{"id":"67fbfd89-e114-46fc-be33-97c0fe8e6696","arxiv_id":"2412.21058","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"The inner horizon of Rindler AdS3 maps to the inner RT surface, which encodes timelike entanglement entropy and gravitational anomaly corrections to holographic entanglement.","lead":"This paper identifies a new bulk surface, the inner RT surface, as the image of the inner horizon of a Rindler black hole in AdS3. The surface gives a geometric picture for timelike entanglement entropy and for gravitational anomaly corrections to entanglement in holographic 2d CFTs.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The twist/IRT equivalence in Sec. 6 hinges on the unproved algebraic identity Eq. (156); if it fails, the central anomaly-reproduction claim loses its derivation.","rationale":"The reader's weakest assumption is that the BTZ inner-horizon formula (93) transfers to arbitrary Rindler intervals. That is a real assumption, but the paper does not rely on it alone: Sec. 6 is intended to prove the same result starting from the known twist/worldline action of [50]. The true load-bearing step of that proof is the unproved identity (156), which equates the boundary term of the twist integral to the IRT length parameter. If this identity is wrong, the equivalence between the twist description and the IRT-surface description fails, and the central claim is left only with the heuristic compact-BTZ analogy. I am not claiming the identity is false; I am claiming it is the specific, checkable place where the argument must stand or fall. This partially agrees with the reader, who flagged the twist/IRT equivalence as an assumption but attributed it to the transfer of (93) rather than to the unproved algebraic step in Sec. 6. The TEE replica issue in Sec. 2.2 is genuine but secondary: it concerns the interpretation of the real part as a von Neumann entropy, not the geometric and anomaly-reproduction claims. My recommendation is unchanged: conditional acceptance pending verification of the explicit identity (156).","tokens_in":49436,"tokens_out":23609,"duration_ms":244167,"concrete_test":"Perform an independent symbolic computation of both sides of Eq. (156) for a generic boundary point P = (U1, V1) in the causal development of A with lU != lV, substituting the explicit expressions (104), (150), (64), and (32). Verify that the equality holds identically for arbitrary U1, V1, lU, lV (including the lU = lV limit with a consistent sign convention). If the two sides differ, the twist/IRT equivalence in Sec. 6 is broken and the central reproduction claim loses its derived support.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim that the Chern-Simons correction to holographic entanglement entropy is reproduced by the length of a geodesic chord on the inner RT surface is established in Sec. 6 via the twist description of [50]. The decisive step is the assertion, in Eq. (156), that log((q - ~q)|_H . n|_H) equals the IRT length parameter btau_ÔH, where q, ~q are the reference frame (104), n is the timelike tangent v_bgamma2_P|_H from (150), ÔH is the partner point (64), and btau is defined in (32). This identity is the bridge between the twist integral and the IRT-surface length, but it is stated without derivation. If it fails for a generic interval, the claimed equivalence in Sec. 6 collapses; the earlier argument via Eq. (93) is an analogy from compact BTZ black holes to Rindler intervals and does not by itself prove the interval result. The separate replica interpretation for timelike intervals in Sec. 2.2 is explicitly acknowledged as incomplete and is not needed for the anomaly-reproduction claim.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the pre-image of the inner horizon of the Rindler AdS3 obtained by the bulk Rindler transformation for a spacelike interval A, calling it the inner RT (IRT) surface. It argues that the IRT surface is exactly the spacelike geodesic representing the real part of the holographic timelike entanglement entropy for the partner timelike interval, with the timelike geodesic at the boundary of the extended entanglement wedge representing the imaginary part. In the context of topologically massive gravity, it proposes that the Chern-Simons correction to the holographic entanglement entropy is the regulated length of a geodesic chord on the IRT surface, and that the anomalous part of the balanced partial entanglement entropy is the length of a saddle geodesic chord (the \"inner EWCS\") connecting the two pieces of the IRT surface. Section 6 attempts to prove the equivalence between the twist description of [50] and the IRT length via Eq. (156).","tokens_in":49638,"tokens_out":5822,"duration_ms":57128,"significance":"If established, the paper's central claim would be a genuinely useful result: the anomalous correction to holographic entanglement entropy, previously encoded in a normal-frame \"twist\" along the RT surface, would become the length of a purely geometric geodesic chord on the inner RT surface, and the inner EWCS would give a geometric picture for the anomalous mixed-state correlation. The paper contains many explicit analytic computations, including the Rindler mapping, the fine structure of modular momentum slices, the explicit IRT surface equations, and a flat-limit comparison with the swing surface prescription. The derivation in Section 6 is a genuine attempt to connect the twist and IRT descriptions, and