{"id":"39f58272-fc68-42ed-8e48-1bf2f33ad041","arxiv_id":"2412.07323","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In the two-dimensional extended Hubbard model, the charge susceptibility obeys an RPA-like formula built from the U'=0 polarization, because the density fermion-boson coupling is nearly independent of U'.","lead":"This paper extends a powerful diagrammatic method, single-boson exchange, to the extended Hubbard model where electrons repel on the same site and between neighboring sites. The authors find that the charge response follows a simple RPA-like formula, and trace this to cancellations in the renormalized charge coupling.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (45) requires the full polarization P^D to be U'-independent, but the paper only documents U'-independence of λD; the bubble ΠD inherits U'-dependence through the self-energy and chemical-potential shift, so the RPA-like claim rests on an unverified premise.","rationale":"The reader's weakest assumption targets the limited parameter evidence for λ^D independence. My stress test goes one step further: even where λ^D is independent, the polarization P^D can depend on U' through the bubble, and the paper neither checks this nor acknowledges it. This makes the central claim under-derived rather than wrong; the proposed numerical check is cheap and decisive. The paper's methodological contribution — the SBE extension to nonlocal interactions, the form-factor reduction, and the systematic approximation tests — is solid and independently valuable, so a conditional acceptance with the P^D check as a condition remains the appropriate verdict. I agree with the reader's conditional stance but identify a different, more direct missing verification: the U'-independence of P^D itself, not just of λ^D.","tokens_in":28710,"tokens_out":8490,"duration_ms":86205,"concrete_test":"From the fRG output used for Figs. 9-10, compute P^D(Ω=0, q=(π,π)) via Eq. (35) for U' = 0, 0.1, 0.2, 0.25, 0.27 at U=2, β=10, and cross-check with P^D = χ^D/(1+B^Dχ^D). If P^D varies by more than ~10% across this range, Eq. (45) loses quantitative support. To isolate the bubble's role, repeat the U'>0 flow with the self-energy fixed to its U'=0 value; a significant change in χ^D would confirm that Π^D, not just λ^D, responds to U'.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central physical claim, Eq. (45), states χD ≈ P^D_{U'=0}/(1+P^D_{U'=0}B^D). For this to hold, the polarization P^D(Ω,q) = Σ_{ν,m} λ^D(Ω,q,ν,m) Π^D(Ω,q,ν,m) must be essentially independent of U'. The paper verifies only that the fermion-boson coupling λ^D is almost U'-independent (Figs. 9, 12), and then asserts that this independence 'translates to the polarization P^D' (Section 4.2). This is a non-sequitur: Π^D is built from the renormalized propagator G containing the self-energy, and at U'>0 the particle-hole symmetry is broken, so the chemical potential must be shifted to fix the filling (Section 3.3). Both Σ and Π^D therefore acquire U'-dependence that is never plotted or quantified. The χ^D curves alone do not discriminate between a constant P^D and a mildly varying one, because the divergence is dominated by B^D(π,π) = U - 8U'. The fluctuation diagnostics do not close the gap: at 1ℓ the post-processed λ^D has a slight U'-dependence and becomes constant only after adding 2ℓ corrections (Fig. 15), while the susceptibilities in the claim are 1ℓ and neglect the rest-function flow, with documented ~8% deviations in χ^D at T=0.1 near the divergence (Fig. 6). The quantitative support for Eq. (45) is therefore weaker than the narrative suggests.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript generalizes the single-boson exchange (SBE) formulation of the functional renormalization group to the extended Hubbard model with a nearest-neighbor interaction U'. The central methodological step is a modified notion of bare-interaction reducibility, splitting the bare interaction into a bosonic part B^X(q) and a fermionic part F^X(k,k'), which preserves the multiplicative form of the single-boson exchange while avoiding the numerically costly form-factor sums of a naive extension. The authors then establish a simplified computational scheme that keeps only an s-wave form factor, neglects the flow of the rest function, and omits non-trivial high-frequency asymptotics, and they test this scheme against fuller calculations at half filling and at van Hove filling for U=2. Their main physical claim is that, although the charge susceptibility chi^D is strongly enhanced by U' and eventually diverges near 4U'=U, the density fermion-boson coupling lambda^D is almost independent of U'. From this they conclude, via Eq. (45), that chi^D is approximately RPA-like with respect to the polarization of the U'=0 Hubbard model, and they trace the insensitivity of lambda^D to a cancellation between