM = Δ_c Φ with respect to the massive coproduct Δ_c. The massless simplification rests on the (1)^k decoration contributing k^2, which cancels the extra propagator; for m ≠ 0 it contributes k^2 + m^2 and the cancellation fails, so Δ_c differs from the massless coproduct on the decorated graphs. The paper does not show that Δ_c applied to the α_n agrees with the massless computation. Either qualify Theorem 10 (and Theorem 14, whose proof uses ρ as a group anti-homomorphism) to the massless case, or supply the argument that the extra decorated terms pair trivially with the α_n.

Minor objections (7)

[§4, Eqs. (51)–(52)] There are duplicate equation numbers in this section: two equations numbered (51) and two numbered (52), the second pair being Φ∘α_t = Θ_{2t}∘Φ and the Butcher-series identity. These should be renumbered.
[Appendix] The explicit check is modulo primitive three-loop vertex graphs, which are said to 'fulfil all desired identities below trivially'. Since the reader cannot verify the omission, a sentence explaining why primitive terms are trivial for the coproduct identities would strengthen the verification.
[§4, passage from Lemma 9 to Theorem 10] The sentence 'The proof follows from Lemma 9 which shows that the map from Gc to G2 given by the transpose of the algebra morphism Φ is the Lie group morphism ρ' is terse. A sentence noting that for connected graded Hopf algebras intertwining all primitive derivations implies comultiplicativity would make the logic explicit.
[§2, Eq. (17)] The hypothesis that the limit exists for all t is strong; it would help to note that Corollary 5 supplies it for the QFT loop via Lemma 4, so the general Theorem 2 is not vacuous in the intended application.
[§5, Eq. (5)] The factor 2 in Θ_{2t} with t = log g deserves a one-line check against the (μ^{ε/2}g)^{n−2} dependence of the effective action in Eq. (3).
[§1 and §6] Typos: 't'Hooft' should be ''t Hooft'; 'burried' should be 'buried'. The coinage 'diﬀeographisms' should be defined once at first use.
[§3, Lemma 4] Reference [4] (Collins) is used for the m-independence of MS counterterms; citing the specific section or page would help the reader locate the statement.

How Pith checks papers → · Full audit of this paper →

The paper

The paper claims

The paper claims that the formula for the bare coupling constant, $g_0=gZ_1Z_3^{-3/2}$, defines a Hopf algebra homomorphism from formal diffeomorphisms to Feynman graphs, so that the bare and renormalized couplings, the renormalizationgroup

The reading

The paper tries to show that the familiar formula for the bare coupling constant, $g_0=gZ_1Z_3^{-3/2}$, has a structural meaning: treated as a formal power series, it defines a Hopf algebra homomorphism from the algebra of formal coordinate changes to the Hopf algebra of Feynman graphs. If that is right, one can forget the graph-by-graph group of characters and read the bare and renormalized couplings directly from the Birkhoff decomposition of the unrenormalized effective coupling, now viewed as a loop of formal diffeomorphisms. The same map carries the renormalization group flow and the $\beta$ function into the group of formal diffeomorphisms. The payoff is a conceptual proof that the renormalization group exists, plus a closed scattering formula that fixes all higher minimal-subtraction poles from the residue of the loop.

What carries the argument

The load-bearing mechanism is the Birkhoff decomposition $gamma(epsilon)=gamma_-(epsilon)^{-1}gamma_+(epsilon)$ in the character group of the Feynman-graph Hopf algebra, together with the loop-number grading $theta_t$ and the residue $Res gamma$. The negative part $gamma_-$ is independent of the mass unit $mu$, and Theorem 2 turns this into the closed formula $gamma_-(epsilon)=lim_{t->infty}e^{-t(beta/epsilon+Z_0)}e^{tZ_0}$ with $beta=Y Res gamma$. That formula is what converts the residue into the full tower of minimal-subtraction poles and, after applying $rho$, into the $\beta$ function and renormalization-group flow.

