{"id":"94a36377-c541-4c13-8ee6-1b78c45c590d","arxiv_id":"1908.00988","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"Repulsive interactions drive a direct transition between a Chern insulator with Hall conductance 4 and a 1/3 Laughlin state in the Harper-Hofstadter model.","lead":"At a magnetic flux density of 3/11 per plaquette, a lattice model of interacting fermions switches between an integer quantum Hall state and a fractional quantum Hall liquid when repulsion is increased. This is the first numerical evidence for a new kind of interaction-driven topological transition and includes a new way to measure topological order.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Directness of the CI-to-Laughlin transition rests on Ly=6 data; larger cylinders may reveal an intermediate phase, and the paper's own convergence checks do not exclude this.","rationale":"I read the paper as making a specific numerical claim: at nφ=3/11, repulsive interactions drive a direct transition between a C=4 CI and a ν=1/3 Laughlin state. What must be true for that claim: (i) the Laughlin phase is correctly identified, and (ii) no other phase intervenes along the V axis. The Laughlin identification is well supported by flux pumping, entanglement-spectrum counting, and the TEE collapse. The directness is the weaker link. The paper itself flags the limitations: convergence at Ly>6, ambiguous order, and an intermediate phase at nφ=3/10. Thus the reader's weakest-assumption identification is correct. I would not change the CONDITIONAL verdict: the claim is plausible and well-evidenced but not fully established. The proposed test directly addresses the missing evidence.","tokens_in":13849,"tokens_out":4000,"duration_ms":40315,"concrete_test":"Repeat the iDMRG simulation at nφ=3/11 with Ly=8 and Ly=10 (e.g., bond dimension up to χ=2000 or with a more efficient variational update) and sweep V finely across the apparent transition, monitoring the Hall response via flux insertion and the entanglement spectrum. If a third phase with different Hall conductance or entanglement structure appears between the CI and Laughlin states, the direct-transition claim is falsified; if the two phases remain adjacent with a single crossing, the claim is supported. As a complementary check, perform exact diagonalization on finite tori at the same flux density and locate the phase boundaries as a function of V.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim is a direct, interaction-driven transition between a C=+4 CI and a ν=1/3 Laughlin state at nφ=3/11 (abstract, Sec. I). The evidence for directness is drawn almost entirely from iDMRG at Ly=6 (Fig. 2): correlation-length growth with χ, an entanglement-gap discontinuity, a fidelity-susceptibility peak, a jump in the single-particle density-matrix spectrum, and a discontinuous Hall-conductance change. On a cylinder of circumference Ly, any intermediate phase whose width in V is smaller than ~1/Ly will be missed; the paper's own Appendix C and Fig. 7 show that convergence becomes progressively worse for Ly=8 and Ly=10, and the authors explicitly state that iDMRG on cylinders 'is incapable to unambiguously distinguish the nature of a quantum phase transition in two dimensions' (Discussion). They also report that at the nearby flux density nφ=3/10 the transition occurs via an uncharacterized intermediate phase, so an intermediate phase at 3/11 is a realistic possibility. The topological-entanglement-entropy identification of the Laughlin state (Fig. 3) is obtained by combining data over many nφ and Ly and excludes an outlier (p/q=2/7, Ly=5) and all p/q≥3/10; while this supports the existence of the Laughlin phase, it does not constrain the directness of the specific transition. Therefore the load-bearing assumption that no intermediate phase exists between the CI and Laughlin state is not secured by the presented data.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies interacting fermions in the Harper-Hofstadter model at flux density n_φ = 3/11, where the lowest Landau level splits into three magnetic sub-bands. Using iDMRG on cylinders, the authors present evidence for a direct interaction-driven transition between a C = +4 integer Chern insulator and a ν = 1/3 Laughlin state. The evidence includes correlation-length growth with bond dimension, a discontinuous entanglement-gap opening, restructuring of the single-particle density matrix, a fidelity-susceptibility peak, and a jump in the pumped Hall charge from 4 to 1/3. The Laughlin phase is identified through a multi-flux-density topological entanglement entropy fit giving S_top = −0.551(15), momentum-resolved entanglement spectra with the expected counting, density correlations, and flux-pumping response. The paper also discusses the relation to composite-fermion theories and argues that the transition differs from previously studied plateau transitions because the two states belong to different Jain series.","tokens_in":14326,"tokens_out":3370,"duration_ms":37562,"significance":"If