the appendices provide useful technical detail. However, the central equivalence rests on an unproved algebraic identity, and the transfer of the BTZ inner-horizon entropy formula to Rindler intervals is assumed; these gaps must be closed before the main claim can be regarded as established.","major_comments":[{"comment":"The equivalence between the twist description and the IRT length rests entirely on the identity log((q - \\tilde{q})|_H \\cdot n|_H) = \\hat{\\tau}_{\\hat{H}}, stated in Eq. (156). This identity is asserted without derivation, and it is the bridge between the twist integral in Eqs. (154)-(155) and the IRT length parameter defined in Eq. (32). Since Eq. (157) and the claimed equivalence in Section 6, as well as the appendix-G explanation of Eq. (125), all depend on this identity, the paper should provide a proof. If the identity fails for a generic interval, the central anomaly-reproduction claim loses its derivation, and the earlier argument via Eq. (93) remains only an analogy from compact BTZ black holes to Rindler intervals.","section":"Section 6, Eq. (156)"},{"comment":"The derivation of the anomalous holographic entanglement entropy (95) starts from the known BTZ inner-horizon entropy formula (93) and the modular Hamiltonian correction (107), and then transfers this result through the Rindler mapping to arbitrary boundary intervals. The text explicitly calls these \"our starting points.\" This transfer is load-bearing: without it, the claim that the IRT geodesic chord computes the anomalous part of the entropy for general intervals is not established. The authors should either justify the transfer from compact BTZ horizons to the non-compact Rindler black string with the regulated cutoffs, or clearly state this step as an assumption whose failure would leave the reproduction claim heuristic.","section":"Section 4.3, Eqs. (93), (95), (107)"},{"comment":"The paper proposes that the timelike entanglement entropy can be interpreted as a holographic von Neumann entropy by applying the Lewkowycz-Maldacena replica prescription to the IRT surface. The text itself later acknowledges that this replica interpretation is incomplete, stating that the role of the timelike geodesic in the analog replica story is \"unclear\" and that the point will be revisited in the future. Because the abstract and summary present the timelike-entanglement interpretation as one of the paper's main results, this limitation should be stated more prominently. If this replica interpretation is not needed for the anomaly-reproduction claim, the paper should say so explicitly; if it is part of the claim, it needs a concrete derivation or a clear relegation to conjecture.","section":"Section 2.2, around Eqs. (36)-(40)"}],"minor_comments":[{"comment":"The sentence \"where T~U and T~U are the parameters\" should read \"T~U and T~V\"; as printed, the second symbol repeats the first.","section":"Section 2.1, text after Eq. (10)"},{"comment":"\"One the other hand\" should be \"On the other hand\".","section":"Section 5.1"},{"comment":"\"were we also take L(...)\" should be \"where we also take L(...)\".","section":"Section 5.2.2"},{"comment":"The phrase \"interaction line\" should be \"intersection line\" in the sentences following Eqs. (30) and (31), and in the related discussion of M_\\pm and the IRT surface.","section":"Sections 2.2 and 3.2.2"}],"recommendation":"major_revision","confidential_remarks":"To the editor: the manuscript is within the scope of SciPost Physics, and the computations are explicit and mostly reproducible. The main concern is the lack of a proof of Eq. (156), which the authors should be asked to supply or to clearly relegate to a conjecture. There is no indication of a novelty-disclosure problem; the reliance on the authors' prior work [51,59,67] is transparent."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nBottom line: this is a serious paper, not a desk reject. The genuinely new object is the pre-image of the inner horizon of Rindler AdS3, which they call the inner RT surface, and the paper shows it does real work: it anchors the real part of timelike entanglement entropy, and with a proper regulation its geodesic chord length reproduces the Chern-Simons correction to holographic entanglement entropy in TMG. That unification of the twist/normal-frame description with lengths on a purely geometric surface is the main contribution. The construction of the extended entanglement wedge and the inner EWCS for mixed-state correlations is also new and fits naturally into the authors' prior PEE framework.\n\nWhat is good: the computations are explicit and mostly self-consistent. The partner-point identification on the IRT surface via modular momentum slices is worked out in detail, the flat limit is treated carefully, and the final entropy and BPE numbers match known results. The paper is honest about its limitations: Sections 2.2 and 7 flag that the role of the timelike geodesic in the replica story remains unclear, and Section 4.3 explicitly names Eq. (93) and the modular-Hamiltonian shift (107) as starting points.