Extended reading notes

Core claim

The central claim is Theorem 10: the effective-coupling series $g_0=gZ_1Z_3^{-3/2}$ defines a Hopf algebra homomorphism $Phi: H_{CM} -> H$ from the coordinates on formal diffeomorphisms $varphi$ with $varphi(0)=0$, $varphi'(0)=1$ to the Feynman-graph Hopf algebra, and its transpose is a group homomorphism $rho:G_c -> G_2$ into formal diffeomorphisms. The image of $beta=Y Res gamma$ is the usual $\beta$ function. Consequently (Theorem 14) the Birkhoff decomposition $g_{eff}(epsilon)=(g_{eff-})^{-1}(epsilon)g_{eff+}(epsilon)$ of the unrenormalized effective coupling, viewed as a loop of formal diffeomorphisms, gives the bare coupling as $g_{eff-}(epsilon)$ and the renormalized effective coupling as $g_{eff+}(0)$. The proof of the homomorphism is carried out in the massless case, with the massive case deferred to the cited mass-independence of counterterms.

Load-bearing premise

The proof of the Hopf algebra homomorphism is carried out only in the massless case, and its extension to massive theories depends on the asserted mass-independence of minimal-subtracted counterterms, which the paper cites from Collins rather than proves.

Editorial extensions

If this is right

The bare coupling and the renormalized effective coupling can be read off from the two factors in the Birkhoff decomposition of the unrenormalized effective coupling, without first constructing the full character group of Feynman graphs.
The renormalization-group flow is the image under $rho$ of the one-parameter subgroup $F_t=exp(tbeta)$ of the character group, and the beta function is $rho(beta)$.
All higher pole terms of minimal-subtraction counterterms are determined by the residue of the loop through the scattering formula, giving a strong form of the 't Hooft relations.
Renormalization acquires a bundle interpretation: the loop of formal diffeomorphisms is the clutching data for a non-linear complex bundle over the Riemann sphere of the dimensional-regularization parameter, and the Birkhoff decomposition is its trivialization.
The group of characters acts on the space of dimensionless coupling constants by 'diffeographisms', refining ordinary formal diffeomorphisms.

Reading between the lines

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

If the homomorphism is exact, the first beta-function coefficients should be obtainable from the residue of the one-loop character alone; this gives an independent cross-check of multi-loop beta-function calculations.
The bundle picture suggests that changing renormalization scheme corresponds to changing the trivialization of the same bundle, so scheme dependence might be classified by the same clutching data rather than by a separate construction.
The scattering formula has the shape of an operator exponentiation; one could test whether it reproduces known leading-log resummations in massless $varphi^3_6$.

Original abstract

We showed in part I (hep-th/9912092) that the Hopf algebra ${\cal H}$ of Feynman graphs in a given QFT is the algebra of coordinates on a complex infinite dimensional Lie group $G$ and that the renormalized theory is obtained from the unrenormalized one by evaluating at $\ve=0$ the holomorphic part $\gamma_+(\ve)$ of the Riemann-Hilbert decomposition $\gamma_-(\ve)^{-1}\gamma_+(\ve)$ of the loop $\gamma(\ve)\in G$ provided by dimensional regularization. We show in this paper that the group $G$ acts naturally on the complex space $X$ of dimensionless coupling constants of the theory. More precisely, the formula $g_0=gZ_1Z_3^{-3/2}$ for the effective coupling constant, when viewed as a formal power series, does define a Hopf algebra homomorphism between the Hopf algebra of coordinates on the group of formal diffeomorphisms to the Hopf algebra ${\cal H}$. This allows first of all to read off directly, without using the group $G$, the bare coupling constant and the renormalized one from the Riemann-Hilbert decomposition of the unrenormalized effective coupling constant viewed as a loop of formal diffeomorphisms. This shows that renormalization is intimately related with the theory of non-linear complex bundles on the Riemann sphere of the dimensional regularization parameter $\ve$. It also allows to lift both the renormalization group and the $\beta$-function as the asymptotic scaling in the group $G$. This exploits the full power of the Riemann-Hilbert decomposition together with the invariance of $\gamma_-(\ve)$ under a change of unit of mass. This not only gives a conceptual proof of the existence of the renormalization group but also delivers a scattering formula in the group $G$ for the full higher pole structure of minimal subtracted counterterms in terms of the residue.