the direct CI-to-Laughlin transition at n_φ = 3/11 is confirmed, it would be a genuinely new type of interaction-driven plateau transition, one that cannot be described as a Chern-number change in the composite-fermion band within a single Jain sequence. This would broaden the theoretical landscape of quantum Hall plateau transitions and motivate new field-theoretic descriptions. The paper's technical contributions are also notable: the simultaneous scaling of cylinder circumference and flux density to extract topological entanglement entropy is a useful methodology, and the Laughlin-state identification is checked against external exact values (S_top = −ln√3, σ_H = 1/3) rather than fitted. At the same time, the central directness claim rests on Ly = 6 data, while the authors themselves report increasingly poor convergence for Ly = 8 and Ly = 10 and explicitly acknowledge that iDMRG cannot unambiguously determine the nature of a 2D quantum phase transition. The claim is therefore plausible but not fully secured.","major_comments":[{"comment":"The direct nature of the transition between the C = +4 CI and the Laughlin state is inferred almost entirely from Ly = 6 iDMRG data. As the authors note in the Discussion and Appendix C, iDMRG on cylinders cannot unambiguously distinguish a direct transition from a sequence of transitions in 2D, convergence becomes progressively worse for Ly = 8 and Ly = 10, and at the nearby flux density n_φ = 3/10 the transition occurs via an uncharacterized intermediate phase. Because any intermediate phase whose width in V is smaller than ∼1/Ly would be missed at Ly = 6, the presented data do not secure the load-bearing claim of directness. The authors should either provide a quantitative finite-size analysis that bounds the width of a possible intermediate phase at n_φ = 3/11, or explicitly reword the central claim to state that the data are consistent with a direct transition without asserting that intermediate phases are excluded.","section":"Fig. 2 and Appendix C"},{"comment":"The Hall response is measured on the two sides of the transition, but near V_c the authors state that the flux-insertion procedure violates adiabaticity and cannot be reliably performed. A discontinuous jump between σ_H = 4 and σ_H = 1/3 therefore does not by itself exclude a narrow intermediate phase or a region where the Hall response is ill-defined. The claim of a direct transition requires either a treatment of the near-critical flux-pumping data or an explicit statement that the transition region is narrower than the resolution of the flux-pumping diagnostic.","section":"Fig. 2(f) and flux-pumping discussion"},{"comment":"The topological entanglement entropy estimate combines data across several flux densities and excludes the p/q = 2/7, Ly = 5 outlier and all p/q ≥ 3/10. This analysis supports the existence of a Laughlin phase at n_φ = 3/11, but it does not constrain the directness of the specific transition at that flux density. The text should state this limitation explicitly wherever the TEE result is invoked in support of the transition claim, rather than presenting it as evidence for the directness of the transition.","section":"Fig. 3 and footnote [70]"}],"minor_comments":[{"comment":"The phrase 'direct transition' is used in the abstract and conclusion, while the Discussion acknowledges that the data could reflect either a continuous or a weakly first-order transition; please define what 'direct' means in terms of the absence of any intermediate phase, and consistently distinguish it from the order of the transition.","section":"Abstract and Sec. I"},{"comment":"Panel (a) uses bond dimensions 600–1000 while panels (b)–(f) use χ = 500; the text should state the bond dimension used for each panel at first mention to avoid ambiguity about which data are converged.","section":"Fig. 2 caption"},{"comment":"The definition of the entanglement gap Δξ refers to 'the same quantum number and momentum sector', but the quantum number sectors in Fig. 2(b) are labeled q = −4, …, 2; clarifying the convention for q (e.g., total particle-number sector relative to a reference filling) would improve reproducibility.","section":"Fig. 2(b) and text"},{"comment":"The speculation about a 2D Fermi surface at the critical point, motivated by the growing inferred central charge, should be labeled explicitly as a speculative interpretation, especially given the non-asymptotic finite-entanglement-scaling caveats stated in Appendix C.","section":"Discussion, last paragraph"},{"comment":"There are minor typographical and formatting inconsistencies, including ligature-based spelling variants such as 'effect' and inconsistent use of 'density profile' versus 'density profile'; a careful proofreading pass would be helpful.","section":"Throughout"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is carefully executed and the Laughlin-state identification is credible. The main obstacle to acceptance is that the central claim of a direct transition is load-bearing but rests on Ly = 6 data, with the paper's own convergence checks and the observed intermediate phase at n_φ = 3/10 leaving a realistic possibility of a narrow intermediate phase at 3/11. I would be willing to accept a revised version that either supplies additional finite-size evidence for directness or appropriately softens the claim and its framing throughout, including the abstract."