{"id":"1757d963-8a07-4169-98b0-3907b38849fc","arxiv_id":"2501.11115","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":5.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"A dressed-site rishon formalism enables bosonic, gauge-invariant tensor network simulations of SU(2) lattice gauge theory in two dimensions, yielding finite-density phase diagrams and quantum many-body scarring dynamics.","lead":"This PhD thesis develops tensor network methods for Hamiltonian lattice gauge theories, centered on a dressed-site formalism that builds gauge invariance into each lattice site. It applies the approach to SU(2) Yang-Mills in 2D, reporting first tensor network simulations, a finite-density phase diagram, and non-ergodic scarring dynamics.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The reported phase diagram and scar dynamics rest on the jmax=1/2 hardcore-gluon truncation, whose validity outside the strong-coupling regime is asserted but never quantified; at the magneto-electric crossover the results may describe the truncated model, not SU(2) Yang-Mills.","rationale":"The reader's weakest assumption is the same as mine: the hardcore-gluon jmax=1/2 truncation is assumed to faithfully capture the low-energy physics in the regimes where the phase diagram and scar dynamics are computed. I agree with the conditional verdict. The central argument has independent support: the dressed-site formalism has appeared in peer-reviewed venues, and the ed-lgt code is released, so the exact mapping for a fixed truncation is not the main risk. However, the abstract's strongest claim, 'first TN simulations... revealing critical aspects of its phase diagram and non-equilibrium behavior, such as a QMB scarring dynamics,' is not protected by a truncation-error analysis. The thesis itself states in Sec. 1.3.2 and Sec. 1.3.5 that jmax=1/2 is a strong-coupling approximation and that weak-coupling physics requires larger representations. Since a magneto-electric transition is by definition in the crossover between electric- and magnetic-dominated regimes, invoking the hardcore truncation at that transition without a jmax study is a real soft spot. This is a specific, addressable concern, not an internal inconsistency, and it does not change the reader's verdict from CONDITIONAL.","tokens_in":84078,"tokens_out":6208,"duration_ms":64074,"concrete_test":"Compute, for a single SU(2) plaquette with open boundary conditions, the ground-state expectation of the magnetic plaquette term Tr U_□ for jmax=1/2, 1, and 3/2 on the same coupling grid used in Ch. 3, and determine the minimal truncation level ℓ* at tolerance ϵ=10^-5 as defined in Sec. 1.4.5. If ℓ* > 2 at any coupling where the phase diagram is drawn, the hardcore-gluon results are not converged and the phase boundaries cannot be attributed to the untruncated SU(2) theory without further extrapolation.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The most load-bearing step is the unvalidated use of the hardcore-gluon truncation, jmax=1/2, for the physics claimed in the abstract. Sec. 1.3.5 defines this as the smallest non-trivial SU(2) link truncation and explicitly limits its reliability to g >> 1; Sec. 1.3.2 states that weak-coupling continuum physics requires larger representations. Yet Ch. 3 reports a magneto-electric transition and a phase diagram as properties of (2+1)D SU(2) Yang-Mills, and Ch. 4 reports quantum many-body scarring in the same truncated model. The magnetic plaquette term is suppressed by 1/g^2, so a magneto-electric transition occurs where the magnetic term is not small; that is precisely where jmax=1/2 is not justified. No convergence check in jmax is reported. The exactness of the dressed-site mapping for a fixed truncation does not remedy this: the physical question is whether the truncation captures the relevant electric and magnetic fluctuations. If the transition coupling is near g ~ O(1) or below, higher j=1,3/2 shells will contribute and the reported boundaries and scar revivals could be artifacts of the hardcore model. The thesis also leaves the rishon decomposition (Eqs. 1.3.23-1.3.31) unproved, but that is a secondary, more tractable correctness check; the truncation question is the one that determines whether the headline physics is SU(2) Yang-Mills.