magnetic, density, and superconducting fluctuation channels using fluctuation diagnostics and a sign-based argument.","tokens_in":29127,"tokens_out":5220,"duration_ms":57441,"significance":"The methodological contribution is potentially valuable: a B-reducibility-based SBE scheme that retains the computational advantages of the local SBE formalism while accommodating nonlocal interactions. The conceptual claim, Eq. (45), is also significant if it survives scrutiny: it states that the nonlocal interaction enters the charge response only at the level of an RPA denominator, while all vertex corrections are encoded in a U'-independent polarization. This is a falsifiable and physically transparent prediction, and the authors are careful to distinguish it from bare RPA by noting that P^D contains vertex corrections. The paper reports systematic tests of several approximations (mixed bubbles, form-factor truncation, rest-function flow, high-frequency asymptotics) and includes fluctuation diagnostics with 2-loop corrections, with no fitted parameters. These are real strengths. However, the quantitative evidence for Eq. (45) is currently thinner than the narrative suggests: the U'-independence is demonstrated for lambda^D at a single value U=2, and the step from lambda^D independence to polarization independence is asserted rather than verified.","major_comments":[{"comment":"The statement in §4.2 that the U'-independence of lambda^D 'translates to the polarization P^D' is not demonstrated. By Eq. (35), P^D is a sum over lambda^D times the bubble Π^D, and Π^D is built from fully renormalized propagators that acquire U'-dependence through the self-energy and through the chemical-potential shift introduced in §3.3 (Eqs. (25)-(27)); at finite U' the particle-hole symmetry is broken, so δμ changes with U'. Neither the U'-dependence of Σ nor that of Π^D is plotted or quantified. Since Eq. (45) is the central physical claim, I ask the authors to show P^D(Ω=0,q) directly as a function of U' (or at least the ratio P^D(U')/P^D(0)), and to compare the right-hand side of Eq. (45) with the numerically computed chi^D from the same fRG flow, reporting residuals as functions of U' and temperature.","section":"§4.2, Eqs. (35) and (45)"},{"comment":"The cancellation mechanism is argued for lambda^D, not for the polarization, and the numerical support is limited to a narrow parameter window. The 1-loop post-processed lambda^D in Fig. 14 shows a slight U'-dependence and only becomes approximately constant after adding 2-loop corrections in Fig. 15, while the susceptibilities used for Eq. (45) are obtained in the 1-loop scheme with the rest-function flow neglected; Fig. 6 documents a relative deviation of about 8% in chi^D at T=0.1 near the divergence. The sign-based diagnostic matrices in Eqs. (47)-(56) give only signs, not magnitudes, as the authors themselves note. Please state whether Eq. (45) is intended to hold at 1-loop or at the converged (multiloop) level, and test it directly in both cases rather than inferring it from lambda^D alone.","section":"§4.3, Figs. 14 and 15"},{"comment":"The abstract claims that the flow of the rest function can be neglected 'up to moderate interaction strengths', but the quantitative evidence is limited to U=2 at half filling and at one van Hove filling, with temperatures down to T=0.1. To make the scope claim load-bearing, the authors should either add a second interaction strength (e.g. U=4) or provide an analytic argument delimiting the regime in which the U'-independence of P^D and the neglect of the rest function remain valid. The qualitative sentence in §5 that at larger couplings and lower temperatures the rest function should be included currently defines the boundary only in words.","section":"Abstract and §5"}],"minor_comments":[{"comment":"The caption says the superconducting (red) and density (green) contributions cancel the magnetic (red) one, but the color labeling appears inconsistent: the magnetic contribution should presumably not share the color 'red' with the superconducting one, and the density contribution is described as green.","section":"Fig. 14 caption"},{"comment":"The text says the analysis in §4.1 uses beta=5 unless otherwise stated, while Fig. 5 is described as T=0.2; please make the notation between beta and T consistent in the captions and the text.","section":"Section 4.1"},{"comment":"There are several typographical and formatting issues, including 'F unding information' in the acknowledgments, 'na ¨ ıve' and 'responsible of'; a careful proofread would improve the presentation.","section":"Throughout"},{"comment":"The derivation of Eq. (55) is compressed; since this sign rule plays a supporting role in the central cancellation argument, a short derivation or a reference to the diagrammatic enumeration in Fig. 16 would help the reader verify the signs without reverse-engineering them.","section":"Eq. (55)"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is within the scope of SciPost Physics and the SBE generalization is a useful step, but the central RPA-like claim rests on an unverified transfer of a lambda^D property to the polarization P^D. I would ask the editor to require the authors to provide direct evidence for the U'-independence of P^D and for Eq. (45) itself, rather than only lambda^D. The paper also builds heavily on the authors' own SBE-fRG framework, including a thesis (Ref. [20]) and a companion paper (Ref. [88]); the novelty relative to those works should be made explicit in the final version."