The compiled record

The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.

Compiled by classifier-1.13-20260917 via OpenRouter, question bank abstract-p1-20260923, 2026-09-23: 1 request, 6,807 input tokens, $0.0003 reported.

What the abstract itself states, as probabilities: claims first 0.05 · states limitation 0.06 · human participants 0.01 · controlled comparison 0.02 · randomized assignment 0.01 · states quantitative result 0.04 · compares with prior results 0.78 · mentions code or data availability 0.01.

The abstract, sentence by sentence

arXiv listing abstract v1. Results and the main claim are underlined; show every role.

We showed in part I (hep-th/9912092) that the Hopf algebra ${\cal H}$ of Feynman graphs in a given QFT is the algebra of coordinates on a complex infinite dimensional Lie group $G$ and that the renormalized theory is obtained from the unrenormalized one by evaluating at $\ve=0$ the holomorphic part $\gamma_+(\ve)$ of the Riemann-Hilbert decomposition $\gamma_-(\ve)^{-1}\gamma_+(\ve)$ of the loop $\gamma(\ve)\in G$ provided by dimensional regularization. We show in this paper that the group $G$ acts naturally on the complex space $X$ of dimensionless coupling constants of the theory. More precisely, the formula $g_0=gZ_1Z_3^{-3/2}$ for the effective coupling constant, when viewed as a formal power series, does define a Hopf algebra homomorphism between the Hopf algebra of coordinates on the group of formal diffeomorphisms to the Hopf algebra ${\cal H}$. This allows first of all to read off directly, without using the group $G$, the bare coupling constant and the renormalized one from the Riemann-Hilbert decomposition of the unrenormalized effective coupling constant viewed as a loop of formal diffeomorphisms. This shows that renormalization is intimately related with the theory of non-linear complex bundles on the Riemann sphere of the dimensional regularization parameter $\ve$. It also allows to lift both the renormalization group and the $\beta$-function as the asymptotic scaling in the group $G$. This exploits the full power of the Riemann-Hilbert decomposition together with the invariance of $\gamma_-(\ve)$ under a change of unit of mass. This not only gives a conceptual proof of the existence of the renormalization group but also delivers a scattering formula in the group $G$ for the full higher pole structure of minimal subtracted counterterms in terms of the residue.

background contribution method result interpretation main claim

What the classifier reports for each sentence. The number is the probability it gave the label.
#	sentence	role	the authors' own work	states the main conclusion	strength
s0	We showed in part I (hep-th/9912092) that the Hopf algebra ${\cal H}$ of Feynman graphs in a given QFT is the algebra of coordinates on a complex infinite dimensional Lie group $G$ and that the renormalized theory is obtained from the unrenormalized one by evaluating at $\ve=0$ the holomorphic part $\gamma_+(\ve)$ of the Riemann-Hilbert decomposition $\gamma_-(\ve)^{-1}\gamma_+(\ve)$ of the loop $\gamma(\ve)\in G$ provided by dimensional regularization.	background 0.99	0.78	0.09	asserted 1.00
s1	We show in this paper that the group $G$ acts naturally on the complex space $X$ of dimensionless coupling constants of the theory.	result 0.63	0.94	0.21	asserted 1.00
s2	More precisely, the formula $g_0=gZ_1Z_3^{-3/2}$ for the effective coupling constant, when viewed as a formal power series, does define a Hopf algebra homomorphism between the Hopf algebra of coordinates on the group of formal diffeomorphisms to the Hopf algebra ${\cal H}$.	result 0.78	0.83	0.15	asserted 0.96
s3	This allows first of all to read off directly, without using the group $G$, the bare coupling constant and the renormalized one from the Riemann-Hilbert decomposition of the unrenormalized effective coupling constant viewed as a loop of formal diffeomorphisms.	result 0.66	0.89	0.17	asserted 0.93
s4	This shows that renormalization is intimately related with the theory of non-linear complex bundles on the Riemann sphere of the dimensional regularization parameter $\ve$.	interpretation 0.97	0.85	0.18	asserted 0.98
s5	It also allows to lift both the renormalization group and the $\beta$-function as the asymptotic scaling in the group $G$.	result 0.85	0.89	0.23	asserted 0.91
s6	This exploits the full power of the Riemann-Hilbert decomposition together with the invariance of $\gamma_-(\ve)$ under a change of unit of mass.	method 0.81	0.79	0.07	asserted 0.97
s7	This not only gives a conceptual proof of the existence of the renormalization group but also delivers a scattering formula in the group $G$ for the full higher pole structure of minimal subtracted counterterms in terms of the residue.	result 0.84	0.92	0.26	asserted 1.00