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear [Name],\n\nShort version: this is a solid numerical paper that likely reports a genuinely new transition, but the central \"direct transition\" claim is not as secure as the abstract suggests. Read it for the TEE method and the clean Laughlin identification; treat the directness as provisional.\n\nWhat's new: previous plateau transitions between FQH states (e.g., Lee et al.) stayed within one Jain series. This paper targets a transition between a C=4 Chern insulator and a ν=1/3 Laughlin state, which belong to different Jain series, at flux density 3/11. If right, it's a different kind of interaction-driven transition. The paper also introduces a useful technical trick: combining data at different Ly and n_phi to collapse the entanglement entropy onto a single magnetic-length scaling curve, giving Stopo = -0.551(15) against the exact -0.549. That's a strong, externally benchmarked result.\n\nThe numerics on the transition itself are honestly presented but limited. The main evidence for directness is at Ly=6: correlation length growth, entanglement spectrum discontinuity, fidelity susceptibility peak, and a Hall jump from 4 to 1/3. All consistent, but on a cylinder of circumference Ly any intermediate phase narrower than ~1/Ly is invisible. The paper's own Fig. 7 and Appendix C show convergence degrades for Ly=8 and 10, and the extracted central charge roughly doubles with Ly before the fit breaks down. The authors themselves say iDMRG cannot unambiguously determine the order in 2D, and they note that at nearby flux density 3/10 the transition goes through an uncharacterized intermediate phase. So the stress-test concern is real: an intermediate phase at 3/11 is not excluded.\n\nThe Laughlin phase identification itself is much better supported. The TEE, the entanglement spectrum counting, and the flux-pumped Hall responses across many p/q with p=2,3,4 all point to a stable Laughlin state. That part deserves credit.\n\nMinor soft spots: no code or data released, which makes the convergence claims harder to check; and the TEE fit drops an outlier (2/7, Ly=5) and all p/q ≥ 3/10, which is justified in the text but means the fit is not fully model-free.\n\nBottom line: this paper deserves a serious referee. The directness claim needs either larger Ly or a torus calculation to rule out an intermediate phase, but the Laughlin stabilization and the TEE method are credible and useful. I'd accept it with the caveat that the abstract's \"direct\" should be softened or backed by more evidence.\n\nBest,\n[Name]","headline":"Credible new transition with a strong TEE method, but the directness claim rests on Ly=6 and may not survive larger cylinders.","tokens_in":14703,"tokens_out":2111,"would_cite":true,"duration_ms":19288,"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":"Numerical evidence points to a direct transition from a C=4 Chern insulator to a 1/3 Laughlin state as repulsion grows.","keywords":["Chern insulator","Laughlin state","Harper-Hofstadter model","plateau transition","fractional Chern insulator","topological entanglement entropy","iDMRG","interaction-driven transition"],"falsifier":"A calculation on a wider cylinder (for example $L_y = 12$ or larger) or with higher bond dimension that resolves an intermediate phase between the $C=4$ insulator and the Laughlin state, or a torus exact-diagonalization showing that the ground-state degeneracy changes through an intermediate region, would refute the direct-transition claim. Likewise, observing Hall conductance values other than 4 and 1/3 during flux insertion near $V_c$ would show the transition is not direct.","tokens_in":13667,"feed_emoji":"🧲","tokens_out":4987,"duration_ms":47020,"temperature":0.7,"pith_summary":"The paper argues that in the Harper-Hofstadter model at flux density $n_\\phi = 3/11$, increasing nearest-neighbor repulsion drives a direct quantum Hall plateau transition from a $C = +4$ integer Chern insulator to a $\\nu = 1/3$ Laughlin state. This matters because the two states belong to different composite-fermion Jain series, so the transition cannot be captured by the previously studied Chern-number-changing transitions of composite fermions. The authors support the claim with infinite DMRG simulations on cylinders, showing a correlation-length peak, entanglement-spectrum rearrangement, a discontinuous Hall-conductance jump from 4 to 1/3, and a topological entanglement entropy matching the Laughlin value. They also show that Laughlin states remain stable when