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Two things to know. The B-reducibility splitting is a real methodological step: it keeps the single-boson factorization for nonlocal U' by absorbing the bosonic part of the bare interaction into w^D, and the numerical shortcuts built on it (s-wave form factor only, negligible rest-function flow, trivial high-frequency asymptotics) are tested carefully, with deviations quantified. That part is solid and useful. The second thing is the physical claim: chi^D is RPA-like with respect to the U'=0 polarization. The story is plausible and the cancellation mechanism via fluctuation diagnostics is a nice insight, but the paper does not directly verify the load-bearing premise.\n\nThe central equation (45) needs P^D to be essentially U'-independent. What the paper shows is that the fermion-boson coupling lambda^D is nearly U'-independent (Figs. 9, 12), and then asserts this translates to P^D. It doesn't plot P^D as a function of U', and it doesn't quantify the U'-dependence of Pi^D via the self-energy and chemical-potential shift. The footnote acknowledging that self-energy corrections are absent in RPA is honest but undersells the point: those corrections can enter P^D. The chi^D curves alone don't discriminate, because the divergence is dominated by B^D = U - 8U'. So Eq. (45) is a reasonable conjecture that is consistent with the numerics, not a demonstrated identity.\n\nOther soft spots are minor. Tests are at U=2 only; the claim about moderate interaction strengths is not established beyond that. The cancellation argument is a sign-based poor-man's heuristic, though it agrees with the full fluctuation diagnostics and the 2-loop post-processing recovers the constant lambda^D. The rest-function neglect produces up to 8% deviations near the divergence at T=0.1, so the claim is approximate precisely where it matters most. No code or data is released, which is common but would help here. The heavy self-citation is present but mostly covers the SBE framework this paper extends; not a real problem.\n\nWho is this for: people using fRG for models with nonlocal interactions, and anyone interested in when RPA-like formulas hold beyond bare RPA. It deserves a serious referee. The referee should ask for a direct test of Eq. (45): plot P^D as a function of U' and compare chi^D from Eq. (45) against the full fRG result across U', separating the self-energy contribution to the bubble's U'-dependence. That would turn a plausible explanation into a verified one.","headline":"A solid and genuinely useful extension of SBE fRG to nonlocal interactions, but the RPA-like claim for the charge susceptibility rests on an unverified premise about the U'-independence of the polarization.","tokens_in":29588,"tokens_out":3132,"would_cite":true,"duration_ms":31286,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper claims that the charge susceptibility of the extended Hubbard model is described by an RPA-like formula built from the U'=0 polarization, and traces this to a cancellation that leaves the density fermion-boson coupling almost…","keywords":["extended Hubbard model","single-boson exchange","functional renormalization group","charge susceptibility","random phase approximation","fermion-boson coupling","fluctuation diagnostics","charge-density wave"],"falsifier":"Run the same fRG calculation at stronger coupling and lower temperature (for example $U=4$, $\\beta=20$, $U'$ near the density-wave boundary) while keeping the full rest-function flow and d-wave form factors, and compare $\\chi^D$ with $P^D_{U'=0}/(1+P^D_{U'=0}B^D)$; if $\\lambda^D$ shows a $U'$-dependence comparable to $\\lambda^M$, or if the susceptibility departs from the RPA-like formula by more than the few percent seen at $U=2$, the central claim fails.","tokens_in":28562,"feed_emoji":"⚡","tokens_out":14564,"duration_ms":185332,"temperature":0.7,"pith_summary":"The paper generalizes the single-boson exchange (SBE) decomposition of the two-particle vertex to nonlocal interactions and applies it in a functional renormalization group (fRG) calculation of the two-dimensional extended Hubbard model. Its central claim is that the charge (density) susceptibility is controlled by