What kind of paper the classifier reports this is: theoretical result 1.00 · evidence from proof or derivation 1.00.

What was examined, and what was not

sources read

arXiv listing abstract v1

not examined

full text
figures and tables
references and citing sentences
relations to other papers

levels run

P abstract

not run

S sections
T body sentences
C citations
R cross-paper relations

Every quoted sentence on this record is the authors'. Every label is from Pith's closed vocabulary, with the probability shown beside it. Every number in the measurements table is the authors', checked by code. Record as JSON

References

Works cited (9)

[4]

J. Collins, Renormalization , Cambridge University Press (1984)

View on Pith 1984
[1]

A. Connes, D. Kreimer, Comm. Math. Phys. 210 249 (2000), hep-th/9912092

View on Pith Pith review arXiv 2000
[2]

A. Connes, D. Kreimer, Comm. Math. Phys. 199 203 (1998), hep-th/9808042

View on Pith Pith review arXiv 1998
[3]

H. Araki, Ann. Sci. \'Ecole Norm. Sup. (4) 6 67 (1973)

View on Pith 1973
[5]

A. Connes, H. Moscovici, Commun. Math. Phys. 198 199 (1998), math.dg/9806109

View on Pith arXiv 1998
[6]

T. Krajewski, R. Wulkenhaar, Eur.Phys.J. C7 , 697-708

View on Pith
[7]

M. Kontsevich, Lett.Math.Phys

View on Pith
[8]

D. Kreimer, Adv. Theor. Math. Phys. 3.3 (1999), hep-th/9901099

View on Pith Pith review arXiv 1999
[9]

D.J. Broadhurst, D. Kreimer, Phys. Lett. B475 63 (2000), hep-th/9912093

View on Pith Pith review arXiv 2000

Forward citations

Cited by 5 Pith papers

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

Algebraic renormalisation of regularity structures
math.RA 2016-10 accept novelty 8.0 of 10

A new Hopf-algebraic framework implements BPHZ renormalisation for all subcritical stochastic PDEs that fit the regularity-structures setting.
Feynman Polytopes and the Tropical Geometry of UV and IR Divergences
hep-th 2022-02 conditional novelty 7.0 of 10

A single polytope per Feynman graph, cut out by simple linear inequalities, captures both UV and IR divergences and yields their leading coefficients as volumes of dual cones.
Homomorphism, substructure, and ideal: Elementary but rigorous aspects of renormalization group or hierarchical structure of topological orders
hep-th 2025-06 unverdicted novelty 5.0 of 10

An algebraic RG formalism for topological orders uses ideals in fusion rings to encode noninvertible symmetries and condensation rules between anyons.
Some very low-dimensional algebraic topology
math.AT 2024-11 unverdicted novelty 4.0 of 10

Proposes that the group completion of planar configuration spaces, equivalent to ΩS^2, is a moduli space whose Jordan-curve states encode renormalized Feynman integrals as residues.
Feynman Integrals
hep-th 2022-01 accept novelty 2.0 of 10

A self-contained graduate course assembling the modern methods for computing Feynman integrals, from graph polynomials and differential equations through symbols, cluster algebras, elliptic curves and motives.

How the literature uses it

50 arXiv papers cite this in 91 sentences

Each citing sentence is taken from the citing paper's LaTeX source and typed by Pith's classifier. A label appears at probability 0.8 or more (0.95 for support), a contest at 0.7 or more; the quoted words are the citing authors'. How each label was checked · JSON

Papers citing it · 59 sentences without a confident label

Builds on it 2 papers

Section three introduces the Hopf algebra of Feynman graphs as developed by Connes and Kreimer in [ref] and shows that the geometrization of BPHZ renormalization developed by the same in [ref] can be extended to a curved background space-time.