\n\nThe soft spots are fixable but real. First, the load-bearing assumption is that the BTZ inner-horizon entropy formula (93) transfers, via the Rindler map, to arbitrary boundary intervals. That is plausible because the Rindler wedge is locally the same BTZ-like metric, but the paper treats it as obvious rather than proven; a referee should ask for a direct justification. Second, the decisive equivalence in Section 6, Eq. (156), between the twist expression log((q-~q)·n) and the IRT length parameter bτ is stated without derivation. It may be a straightforward substitution using (104), (150), and (64), but the paper should show the steps; as written, the central anomaly-reproduction claim rests on an unproven algebraic identity. The stress-test note is on target here. Third, the Lewkowycz-Maldacena replica interpretation for timelike intervals is asserted, not derived, and the paper itself acknowledges the incompleteness.\n\nOverall, the central geometric picture is coherent and the weaknesses are repairable. This is exactly the kind of paper that should go to referees rather than be desk rejected. I would bring it to the reading group and would cite it for the IRT surface construction.","headline":"A genuinely new geometric object (the inner RT surface) with real explanatory payoff, but the key algebraic bridge to the twist description and the replica interpretation are asserted rather than proven.","tokens_in":50163,"tokens_out":2063,"would_cite":true,"duration_ms":22448,"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":"The paper argues that the inner horizon of Rindler AdS3 maps to a bulk geodesic—the inner RT surface—that simultaneously reproduces the real part of holographic timelike entanglement entropy and the Chern-Simons correction to holographic…","keywords":["inner horizon","inner RT surface","timelike entanglement entropy","Rindler AdS3","topologically massive gravity","gravitational anomaly","entanglement wedge cross section","partial entanglement entropy"],"falsifier":"Compute the anomalous part of the entanglement entropy for a boundary interval using the normal-frame worldline action (100) under two different but equally valid smooth normal-frame configurations along the RT surface; if the result is frame-dependent and not equal to the regulated inner-RT chord length (113), the claimed equivalence between twist and IRT descriptions fails. A second check would be to test whether the inner-EWCS saddle (148) still equals the twist-based correction when the balance conditions (116) are violated; the equality should break precisely there.","tokens_in":49192,"feed_emoji":"🕳️","tokens_out":7707,"duration_ms":73656,"temperature":0.7,"pith_summary":"This paper tries to show that a single bulk object, the pre-image of the inner horizon of Rindler AdS3—called the inner RT surface—carries two quantum-information meanings that were previously described separately. For any spacelike boundary interval, its causal development has two tips; the timelike interval joining those tips has a holographic timelike entanglement entropy whose real part is a spacelike geodesic, and the paper identifies that geodesic with the inner RT surface. In topologically massive gravity, the Chern-Simons correction to holographic entanglement entropy has been computed by a twist along the RT surface; the paper claims this correction is simply the regulated length of a geodesic chord on the inner RT surface, and that the anomalous part of the balanced partial entanglement entropy is the length of a saddle geodesic connecting two pieces of that surface, the inner entanglement wedge cross section. If correct, anomaly corrections become purely geometric lengths rather than normal-frame-dependent worldline data, and timelike entanglement acquires a bulk replica interpretation.","feed_headline":"Inner horizon geodesics carry timelike and anomalous entanglement","feed_subtitle":"The same geodesic surface gives timelike entanglement and Chern-Simons corrections in AdS3/CFT2.","key_machinery":"The load-bearing object is the inner RT surface $\\widehat E$, defined as the inverse Rindler image of the inner horizon $\\tilde\\rho=-T_{\\tilde U}T_{\\tilde V}$ of Rindler $\\widetilde{\\mathrm{AdS}}_3$. It is an extremal surface, the fixed-point set of the modular momentum flow $k_t^{\\beta,\\text{bulk}}$, and its length parameter $\\hat\\tau$ organizes geodesic chords whose lengths reproduce partial entanglement entropies. The argument works by mapping the interval to a thermal state, computing thermal entropy on the outer and inner horizons, and then mapping back; because the inner horizon length is already known to give the Chern-Simons entropy correction in topologically massive gravity, the IRT chord length inherits that role for arbitrary boundary intervals.","core_discovery":"On the paper's own terms, the central discovery is that the pre-image of the inner horizon of the Rindler $\\widetilde{\\mathrm{AdS}}_3$ in the original Poincaré AdS3, denoted $\\widehat E$, is exactly the spacelike geodesic used in the holographic description of the timelike entanglement entropy of the partner interval, and moreover that with a suitable cutoff this same surface computes the Chern-Simons correction to entanglement entropy. In formulas, for an interval $A: (-l_U/2,-l_V/2)\\to(l_U/2,l_V/2)$, the inner RT surface is $\\widehat E:\\ \\rho=-2l_V/(l_U(l_V^2-4V^2)),\\ U=-(l_U/l_V)V$, anchored at the two tips of the causal development $D_A$. The regulated chord on $\\widehat E$ gives the anomalous entropy $S^a_A=\\frac{1}{4\\mu G}\\log(l_U\\varepsilon_V/(l_V\\varepsilon_U))$, matching both the replica-method result and the twist-description result. The paper further claims that the mixed-state correlation dual to the entanglement wedge cross