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"This manuscript, a PhD thesis posted on arXiv, develops a dressed-site formalism for Hamiltonian lattice gauge theories in which gauge links are truncated by an energy cutoff, decomposed into fermionic rishon modes, and fused with matter sites into gauge-invariant dressed sites with bosonic statistics. The formalism is applied to SU(2) Yang-Mills with staggered matter in two spatial dimensions, where the author reports ground-state phase diagrams, baryonic spectra, a baryon-liquid phase, topological observables, and, in a one-dimensional truncated model, quantum many-body scarring dynamics. The thesis also presents a fermion-to-qubit mapping for general lattice fermion theories, an analysis of space-filling curves for tensor-network locality, and a roadmap for high-performance tensor-network simulations of lattice gauge theories.","tokens_in":84434,"tokens_out":3146,"duration_ms":35079,"significance":"If the central formal claim is correct, the dressed-site construction provides an exact, gauge-invariant, bosonic encoding of truncated non-Abelian gauge theories, and the reported (2+1)D SU(2) simulations would be a genuinely new tensor-network application. The manuscript is also useful as a systematic review of tensor-network methods for lattice gauge theories, and the accompanying ED-LGT code and the quantum-simulation oriented qudit formulation are concrete contributions that go beyond a purely pedagogical treatment. However, the significance of the headline physical results—phase diagram and many-body scarring—depends on two points that are not adequately established in the manuscript: the exactness of the rishon decomposition of the parallel transporter for arbitrary truncation, and the quantitative validity of the hardcore-gluon jmax=1/2 truncation in the regime where the reported transitions and dynamics occur. The paper should be credited for including numerical evidence of truncation convergence for a single QED plaquette, but that evidence is not carried over to the SU(2) calculations.","major_comments":[{"comment":"The central formal step—the rishon decomposition of the truncated SU(2) parallel transporter—is asserted rather than proved. After Eq. (1.3.27) the text says “It is possible to show that this construction is indeed compatible with the explicit form of the parallel transport reported in Eq. (1.3.10),” but no proof or explicit algebraic verification is given. Since all subsequent dressed-site operators and all numerical results in Chapters 3 and 4 inherit this equivalence, this is load-bearing. The author should either provide a complete derivation, or a reproducible symbolic/numerical verification that the right-hand side of Eq. (1.3.23) equals the Clebsch-Gordan matrix elements of Eq. (1.3.10) for all allowed j and for generic jmax.","section":"Sec. 1.3.3, Eqs. (1.3.23)–(1.3.31)"},{"comment":"The ‘operative defermionized Hamiltonian’ is written down without a complete step-by-step derivation. In particular, the passage from the rishon form of the hopping and plaquette terms to the projected dressed-site operators uses the projection Oeff = M†OM of Eq. (1.2.5), but the text does not show how the 5×5 or 30×30 dressed-site matrices are obtained, what the explicit coefficients of the corner operators are, or how the Gauss-law kernel M is computed in practice. This is not merely a presentation issue, because the correctness of Eq. (1.3.45) is the basis for every reported numerical result. The author should add a derivation or an appendix with the operator construction, and should state explicitly which results are independently reproducible from the released ed-lgt code.","section":"Sec. 1.3.4, Eq. (1.3.45)"},{"comment":"The hardcore-gluon truncation jmax=1/2 is described in Sec. 1.3.5 as a good approximation only in the strong-coupling limit g >> 1, and Sec. 1.3.2 states that weak-coupling continuum physics requires larger representations. Yet Chapter 3 reports a magneto-electric transition and a phase diagram for (2+1)D SU(2) Yang-Mills. The magneto-electric crossover occurs where the magnetic plaquette term, suppressed by 1/g^2, balances the electric term; by the author’s own criterion this is precisely the regime where jmax=1/2 is least justified. No convergence check in jmax is reported for the 2D equilibrium results, and the single-plaquette convergence study of Fig. 1.3 is performed for U(1), not SU(2). The author should either (i) identify the coupling range of the reported transition and demonstrate that jmax=1/2 is reliable there, or (ii) explicitly rephrase the Chapter 3 results as properties of the truncated hardcore-gluon model rather than of SU(2) Yang-Mills.","section":"Sec. 1.3.5 and Ch. 3, esp. Sec. 3.2 and 3.7"},{"comment":"The text uses the QED plaquette convergence result ℓ* ~ g^-1 to motivate the statement that “an analogous inverse dependence of the minimal gauge truncation on the coupling is expected for non-Abelian LGT in arbitrary dimensions.” This expectation is not demonstrated, and it is invoked in discussing the need for truncation compression. The author should either supply a corresponding single-plaquette or small-lattice convergence study for SU(2), or clearly label this statement as an unsupported conjecture. Since the abstract claims the first TN simulations of the 2D SU(2) system, this missing truncation benchmark is directly relevant to whether the reported physics is the physics of the full gauge theory.","section":"Sec. 1.4.5, Fig. 1.3"}],"minor_comments":[{"comment":"The manuscript contains numerous typographical and notational