{"id":"c34e5199-a0cb-45a6-988a-e48537c6a337","arxiv_id":"1908.05705","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Rescaled short-range Schrodinger operators converge in norm resolvent sense to the N-boson delta-interaction Hamiltonian in 1D, with explicit O(epsilon^s) rates.","lead":"This paper proves that a one-dimensional gas of N quantum bosons with smooth short-range pair forces converges, as the force range shrinks, to the idealized zero-range contact-interaction model, in a strong operator-norm sense. It also provides the first explicit convergence rate for arbitrary particle number N.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Limit identification is formally circular: Theorem 1.1 invokes Corollary 5.1 to show R(z) is V-independent, but the corollary as stated concerns (H+z)^-1 and is derived only after the theorem is assumed.","rationale":"The reader's weakest-assumption analysis focused on the Gamma-convergence liminf inequality (C.6). That part of the proof appears sound: the trace inequalities (C.1)-(C.2) do give the needed H1 control, and the uniform estimate (C.9) follows from the moment condition for the compactly supported comparison potential V'. The more concrete problem in the written proof is the logical order of the identification step. After (5.4), the theorem proof invokes Corollary 5.1 to assert V-independence of R(z), but the corollary's statement concerns (H+z)^-1 and its proof is located after the theorem and relies on the expression (5.4) being already known as the resolvent of H. This is repairable by promoting the algebra in (5.7)-(5.13) to a lemma about R(z) before the theorem proof; no mathematical estimate seems broken. Because the central claim is very likely correct but the written proof has a formal circularity in the step that connects the Krein limit to the delta Hamiltonian, the appropriate verdict is CONDITIONAL rather than REJECT or ACCEPT as is.","tokens_in":24576,"tokens_out":55694,"duration_ms":539078,"concrete_test":"Reorder the proof as follows: before proving Theorem 1.1, derive from (5.4) and the factorizations (5.7)-(5.8) the identity R(z) = R0(z) + alpha ~S(z)^* (1 - alpha ~phi(z))^-1 ~S(z), using only Sections 2-4 and the algebra of (5.12). Then choose a compactly supported even V' with the same integral as V and apply Corollary C.4 to conclude R(z) = (H+z)^-1. Check that no line of this reordered derivation invokes Theorem 1.1 or the equality R(z) = (H+z)^-1. If the reordered proof is complete, the issue is expository; if not, a genuine circular gap remains in the identification argument.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central theorem hinges on identifying the norm limit R(z) from (5.4) with the resolvent (H+z)^-1. After defining R(z), the proof says: \"To see that R(z) = (H+z)^-1 we use Corollary 5.1, below. By this corollary, the operator R(z) depends on alpha only...\" But Corollary 5.1 is stated as an identity for (H+z)^-1, not for R(z), and its proof appears after Theorem 1.1, introduced as \"the expression (5.4) for the resolvent of H\". Taken literally, the theorem proof uses the corollary to make the identification, while the corollary's proof presupposes that identification. The gap is repairable: the algebra in (5.7)-(5.13) can be reorganized as an independent lemma showing R(z) = R0(z) + alpha ~S(z)^* (1 - alpha ~phi(z))^-1 ~S(z), which is manifestly independent of V; then a compactly supported V' with the same integral and Corollary C.4 give R(z) = (H+z)^-1. But as written, the forward reference does not supply the needed assertion before the theorem is used, so the proof of the crucial identification is formally circular.