the lowest Landau level is split into up to four sub-bands, and they introduce a technique that scales cylinder circumference and flux density together to extract the topological entanglement entropy more efficiently.","feed_headline":"Repulsive interactions drive a direct Chern-to-Laughlin transition","feed_subtitle":"Numerics show a 1/3 Laughlin liquid replacing a C=4 integer Chern insulator as repulsion grows.","key_machinery":"The central object is the Harper-Hofstadter model at rational flux density $n_\\phi = p/q$: a tight-binding model with Aharonov-Bohm phases and nearest-neighbor repulsion $V$. At $p/q$, the lowest Landau level splits into $p$ magnetic sub-bands, which lets integer sub-band filling and fractional LLL filling compete at the same particle density when $p = t$. The computational engine is infinite DMRG on a cylinder, and the decisive diagnostics are flux insertion for the Hall conductance, the entanglement spectrum and gap, the fidelity susceptibility, and the scaling of entanglement entropy with cylinder circumference measured in units of the magnetic length $\\ell_B = a\\sqrt{q/2\\pi p}$. The methodological trick is to vary $n_\\phi$ and $L_y$ together so that entropy data from several flux densities collapse onto one curve, yielding a precise topological entanglement entropy.","core_discovery":"The central claim is that repulsive interactions alone can induce a direct transition between an integer Chern insulator and a fractional quantum Hall state at the same particle density. At $n_\\phi = 3/11$, the lowest Landau level splits into three magnetic sub-bands; the lowest band has Chern number 4, so weakly interacting fermions form a $C = +4$ integer Chern insulator. For strong repulsion, the ground state is a $\\nu = 1/3$ Laughlin state, identified by a flux-pumped Hall response of 1/3 and a topological entanglement entropy of $-\\ln \\sqrt{3}$. The numerical signatures, including correlation-length divergence with bond dimension, a sudden opening of the entanglement gap, a restructuring of the single-particle density matrix, a fidelity-susceptibility peak, and a discontinuous jump in Hall conductance, all point to a direct transition at a critical interaction strength $V_c$, at least on a cylinder of circumference $L_y = 6$. The paper is careful to note that iDMRG on cylinders cannot unambiguously determine whether the transition is continuous or weakly first order in two dimensions.","pith_inferences":["If the direct transition survives in the thermodynamic limit, cold-atom or moir\\'e platforms realizing the Hofstadter model could tune between integer and fractional Hall plateaus by adjusting interaction strength, a knob that is difficult to access in conventional two-dimensional electron gases.","The observed growth of the inferred central charge with cylinder circumference hints at a critical point with a two-dimensional Fermi surface; a testable consequence would be entanglement scaling that depends on circumference and saturates only at large $L_y$.","The joint scaling of $n_\\phi$ and $L_y$ suggests a systematic route to map the Hofstadter phase diagram across many $p/q$ values, potentially revealing other direct Chern-insulator-to-Laughlin transitions whenever $p = t$."],"forward_implications":["Interaction strength becomes a control parameter for quantum Hall plateau transitions, complementing the usual tuning of filling factor or magnetic field.","The transition falls outside existing composite-fermion Chern-number-changing critical theories, so new effective field theories are needed to describe it.","Laughlin physics survives when the lowest Landau level is fragmented into several bands, broadening the class of lattice models that can host fractional Chern insulators.","The magnetic-length scaling method reduces the computational cost of extracting topological entanglement entropy and can be applied to other flux densities.","The indications of an intermediate phase at $n_\\phi = 3/10$ suggest that the phase diagram near such transitions is richer than a single direct transition."],"supporting_citations":[{"why":"Supplies the Hofstadter spectrum and its self-similar structure, which fixes the choice of flux density $n_\\phi = 3/11$ with three sub-bands in the lowest Landau level.","marker":"[44]"},{"why":"Provides the composite-fermion prediction of fractional Chern insulator states in Harper-Hofstadter bands with higher Chern number, which the paper extends to a transition scenario.","marker":"[19]"},{"why":"Describes the prior plateau transition between fractional Chern insulators within the same Jain series, which the present transition is contrasted with.","marker":"[40]"},{"why":"Defines the Laughlin state whose Hall conductance of 1/3 and topological entanglement entropy are matched by the numerics.","marker":"[45]"},{"why":"Establishes quantized Hall conductance and Chern numbers, forming the basis for the flux-insertion measurement of $\\sigma_H$.","marker":"[54]"},{"why":"Provides