infelicities, including “Cliffor’s algebra” (Sec. 1.1.4), the anticommutator sign in Eq. (1.3.4), duplicated figure labels (Fig. 1.1 and Fig. 1.2), and inconsistent placement of subscripts such as ψˆ†n,α vs ψˆ†n,α. A careful proofreading pass would substantially improve readability.","section":"Throughout"},{"comment":"The definition of the rishon operator ζˆg(r) is hard to parse: the lower limit of the sum is written as “jmax− 1/2” and the index m− in the ket ⟨j+1/2, m−+1/2| is not defined. This should be restated with explicit bounds and a clear explanation of the truncated Hilbert space to allow the reader to verify Eq. (1.3.27).","section":"Sec. 1.3.3, Eq. (1.3.27)"},{"comment":"The abstract’s claim of “first TN simulations” relies on the author’s own publication [2]. A short review of prior tensor-network or other Hamiltonian approaches to (1+1)D and (2+1)D non-Abelian gauge theories would help place this claim in context and distinguish a first in a specific truncation scheme from a first for the full model.","section":"Sec. 3.1 and Abstract"},{"comment":"In the 1D qudit Hamiltonian Eq. (1.3.59), the operator Mˆ n appears in the mass term but was not explicitly defined in the preceding equations; Eq. (1.3.57c) defines Nˆ n, and the text later uses Mˆ n. The author should define Mˆ n explicitly or replace it by Nˆ n for consistency.","section":"Sec. 1.3.5, Eqs. (1.3.57a)–(1.3.57d)"}],"recommendation":"major_revision","confidential_remarks":"This manuscript is a PhD thesis assembled from several already-published papers ([1]–[5]). The referee report focuses on the parts that are presented as new or load-bearing. The most serious issue is the gap between the formal dressed-site construction and the physical claims: the exactness of the rishon decomposition is asserted, the projected Hamiltonian is not fully derived, and the truncation whose validity is most questionable is precisely the one used for the headline phase diagram and scarring results. These points are fixable within the scope of the manuscript—by adding proofs, appendices, and at least one SU(2) jmax-convergence check—so I do not recommend rejection, but the manuscript is not ready in its present form. I would also suggest the editor ask the author to state clearly which parts of the numerical results are reproducible from the released code and which parts were already reported in the cited publications."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Giovanni, quick take on arXiv:2501.11115. It's a PhD thesis that compiles the author's own published work, mostly already in peer-reviewed venues (PRR, Quantum). What is actually new here is the exposition: a self-contained derivation of the dressed-site formalism for SU(2) and U(1), with explicit bases for the hardcore-gluon model, plus a chapter on fermion-to-qubit mapping and realistic discussion of Hilbert curve ordering for 2D tensor networks. The thesis is honest about being a compilation, and it ships code (ed-lgt, Zenodo DOI) that reproduces the dressed-site operators. That is real.\n\nThe central physics claims are the first tensor-network simulations of 2D SU(2) Yang-Mills at finite density and quantum many-body scarring in a 1D truncated SU(2) model. If those results are right, they are significant: they extend TN methods to non-Abelian gauge theories beyond 1D, in regimes Monte Carlo can't reach. The simulations themselves are numerical and not independently reproducible from the arXiv artifact, but the published papers and the released code go a long way.\n\nThe soft spots are two. The rishon decomposition of the SU(2) parallel transporter is asserted with 'it is possible to show' (Sec. 1.3.3, around Eqs. 1.3.23-1.3.31), and the projected dressed-site Hamiltonian (Eq. 1.3.45) lacks a full derivation. That's a tractable gap: either prove it in an appendix or cite where the supplementary derivation is available. It didn't stop me from following the construction, but a referee should insist.\n\nThe more serious issue is the use of the jmax=1/2 hardcore-gluon truncation for the phase diagram and scar claims. In Sec. 1.3.5 the thesis itself limits this truncation to g >> 1, where the electric term dominates. The magneto-electric transition in Chapter 3 happens near where the magnetic term becomes important, i.e. not clearly in the strong-coupling regime, and the scarring in Chapter 4 is in the same truncated model. No convergence check in jmax is provided for these observables. The U(1) truncation analysis in Sec. 1.4.5 shows similar truncations converge poorly at small g, so the risk is real: the reported boundaries and scar revivals may be properties of the hardcore-gluon model, not of SU(2) Yang-Mills. The exactness of the dressed-site mapping for a fixed truncation does not address this, and the stress-test note is right on this point.\n\nWho is this for? People entering Hamiltonian LGT simulations, tensor network experts looking for a gauged reference, and quantum simulation groups benchmarking qudit encodings. It does a good pedagogical job.