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"This paper proves that the N-boson Schrödinger operator H_ε with two-body rescaled potentials V_ε(x_j−x_i)=ε^{−1}V((x_j−x_i)/ε) converges in norm resolvent sense, as ε→0, to the δ-interaction Hamiltonian H with strength α=g∫V dr, for any even V∈L^1∩L^2 and g=lim g_ε. The proof is based on a generalized Krein formula, explicit norm convergence of the auxiliary operators A_ε(H0+z)^{−1} and φ_ε(z), and Γ-convergence of the associated quadratic forms. Under the additional decay condition ∫|r|^{2s}|V(r)|dr<∞ and |g_ε−g|=O(ε^s), the resolvent error is O(ε^s). A factorization step shows the limiting resolvent depends only on α, yielding a closed-form resolvent identity (1.13)/(5.13).","tokens_in":24814,"tokens_out":21234,"duration_ms":166992,"significance":"The result is a significant advance over the known N=2 and N=3 cases: it establishes norm resolvent convergence for arbitrary N bosons in one dimension, with explicit rates and a V-independent formula for the limit, without recourse to Faddeev equations. The method (Krein formula plus Γ-convergence) is robust and the lemmas are stated with precise estimates. If the proof's one structural gap is repaired, the paper will be a useful reference for singular perturbation theory and the Lieb-Liniger model.","major_comments":[{"comment":"The proof of Theorem 1.1 identifies the norm limit R(z) with (H+z)^{-1} by invoking Corollary 5.1, which is stated below. Taken literally, the corollary is an identity for (H+z)^{-1}, and its derivation in the closing paragraph of Section 5 uses the phrase \"the expression (5.4) for the resolvent of H\", i.e., it presupposes the very identification R(z)=(H+z)^{-1} that Theorem 1.1 is proving. This is a formal circularity in a load-bearing step. The gap is repairable without changing the results: the factorization and the algebra leading to (5.12) can be reorganized as an independent lemma showing R(z)=R0(z)+α \\tilde S(z)^*(1−α \\tilde φ(z))^{-1}\\tilde S(z) for the operator R(z) defined in (5.4); then, since the right side depends on V only through α, choosing a compactly supported V' with the same integral and applying Corollary C.4 gives R(z)=(H+z)^{-1} by uniqueness of strong limits. Until this reorganization is made, the proof of the central identification is not fully self-contained.","section":"Section 5, Eq. (5.4) and Corollary 5.1"}],"minor_comments":[{"comment":"There are typographical errors (\"semi-bounde d\" and \"S tuttgart\"); please correct them.","section":"Abstract and title page"},{"comment":"The proof of Lemma A.1 is left as an exercise to the reader. Since properties (i)-(vi) are used repeatedly in the main text, a sketch or a reference would improve the self-containedness of the appendix.","section":"Appendix A, Lemma A.1"},{"comment":"The phrase \"the left hand side of (1.13) seems to depend on V\" is only resolved later by Corollary 5.1; consider adding a forward reference at that point.","section":"Section 1, Remark 2"},{"comment":"The identity (5.12) is stated without derivation; a one-line verification using (1−A)^{-1}=1+A(1−A)^{-1} with A=g|u⟩⟨v|⊗\\tilde φ would help the reader.","section":"Equation (5.12)"}],"recommendation":"major_revision","confidential_remarks":"The mathematical content appears sound; the main issue is a fixable circularity in the proof organization. I support publication after a revision that reorganizes the factorization argument. No concerns about attribution or scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nYou should know this: the paper proves the first norm resolvent convergence result for the 1D Bose gas Hamiltonian with delta interactions, for arbitrary N, with explicit rates. Previous work covered N=2 and N=3; the N=3 proof used Faddeev equations that do not generalize. The pair-decomposed Krein formula approach is new, and the convergence estimates in Sections 3 and 4 are detailed and explicit. This is a genuine advance.