section receives an anomalous part equal to the length of a saddle geodesic connecting the two components of the inner RT surface of the mixed state, called the inner EWCS, and that the twist description of [50] and [51] is equivalent to this purely geometric description.","pith_inferences":["The paper does not draw this, but if the replica argument for the inner RT surface is valid, timelike entanglement entropy would be a genuine von Neumann entropy generated by the modular momentum, not merely an analytic continuation of the spacelike formula.","The paper leaves implicit that the equivalence between twist and inner-RT length turns the twist observable in pure AdS3 into a timelike-entanglement diagnostic; a direct test would be to compare twist fluctuations along the RT surface with fluctuations of IRT chord lengths under boundary perturbations.","The construction is specific to three bulk dimensions and locally AdS3 spacetimes; a speculative extension would ask whether an inner-horizon pre-image plays a similar role for gravitational-anomaly corrections in higher-dimensional holography, where no chord-length formula is currently known."],"forward_implications":["The real part of holographic timelike entanglement entropy is given by the length of the inner RT surface, and the imaginary part by the timelike geodesic at the boundary of the extended entanglement wedge.","The Chern-Simons correction to entanglement entropy in TMG/CFT with gravitational anomaly is a regulated geodesic length on the inner RT surface, not a normal-frame-dependent quantity.","The anomalous part of the balanced partial entanglement entropy equals the length of the inner EWCS, a saddle geodesic connecting pieces of the inner RT surface.","The twist description and the IRT description agree because the normal-frame boundary data along the RT surface encode the same point-to-point partnership as the modular momentum slices.","In the flat limit, the IRT picture reduces to the swing-surface picture of holographic entanglement entropy in flat-space holography."],"supporting_citations":[{"why":"Defines holographic timelike entanglement entropy and its two-geodesic picture that the paper identifies with the IRT surface.","marker":"[37,38]"},{"why":"Supplies the twist description of entanglement entropy with gravitational anomaly that the paper reproduces from the IRT chord length.","marker":"[50]"},{"why":"Gives the anomalous BPE/EWCS twist computation whose result the inner EWCS reproduces.","marker":"[51]"},{"why":"Establishes that the Chern-Simons correction to BTZ entropy equals the inner-horizon length, the starting point of Sec. 4.3.","marker":"[31]"},{"why":"Provides the Rindler method mapping interval entanglement entropy to thermal entropy of the outer horizon.","marker":"[13]"},{"why":"Supplies the replica/LM prescription that the paper applies to the IRT surface for timelike intervals.","marker":"[17]"},{"why":"Gives the Rindler-method and swing-surface picture used for the flat-limit comparison and null lines.","marker":"[16]"},{"why":"Provides the modular-slice and geodesic-chord framework for partial entanglement entropy used throughout.","marker":"[59]"}],"fun_headline_variants":["Inner horizon geodesics compute timelike and anomalous entanglement","Inner RT surface yields timelike entropy and CS corrections","From inner horizon to timelike and anomalous entanglement","Inner horizon geodesics carry timelike and anomalous information","Inner RT surface: timelike entropy and Chern-Simons twist"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The argument depends on the known result that the Chern-Simons correction to black hole entropy in topologically massive gravity equals the inner-horizon length, together with the assumption that this relation transfers, via the Rindler map, to arbitrary boundary intervals; it also assumes the replica argument applies to the inner RT surface so that timelike entanglement entropy has a von Neumann entropy interpretation.","fun_headline_variants_meta":{"raw":{"variants":["Inner horizon geodesics compute timelike and anomalous entanglement","Inner RT surface yields timelike entropy and CS corrections","From inner horizon to timelike and anomalous entanglement","Inner horizon geodesics carry timelike and anomalous information","Inner RT surface: timelike entropy and Chern-Simons twist"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000206,"raw_usage":{"total_tokens":1519,"prompt_tokens":1193,"completion_tokens":326,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":809,"completion_tokens_details":{"reasoning_tokens":248}},"tokens_in":809,"tokens_out":326,"duration_ms":3348,"temperature":1.0,"reasoning_tokens":248,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T23:04:16.965512+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the anomalous part of the entanglement entropy for a boundary interval using the normal-frame worldline action (100) under two different but equally valid smooth normal-frame configurations along the RT surface; if the result is frame-dependent and not equal to the regulated inner-RT chord length (113), the claimed equivalence between twist and IRT descriptions fails. A second check would be to test whether the inner-EWCS saddle (148) still equals the twist-based correction when the balance conditions (116) are violated; the equality should break precisely there.","supporting_citations":[],"review_version":1}