\n\nMy recommendation: yes, send it to peer review, but with a referee pitch that requires (1) a complete proof of the rishon decomposition, and (2) truncation-error quantification for the phase diagram, e.g. jmax=1 comparisons or strong-coupling perturbation theory. If those are added, the thesis becomes a useful reference. As is, it's a promising but incomplete presentation of important results.","headline":"A well-organized thesis compiling the author's own significant tensor-network results for SU(2) lattice gauge theories, but the headline phase diagram and scarring claims rest on a truncation whose validity is asserted, not demonstrated, and the rishon decomposition is left unproven.","tokens_in":84944,"tokens_out":3129,"would_cite":false,"duration_ms":31496,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"A dressed-site formalism enables the first tensor-network simulations of two-dimensional SU(2) Yang-Mills lattice gauge theory, with an exact bosonic gauge-invariant encoding.","keywords":["lattice gauge theory","tensor networks","dressed-site formalism","SU(2) Yang-Mills","quantum many-body scars","fermion-to-qubit mapping","Hilbert curve","Hamiltonian simulation"],"falsifier":"Repeat the reported two-dimensional ground-state phase diagram and the one-dimensional scar-revival calculations with link truncations $j_{\\max}=1$, $3/2$, and higher on the same lattice sizes; if the phase-boundary locations and the revival fidelity of the scarred states change substantially or disappear as $j_{\\max}$ is increased, the claimed signatures are artifacts of the truncation rather than properties of full SU(2) Yang-Mills.","tokens_in":83834,"feed_emoji":"⚛️","tokens_out":6419,"duration_ms":63986,"temperature":0.7,"pith_summary":"This thesis claims that the dressed-site formalism, which fuses each lattice site with the neighboring truncated gauge-link degrees of freedom, gives an exact, bosonic, gauge-invariant reformulation of Hamiltonian lattice gauge theories with controllable truncation. Applied to SU(2) Yang-Mills with dynamical staggered fermions, it makes two-dimensional tensor-network simulations possible for the first time, producing a phase diagram at zero and finite baryon density plus a first observation of quantum many-body scarring in a non-Abelian lattice gauge theory. The broader claim is that tensor networks can access regimes, real-time dynamics and finite density, where Monte Carlo methods are blocked by the sign problem. If correct, the formalism provides a practical route from Abelian toy models toward non-Abelian theories relevant for QCD, and a benchmark target for quantum simulators.","feed_headline":"Dressed sites bring 2D SU(2) lattice gauge theory to tensor networks","feed_subtitle":"Exact gauge-invariant truncation yields a phase diagram and many-body scar dynamics beyond Monte Carlo reach","key_machinery":"The central object is the dressed site: a composite degree of freedom formed by fusing a staggered-fermion matter site with the rishon modes of all attached half-links. Each truncated gauge link is split into two rishons; the parallel transporter becomes a rishon bilinear, and the requirement that the two sides of a link sit in the same irreducible representation becomes an Abelian $\\mathbb Z_2$ link symmetry. Gauss law is then a purely internal constraint, and the effective Hamiltonian is obtained by projecting onto its kernel, yielding local, bosonic operators.","core_discovery":"Within the hardcore-gluon truncation ($j_{\\max}=1/2$) of SU(2) Yang-Mills, the parallel transporter on each link is decomposed into two fermionic rishon modes, one per half-link; the rishons are then absorbed into the adjacent matter site, and Gauss' law is imposed exactly by restricting to the kernel of the gauge generators. The resulting dressed-site Hamiltonian is made entirely of bosonic operators acting on a 30-dimensional local basis in two spatial dimensions, so fermionic statistics and gauge constraints no longer need to be enforced dynamically. Using this representation, the thesis reports the first tensor-network ground-state and time-evolution simulations of two-dimensional SU(2) Yang-Mills lattice gauge theory, including a magneto-electric crossover, baryonic spectrum, a finite-density baryon-liquid phase, and topological sectors, together with quantum many-body scarring dynamics in the one-dimensional truncation.","pith_inferences":["If the exactness of the dressed-site mapping holds at every truncation level, the same construction should extend to SU(3) Yang-Mills; the practical obstacle would be the much larger local Hilbert space rather than gauge invariance.","The thesis reports scar signatures at higher link truncations but not extrapolated to the continuum; a direct test is whether revivals and the scar tower survive as $j_{\\max}$ grows toward the