\n\nThe proof is self-contained and mostly well organized. The weak spot is in Section 5, where the limit operator R(z) is identified with the actual Hamiltonian H. The proof defines R(z) as the limit of the resolvents, then says 'To see that R(z) = (H+z)^{-1} we use Corollary 5.1, below.' But Corollary 5.1 is stated as an identity for (H+z)^{-1}, not for R(z), and its derivation appears after the theorem and presupposes that (5.4) is the resolvent of H. As written, that is a formal circularity. It is not fatal: the algebra in (5.7)–(5.12) is independent of the identification and can be reorganized as a lemma showing R(z) equals the V-independent expression. Then a compactly supported V' with the same integral and Corollary C.4 complete the proof. The authors just need to move that calculation before the theorem and state it for R(z). The gap is repairable and does not affect the main convergence estimates.\n\nOther concerns are minor. Lemma A.1 is left to the reader, which is fine. The Gamma-convergence appendix requires only ∫|V||r|^{1/2} < ∞, weaker than the theorem's decay assumption, but it is used only for the strong-resolvent identification, so no problem. The citation pattern is clean; the one self-citation, [13], is unrelated to the central argument.\n\nOverall, the reader's ACCEPT verdict is fair, provided the authors reorder the proof to remove the circularity. This paper is for mathematical physicists working on zero-range limits, Lieb-Liniger-type models, and resolvent convergence; it deserves a serious referee. I would cite it once the logic is cleaned up.","headline":"Solid new result on norm resolvent convergence for the 1D Bose gas with delta interactions; the main proof has a repairable logical gap in the limit identification.","tokens_in":25368,"tokens_out":5540,"would_cite":true,"duration_ms":48764,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81Q10","81Q15","81V70"],"pacs":[],"model":"deepseek-v4-flash","headline":"For any number of bosons in one dimension, delta-interaction Hamiltonians are norm-resolvent limits of rescaled short-range potentials, with explicit rate O(epsilon^s).","keywords":["one-dimensional Bose gas","Lieb-Liniger model","delta interactions","contact interactions","norm resolvent convergence","Gamma-convergence","Krein formula","short-range potentials"],"falsifier":"For $N=2$ and a Gaussian potential $V(r)=e^{-r^2}$, the resolvent difference can be evaluated numerically at fixed $z$ for several $\\varepsilon$; since $\\int |r|^{2s}|V|\\,dr<\\infty$ for every $s<1$, the theorem predicts decay no slower than $O(\\varepsilon^{0.9})$. Observably slower decay, or none, would refute the rate claim. For the identification claim itself, one can check directly that the operator $R(z)$ in (1.13) satisfies the $\\delta$-interaction boundary condition (the jump condition) on a dense set; failure to satisfy it would mean the limit is a different Hamiltonian.","tokens_in":24362,"feed_emoji":"⚛️","tokens_out":13936,"duration_ms":121060,"temperature":0.7,"pith_summary":"The paper proves that the usual Hamiltonian for N bosons in one dimension with two-body $\\delta$-interactions—infinitely short-range potential spikes—is a norm-resolvent limit of genuine Schrödinger operators with rescaled short-range potentials. For an even potential $V\\in L^1\\cap L^2(\\mathbb{R})$, it defines $H_\\varepsilon=-\\Delta - g_\\varepsilon \\sum_{i<j} V_\\varepsilon(x_j-x_i)$ with $V_\\varepsilon(r)=\\varepsilon^{-1}V(r/\\varepsilon)$ and shows that, whenever $g_\\varepsilon\\to g$, the resolvent operators $(H_\\varepsilon+z)^{-1}$ converge in norm to the resolvent of $H=-\\Delta-\\alpha\\sum_{i<j}\\delta(x_j-x_i)$ with $\\alpha=g\\int V\\,dr$. The convergence is quantitative: if $\\int |r|^{2s}|V(r)|\\,dr<\\infty$ and $|g_\\varepsilon-g|=O(\\varepsilon^s)$, then the resolvent error is $O(\\varepsilon^s)$. Norm resolvent convergence is strong enough to transfer spectra and time evolution from the short-range models to the idealized contact-interaction model, and the proof covers every particle number $N\\geq 2$.","feed_headline":"Delta-interaction Bose gas is a norm-resolvent limit","feed_subtitle":"Resolvent convergence holds for any N; the rate is O(epsilon^s), set by potential decay.","key_machinery":"The