The geometric $\beta$-function in curved space-time under operator regularization · Introduction · 0.95
In section [ref], I generalize the $β$ function defined in [ref] .

Dynkin operators, renormalization and the geometric $\beta$ function · Introduction · 0.96

Uses its method or tool 8 papers

In the rest of the paper, I apply the methods of [ref] , [ref] , and [ref] to build a Hopf algebra of Feynman diagrams, define the counterterms using Birkhoff decomposition, and define the $β$ function for cut-off regularization on the corresponding renormalization bundle.

Geometrically relating momentum cut-off and dimensional regularization · BPHZ on momentum cut-off regularization · 0.90
We will closely follow the presentation of [ref] .

Analytical and Geometric approches of non-perturbative Quantum Field Theories · Renormalization group equation · 0.83
Connes and himself in the 1990s and 2000s [ref] .

Algebraic Structures in the Coupling of Gravity to Gauge Theories · Introduction · 0.92

Compares with it 2 papers

For further study it will be interesting to compare the above observations with the similar relations occurring among counterterms in the minimal subtraction scheme as reported in [ref] , or the differential equation for the physicial limit in this scheme derived in [ref] .

Hopf algebraic Renormalization of Kreimer's toy model · The toy model of iterated insertions · 0.98
In a certain sense, it is a Wilsonian analogue of the morphism from the group of graphs into the diffeomorphisms of the coupling constant presented in [ref] .

Wilsonian renormalization, differential equations and Hopf algebras · Hopf algebra of Feynman diagrams and effective actions · 0.94

Cites it as support 1 paper

These are also motivated by the natural examples [ref] .

Pathlike Co/Bialgebras and their Antipodes with Applications to Bi- and Hopf Algebras Appearing in Topology, Number Theory and Physics · Quotients and quantum deformations for grouplike central and invertible characters · 0.96

Cites it as background 18 papers

In [ref] , the authors define a $β$ function, a Lie algebra element representing how a dimensionally regularized QFT depends on the energy scale.

Dynkin operators, renormalization and the geometric $\beta$ function · Introduction · 0.90
Recent literature has emerged geometrically describing this process of renormalization and regularization for a QFT in which the perturbative $β$ function for dimensional regularization is defined by a connection on a renormalization bundle in which different regularization schemes correspond to sections [ref] .

Geometrically relating momentum cut-off and dimensional regularization · Introduction · 0.93
To find the derivation of the geometric $β$-function in the flat geometric context, see [ref] , [ref] or [ref] .

The geometric $\beta$-function in curved space-time under operator regularization · As a geometric object · 0.93

Discussion

Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.

[1] [4]

J. Collins, Renormalization , Cambridge University Press (1984)

View on Pith 1984

[2] [1]

A. Connes, D. Kreimer, Comm. Math. Phys. 210 249 (2000), hep-th/9912092

View on Pith Pith review arXiv 2000

[3] [2]

A. Connes, D. Kreimer, Comm. Math. Phys. 199 203 (1998), hep-th/9808042

View on Pith Pith review arXiv 1998

[4] [3]

H. Araki, Ann. Sci. \'Ecole Norm. Sup. (4) 6 67 (1973)

View on Pith 1973

[5] [5]

A. Connes, H. Moscovici, Commun. Math. Phys. 198 199 (1998), math.dg/9806109

View on Pith arXiv 1998

[6] [6]

T. Krajewski, R. Wulkenhaar, Eur.Phys.J. C7 , 697-708

View on Pith

[7] [7]

M. Kontsevich, Lett.Math.Phys

View on Pith

[8] [8]

D. Kreimer, Adv. Theor. Math. Phys. 3.3 (1999), hep-th/9901099

View on Pith Pith review arXiv 1999

[9] [9]

D.J. Broadhurst, D. Kreimer, Phys. Lett. B475 63 (2000), hep-th/9912093

View on Pith Pith review arXiv 2000