weak-coupling limit.","The fermion-to-qubit mapping developed for general lattice fermion theories could be applied to other fermionic condensed-matter models, where it may reduce the qubit overhead of digital quantum simulation beyond the Hubbard example studied here.","The Hilbert-curve ordering result and the dressed-site formalism are developed in parallel; combining them systematically in two-dimensional lattice gauge theory simulations is a natural next step that the thesis does not itself carry out."],"forward_implications":["Gauss law is satisfied by construction, so no large penalty terms are needed to keep the simulation in the physical gauge-invariant sector.","Because every term in the effective Hamiltonian is bosonic, tensor-network algorithms avoid both the Monte Carlo sign problem and long-range fermion-to-qubit encodings.","The compact local dimensions, 30 per site in the two-dimensional hardcore-gluon case, make exact diagonalization and moderate-bond tensor networks feasible, while the dressed-site dimension grows rapidly with truncation level.","The reported finite-density phase diagram and non-equilibrium scar dynamics are concrete observables that can serve as benchmarks for future quantum simulations of non-Abelian gauge theories."],"supporting_citations":[{"why":"Supplies the first tensor-network simulations of two-dimensional SU(2) Yang-Mills with the dressed-site formalism, including the phase diagram at zero and finite baryon density.","marker":"[2]"},{"why":"Reports the quantum many-body scarring dynamics in truncated SU(2) Yang-Mills and the corresponding tower of scar states.","marker":"[3]"},{"why":"Generalizes the dressed-site idea to a fermion-to-qubit mapping for lattice fermion theories without gauge symmetry.","marker":"[4]"},{"why":"Gives the roadmap, local-dimension scaling, and truncation convergence analysis for U(1) and SU(2) gauge theories.","marker":"[5]"},{"why":"Provides the exact diagonalization code used to construct and benchmark dressed-site Hamiltonians and gauge-invariant bases.","marker":"[6]"},{"why":"Introduces the electric-energy Casimir truncation and dressed-site construction on which the formalism builds.","marker":"[155]"},{"why":"Underlies the dressed-site gauge-invariant construction by which matter and link degrees of freedom are fused.","marker":"[158]"}],"fun_headline_variants":["First tensor network run of 2D SU(2) lattice gauge theory","Tensor networks crack 2D SU(2) lattice gauge theory","Beyond Monte Carlo: tensor networks for 2D SU(2)","Exact gauge truncation enables first 2D SU(2) tensor network","Many-body scars emerge in 2D SU(2) tensor-network simulation"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the minimal hardcore-gluon truncation, keeping only the $j=0$ and $j=1/2$ representations of the SU(2) link field, faithfully captures the low-energy physics in the regimes where the phase diagram and scarring dynamics are computed, although the thesis states this truncation is reliable mainly for strong coupling $g\\gg1$.","fun_headline_variants_meta":{"raw":{"variants":["First tensor network run of 2D SU(2) lattice gauge theory","Tensor networks crack 2D SU(2) lattice gauge theory","Beyond Monte Carlo: tensor networks for 2D SU(2)","Exact gauge truncation enables first 2D SU(2) tensor network","Many-body scars emerge in 2D SU(2) tensor-network simulation"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001893,"raw_usage":{"total_tokens":7411,"prompt_tokens":924,"completion_tokens":6487,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":540,"completion_tokens_details":{"reasoning_tokens":6389}},"tokens_in":540,"tokens_out":6487,"duration_ms":41443,"temperature":1.0,"reasoning_tokens":6389,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T18:37:16.646543+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Repeat the reported two-dimensional ground-state phase diagram and the one-dimensional scar-revival calculations with link truncations $j_{\\max}=1$, $3/2$, and higher on the same lattice sizes; if the phase-boundary locations and the revival fidelity of the scarred states change substantially or disappear as $j_{\\max}$ is increased, the claimed signatures are artifacts of the truncation rather than properties of full SU(2) Yang-Mills.","supporting_citations":[{"cited_title":"Digital Quantum Simulation of Lattice Fermion Theories with Local Encoding","cited_arxiv_id":null,"evidence_quote":"Generalizes the dressed-site idea to a fermion-to-qubit mapping for lattice fermion theories without gauge symmetry."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the exact diagonalization code used to construct and benchmark dressed-site Hamiltonians and gauge-invariant bases."},{"cited_title":"Lattice Gauge Tensor Networks","cited_arxiv_id":null,"evidence_quote":"Underlies the dressed-site gauge-invariant construction by which matter and link degrees of freedom are fused."}],"review_version":1}