load-bearing object is a Krein-type resolvent identity (Appendix B), a formula expressing $(H_\\varepsilon+z)^{-1}$ as the free resolvent $(H_0+z)^{-1}$ plus one additional term built from $\\varphi_\\varepsilon(z)=B_\\varepsilon(H_0+z)^{-1}A_\\varepsilon^*$, where $A_\\varepsilon$ and $B_\\varepsilon$ absorb the scaled potential through multiplication and scaling operators. The proof then establishes operator-norm convergence of $A_\\varepsilon(H_0+z)^{-1}$ to $S(z)$ and of each pair contribution $\\varphi_{ij,\\varepsilon}(z)$ to $\\varphi_{ij,0}(z)$ using explicit Green's function kernels in dimensions 1, 3, and 4 in relative-and-center-of-mass coordinates. A $\\Gamma$-convergence argument on the quadratic forms identifies the strong-resolvent limit with $H$, and the rate $O(\\varepsilon^s)$ comes from Hölder-type estimates on translates of Green's functions (Lemma A.3).","core_discovery":"The central claim, Theorem 1.1, is that $H_\\varepsilon\\to H$ in the norm resolvent sense for every $N\\geq 2$, with the explicit resolvent identity $(H+z)^{-1}=(H_0+z)^{-1}+gS(z)^*(1-g\\varphi(z))^{-1}JS(z)$ for $z\\in\\rho(H_0)\\cap\\rho(H)$. Here $S(z)$ and $\\varphi(z)$ are limits of operators built from the scaled potential, and after integrating out the potential the formula depends on $V$ only through $\\alpha=g\\int V\\,dr$. Under the moment condition $\\int |r|^{2s}|V(r)|\\,dr<\\infty$ and the coupling-rate condition $|g_\\varepsilon-g|=O(\\varepsilon^s)$, the paper proves $\\|(H+z)^{-1}-(H_\\varepsilon+z)^{-1}\\|=O(\\varepsilon^s)$. For $N=2$ and $N=3$ such convergence was known; the contribution here is the general $N$ statement and the explicit rate.","pith_inferences":["Since the limiting resolvent depends on $V$ only through $\\alpha$, the same $O(\\varepsilon^s)$ rate should be universal across different choices of $V$ with equal integrated strength and identical moment decay; this could be checked numerically for Gaussian, exponential, and compactly supported potentials.","The abstract pieces of the proof—the Krein identity and the $\\Gamma$-convergence identification—do not use one-dimensional geometry, so a plausible extension is a two-dimensional analogue for distinguishable particles with a renormalized coupling, where $\\varphi_\\varepsilon(z)$ diverges but its singular part is expected to cancel.","The trace inequalities (C.1)–(C.2) are the only place where collision-plane regularity enters; testing whether the rate $O(\\varepsilon^s)$ can fail when $\\int |r|^{2s}|V|=\\infty$ would clarify whether the moment condition is sharp."],"forward_implications":["Norm resolvent convergence implies convergence of spectra, so eigenvalues and spectral gaps of the short-range models approach those of the $\\delta$-interaction Hamiltonian as $\\varepsilon\\to 0$.","The unitary time evolutions converge in a weighted operator norm: $\\|(e^{-iH_\\varepsilon t}-e^{-iHt})(H+i)^{-1}\\|\\to 0$ uniformly on compact time intervals, and at rate $O(\\varepsilon^s)$ on growing intervals when $s>0$.","The rate estimate needs only an $L^1\\cap L^2$ potential with finite $2s$-th moment; no further smoothness of $V$ is required.","The limiting Hamiltonian does not remember the shape of $V$: any potential with the same integrated strength $\\alpha=g\\int V\\,dr$ produces the same resolvent formula (1.13).","The result holds for all $N\\geq 2$, going beyond the previously settled two- and three-particle cases."],"supporting_citations":[{"why":"Supplies the classical self-adjoint realization of the $\\delta$-interaction Hamiltonian and the known $N=2$ norm-resolvent convergence that this paper generalizes.","marker":"[1]"},{"why":"Gives the previously established norm-resolvent convergence for three particles in one dimension via Faddeev equations, the case that the present argument extends to arbitrary $N$.","marker":"[3]"},{"why":"Used for Theorem 2.19, which extends the explicit resolvent formula from large $z$ to the full common resolvent