an RPA-like formula, $\\chi^D \\approx P^D_{U'=0}/(1+P^D_{U'=0}B^D)$, where $P^D$ is the polarization obtained from the Hubbard model at $U'=0$ with all vertex corrections included, and $B^D$ is the bosonic part of the bare interaction containing $U$ and the nearest-neighbor $U'$. The reason is that the density fermion-boson coupling $\\lambda^D$ is almost independent of $U'$, even while $\\chi^D$ grows strongly with $U'$ and eventually diverges near $U'/U \\approx 1/4$. A fluctuation diagnostics traces this constancy to a cancellation between magnetic and density/superconducting contributions to the renormalization of $\\lambda^D$. The paper also establishes a numerically cheap scheme, using only an s-wave form factor and neglecting rest-function flow and non-trivial high-frequency asymptotics, that is accurate in the weak-to-moderate coupling regime.","feed_headline":"Charge response of the extended Hubbard model is RPA-like","feed_subtitle":"A vertex-corrected polarization at U'=0 controls screening; the density coupling barely moves with U'.","key_machinery":"The argument rides on splitting the bare interaction in each physical channel into a bosonic part $B^X(q)$ and a fermionic part $F^X(k,k')$, so that the single-boson-exchange vertex keeps its factorized form $\\nabla^X = \\lambda^X w^X \\lambda^X$: a fermion-boson vertex $\\lambda^X$ on either side of a screened interaction or bosonic propagator $w^X$. For the density channel $B^D(q)=U+4U'(\\cos q_x+\\cos q_y)$, so the nonlocal interaction enters as the initial value of the bosonic propagator, while the momentum-dependent remainder $F^X$ stays in the irreducible part. The polarization $P^X=\\sum \\lambda^X \\Pi^X$ then determines the screened interaction via $w^X=B^X/(1+B^XP^X)$, and if $\\lambda^D$ does not move with $U'$, substituting the $U'=0$ polarization reproduces Eq. (45). The cancellation that freezes $\\lambda^D$ is exposed by a fluctuation diagnostics that splits it into magnetic, density, and superconducting contributions, and a sign-based poor-man's matrix reproduces the pattern without full numerics. The paper further uses the SBE approximation, neglecting the flow of the multiboson rest function, and checks that this is accurate in the tested regime.","core_discovery":"On the paper's own terms, the central discovery is Eq. (45): for the two-dimensional extended Hubbard model with onsite $U$ and nearest-neighbor $U'$, the charge susceptibility can be written as $\\chi^D \\approx P^D_{U'=0}/(1+P^D_{U'=0}B^D)$, where $B^D(q)=U+4U'(\\cos q_x+\\cos q_y)$ and $P^D$ is the polarization built from the density fermion-boson coupling and the fRG bubble. The point is that $P^D$ is evaluated at $U'=0$ but retains all vertex corrections of the Hubbard model, so the formula is RPA-like in its $U'$-dependence, not bare RPA. The near-independence of $\\lambda^D$ from $U'$ means the nonlocal interaction enters the charge response only through the bosonic bare interaction, which is why $\\chi^D$ grows linearly with $U'$ up to the charge-density-wave divergence near $U'/U \\approx 1/4$. The paper traces this constancy to cancellations between the magnetic and the density plus superconducting contributions in the fluctuation diagnostics of $\\lambda^D$, and verifies a simplified computation scheme that neglects rest-function flow and nonlocal form factors in the weak-to-moderate coupling regime.","pith_inferences":["Going beyond the paper, Eq. (45) offers a cheap interpolation strategy: freeze the Hubbard-model polarization and vary only $B^D$ when scanning $U'$, provided the cancellation in $\\lambda^D$ persists away from half filling and at lower temperature.","The sign-based diagnostic suggests the effect is structural rather than accidental; if so, longer-range nonlocal interactions that enter only through a bosonic density channel would also leave $\\lambda^D$ nearly inert, which is testable in the same fRG setup.","Because $P^D_{U'=0}$ already contains Hubbard vertex corrections, bare-RPA estimates that ignore those corrections will overestimate the charge-density-wave tendency; the paper shows only the dependence on $U'$ is RPA-like, not the absolute value.","The same cancellation logic may carry over to retarded interactions such as phonon-mediated ones, where only the bosonic propagator changes while the density vertex stays close to its Hubbard value; the paper lists the Hubbard-Holstein model as a natural next application."],"forward_implications":["Given the $U'=0$ Hubbard-model polarization, the charge response of the extended model at the tested parameters can be produced by the RPA-like denominator $1+P^D_{U'=0}B^D$ without recomputing the full $U'>0$ flow.","The linear growth of $\\chi^D$ with $U'$ and its divergence near $U'/U \\approx 1/4$ follow directly from the bosonic bare interaction $B^D$, not from interaction-induced vertex