the cylinder DMRG method in mixed real and momentum space used for the infinite-cylinder simulations.","marker":"[63]"},{"why":"Earlier characterization of a fermionic $\\nu=1/3$ fractional Chern insulator, supplying a baseline for the flux-pumping Hall response.","marker":"[69]"},{"why":"Defines the topological entanglement entropy that is extracted and compared with the Laughlin prediction.","marker":"[71]"},{"why":"Predicts the topological entanglement entropy for fractional quantum Hall ground states, used as the reference value of $-\\ln\\sqrt{3}$.","marker":"[74]"},{"why":"The infinite-size DMRG algorithm that underpins all numerical results in the paper.","marker":"[58]"}],"fun_headline_variants":["Interactions trigger direct Chern-to-Laughlin transition","Repulsion alone drives integer-to-fractional Chern switch","C=4 Chern insulator turns 1/3 Laughlin via interactions","Direct plateau transition: Chern insulator to Laughlin state"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The simulations must faithfully represent the two-dimensional phases and the directness of the transition at the cylinder circumferences and bond dimensions used; the paper itself notes convergence difficulties near the critical point and for larger cylinders.","fun_headline_variants_meta":{"raw":{"variants":["Interactions trigger direct Chern-to-Laughlin transition","Repulsion alone drives integer-to-fractional Chern switch","C=4 Chern insulator turns 1/3 Laughlin via interactions","Direct plateau transition: Chern insulator to Laughlin state"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000249,"raw_usage":{"total_tokens":1580,"prompt_tokens":1004,"completion_tokens":576,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":620,"completion_tokens_details":{"reasoning_tokens":509}},"tokens_in":620,"tokens_out":576,"duration_ms":5322,"temperature":1.0,"reasoning_tokens":509,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T15:26:11.943594+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A calculation on a wider cylinder (for example $L_y = 12$ or larger) or with higher bond dimension that resolves an intermediate phase between the $C=4$ insulator and the Laughlin state, or a torus exact-diagonalization showing that the ground-state degeneracy changes through an intermediate region, would refute the direct-transition claim. Likewise, observing Hall conductance values other than 4 and 1/3 during flux insertion near $V_c$ would show the transition is not direct.","supporting_citations":[{"cited_title":"Interaction-driven plateau transition between integer and fractional Chern Insulators","cited_arxiv_id":"1908.00988","evidence_quote":"Supplies the Hofstadter spectrum and its self-similar structure, which fixes the choice of flux density $n_\\phi = 3/11$ with three sub-bands in the lowest Landau level."},{"cited_title":"Fractional Chern In- sulators in Harper-Hofstadter Bands with Higher Chern Number,","cited_arxiv_id":null,"evidence_quote":"Provides the composite-fermion prediction of fractional Chern insulator states in Harper-Hofstadter bands with higher Chern number, which the paper extends to a transition scenario."},{"cited_title":"Emergent Multi-Flavor QED 3 at the Plateau Transition between Fractional Chern Insulators: Applications to Graphene Heterostruc- tures,","cited_arxiv_id":null,"evidence_quote":"Describes the prior plateau transition between fractional Chern insulators within the same Jain series, which the present transition is contrasted with."},{"cited_title":"Anomalous quantum Hall effect: An incompressible quantum fluid with fractionally charged excitations,","cited_arxiv_id":null,"evidence_quote":"Defines the Laughlin state whose Hall conductance of 1/3 and topological entanglement entropy are matched by the numerics."},{"cited_title":"A Result Not Dependent on Rationality for Bloch Electrons in a Magnetic Field,","cited_arxiv_id":null,"evidence_quote":"Establishes quantized Hall conductance and Chern numbers, forming the basis for the flux-insertion measurement of $\\sigma_H$."},{"cited_title":"Density matrix renormalization group on a cylinder in mixed real and momentum space,","cited_arxiv_id":null,"evidence_quote":"Provides the cylinder DMRG method in mixed real and momentum space used for the infinite-cylinder simulations."},{"cited_title":"Characterization and stability of a fermionic ν = 1/3 fractional Chern insulator,","cited_arxiv_id":null,"evidence_quote":"Earlier characterization of a fermionic $\\nu=1/3$ fractional Chern insulator, supplying a baseline for the flux-pumping Hall response."},{"cited_title":"Topological Entangle- ment Entropy,","cited_arxiv_id":null,"evidence_quote":"Defines the topological entanglement entropy that is extracted and compared with the Laughlin prediction."},{"cited_title":"Hofstadter butterfly as quantum phase diagram,","cited_arxiv_id":null,"evidence_quote":"The infinite-size DMRG algorithm that underpins all numerical results in the paper."}],"review_version":1}