set.","marker":"[8]"},{"why":"Provides the abstract equivalence between $\\Gamma$-convergence of quadratic forms and strong resolvent convergence, used to identify the limit operator as $H$.","marker":"[9]"},{"why":"Introduces the Lieb-Liniger Hamiltonian (the trapped version of the model considered here), whose approximation by short-range potentials is the subject of the paper.","marker":"[14]"}],"fun_headline_variants":["Bose gas with delta forces is norm-resolvent limit for any N","Short-range potentials approximate delta-interaction Bose gas with explicit rate","Norm-resolvent convergence proven for all N in 1d Bose gas","Explicit O(ε^s) convergence to delta-interaction Bose gas","General N proof: delta Bose gas as limit of short-range potentials"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the energy cost of a wavefunction whose particles coincide is controlled by its kinetic energy; the Gamma-convergence proof uses this control to pin down the limit. If this control failed, the limit of the approximating operators could be a different self-adjoint operator, and the explicit resolvent formula would not describe the $\\delta$-interaction Hamiltonian.","fun_headline_variants_meta":{"raw":{"variants":["Bose gas with delta forces is norm-resolvent limit for any N","Short-range potentials approximate delta-interaction Bose gas with explicit rate","Norm-resolvent convergence proven for all N in 1d Bose gas","Explicit O(ε^s) convergence to delta-interaction Bose gas","General N proof: delta Bose gas as limit of short-range potentials"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.001114,"raw_usage":{"total_tokens":4567,"prompt_tokens":804,"completion_tokens":3763,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":420,"completion_tokens_details":{"reasoning_tokens":3667}},"tokens_in":420,"tokens_out":3763,"duration_ms":25742,"temperature":1.0,"reasoning_tokens":3667,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T13:07:59.020312+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"For $N=2$ and a Gaussian potential $V(r)=e^{-r^2}$, the resolvent difference can be evaluated numerically at fixed $z$ for several $\\varepsilon$; since $\\int |r|^{2s}|V|\\,dr<\\infty$ for every $s<1$, the theorem predicts decay no slower than $O(\\varepsilon^{0.9})$. Observably slower decay, or none, would refute the rate claim. For the identification claim itself, one can check directly that the operator $R(z)$ in (1.13) satisfies the $\\delta$-interaction boundary condition (the jump condition) on a dense set; failure to satisfy it would mean the limit is a different Hamiltonian.","supporting_citations":[{"cited_title":"Albeverio, F","cited_arxiv_id":null,"evidence_quote":"Supplies the classical self-adjoint realization of the $\\delta$-interaction Hamiltonian and the known $N=2$ norm-resolvent convergence that this paper generalizes."},{"cited_title":"The three- body problem in dimension one: from short-range to contact i nteractions","cited_arxiv_id":null,"evidence_quote":"Gives the previously established norm-resolvent convergence for three particles in one dimension via Faddeev equations, the case that the present argument extends to arbitrary $N$."},{"cited_title":"On inverses of Krein’s Q- functions","cited_arxiv_id":null,"evidence_quote":"Used for Theorem 2.19, which extends the explicit resolvent formula from large $z$ to the full common resolvent set."},{"cited_title":"An introduction to Γ -convergence, volume 8 of Progress in Nonlinear Differential Equations and their Applications","cited_arxiv_id":null,"evidence_quote":"Provides the abstract equivalence between $\\Gamma$-convergence of quadratic forms and strong resolvent convergence, used to identify the limit operator as $H$."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the Lieb-Liniger Hamiltonian (the trapped version of the model considered here), whose approximation by short-range potentials is the subject of the paper."}],"review_version":1}