renormalization.","The simplified scheme with only an s-wave form factor, no rest-function flow, and no non-trivial high-frequency asymptotics reproduces susceptibilities and couplings within about two percent of the full calculation down to $T=0.1$, making parameter scans numerically feasible.","Magnetic and superconducting susceptibilities do not admit the same RPA-like description, because their fermion-boson couplings do depend on $U'$ through inter-channel feedback.","At stronger coupling or lower temperature the rest-function flow can no longer be neglected, so the RPA-like formula is a weak-to-moderate-coupling statement rather than a general identity."],"supporting_citations":[{"why":"This paper introduces the single-boson exchange decomposition of the vertex that the present work generalizes to nonlocal interactions.","marker":"[53]"},{"why":"This paper supplies the SBE-based fRG flow equations and the SBE approximation that neglects the rest-function flow, both used throughout the present analysis.","marker":"[54]"},{"why":"This paper provides the SBE representation of the functional renormalization group for strongly interacting electrons, including the channel decomposition behind the fluctuation diagnostics.","marker":"[13]"},{"why":"This paper establishes the quantitative multiloop and truncated-unity fRG framework against which the one-loop and two-loop results here are checked.","marker":"[5]"},{"why":"This thesis gives the numerical implementation and form-factor expansion of the SBE-fRG, including the straightforward extension that the B-reducible scheme is designed to improve.","marker":"[20]"},{"why":"This paper computes the competition between antiferromagnetic and charge-density-wave fluctuations in the extended Hubbard model and serves as the baseline for the $U'/U$ crossover found here.","marker":"[38]"},{"why":"This paper introduces fluctuation diagnostics, the method used to expose the cancellations that keep the density fermion-boson coupling nearly independent of $U'$.","marker":"[79]"},{"why":"This paper defines the high-frequency asymptotics of the vertex whose non-trivial $U'$-induced parts are shown to be negligible in the tested parameter regime.","marker":"[17]"}],"fun_headline_variants":["RPA-like charge response from cancellations in density coupling flow","Cancellations in coupling renormalization make charge response RPA-like","U' affects charge screening only via bare interaction, yielding RPA-like form","Charge susceptibility in extended Hubbard model is RPA-like from bare U' interaction"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the density fermion-boson coupling $\\lambda^D$ stays independent of $U'$; this is verified numerically only at $U=2$, $\\beta=10$ for two fillings, and the paper itself notes that the neglected rest-function flow would modify the picture at stronger coupling.","fun_headline_variants_meta":{"raw":{"variants":["RPA-like charge response from cancellations in density coupling flow","Cancellations in coupling renormalization make charge response RPA-like","U' affects charge screening only via bare interaction, yielding RPA-like form","Charge susceptibility in extended Hubbard model is RPA-like from bare U' interaction"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001477,"raw_usage":{"total_tokens":5927,"prompt_tokens":929,"completion_tokens":4998,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":545,"completion_tokens_details":{"reasoning_tokens":4920}},"tokens_in":545,"tokens_out":4998,"duration_ms":37994,"temperature":1.0,"reasoning_tokens":4920,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T18:55:36.299821+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Run the same fRG calculation at stronger coupling and lower temperature (for example $U=4$, $\\beta=20$, $U'$ near the density-wave boundary) while keeping the full rest-function flow and d-wave form factors, and compare $\\chi^D$ with $P^D_{U'=0}/(1+P^D_{U'=0}B^D)$; if $\\lambda^D$ shows a $U'$-dependence comparable to $\\lambda^M$, or if the susceptibility departs from the RPA-like formula by more than the few percent seen at $U=2$, the central claim fails.","supporting_citations":[{"cited_title":"Krien, A","cited_arxiv_id":null,"evidence_quote":"This paper introduces the single-boson exchange decomposition of the vertex that the present work generalizes to nonlocal interactions."},{"cited_title":"Fraboulet, S","cited_arxiv_id":null,"evidence_quote":"This paper supplies the SBE-based fRG flow equations and the SBE approximation that neglects the rest-function flow, both used throughout the present analysis."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"This paper provides the SBE representation of the functional renormalization group for strongly interacting electrons, including the channel decomposition behind the fluctuation diagnostics."}],"review_version":1}