{"id":"6a29e5d4-0785-48d8-933f-a1fc488025c1","arxiv_id":"2506.18086","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Every valid linear holographic entropy inequality with integer coefficients has a contraction map, making the contraction-map proof method complete.","lead":"This paper claims to prove that the standard contraction-map method for proving holographic entropy inequalities is complete: every valid linear inequality with integer coefficients has a contraction map. The result would close a long-standing gap in the holographic entropy cone program and justify the method used for all known facet inequalities.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Lemma 4.2/4.3's virtual-edge construction does not produce a holographic geometry that violates the inequality: infinite weights lie outside Definition 4.1, and the min-cut in (2.15) is replaced by a non-minimized RHS sum in (4.23), so the claimed negative sign is not established.","rationale":"The reader's conditional verdict is appropriate, and I agree that the graph-realizability assumption is a weakness. However, I see a more immediate internal problem: even granting the geometry-graph duality, the violating graph built in Lemma 4.2/4.3 is not a partial cube and contains infinite-weight edges, so Theorem 2.1 does not apply to it. More importantly, the proof replaces the true RHS discrete entropy, which is a minimum over cuts, with a non-minimized sum over LHS edge weights. For a contraction map this replacement gives the correct direction because the true RHS entropy is bounded above by the non-minimized sum, but for a non-contraction map the non-minimized difference being negative does not imply the minimized difference is negative. The infinite edge weight masks this gap: it makes both LHS and RHS expressions divergent, and no rigorous finite-limit argument is supplied. Therefore the central completeness theorem is not established by the proof as written. The result may still be true and the gap may be repairable with a careful construction of actual RT-region graphs and a sign-stable finite-limit argument, so I do not move the verdict beyond conditional.","tokens_in":57035,"tokens_out":13521,"duration_ms":152972,"concrete_test":"Take the explicit non-contraction map of Table 2 with the virtual-edge subgraph of Figure 6, assign a finite weight W to each virtual edge, and compute the actual discrete entropies by solving the min-cut programs in (2.15) for every LHS and RHS term rather than using the non-minimized sum in (4.23). Check whether S*_M - S*_N is negative for any finite W; if it is never negative, the sign conclusion in Lemma 4.2 is an artifact of replacing the RHS min-cut by an upper bound. If it is negative for some W, additionally check whether that finite weighted graph admits a realization as a smooth RT-region graph under Theorem 2.1.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Theorem 4.2 depends on Lemma 4.2 and Lemma 4.3, but the object they construct is not shown to be a holographic geometry in the sense of Definition 4.1. Definition 4.1 quantifies over holographic geometries obeying the RT formula, yet the violating subgraph \\tilde G_X contains virtual edges with infinite weight (Definition 4.3) and is described in Section 3.3 as having altered geodesic structure and broken bulk smoothness. Such a graph is not an RT-region graph of a smooth geometry, so it is outside the quantification of Definition 4.1. The footnote's proposed regularization by arbitrarily large finite weights is not sufficient: first, \\tilde G_X violates the adjacency condition d_H_adj = d_G, so Theorem 2.1's geometry-graph duality, which applies to partial cubes, does not cover it; second, the discrete entropy in (2.15) is a minimum over cuts, and a very large edge can be avoided by the minimizing cut. However, (4.23) substitutes the non-minimized RHS sum \\sum d_H(f(x),f(x'))|...| for S*_N and concludes S*_M - S*_N = -\\infty. Since the actual S*_N is less than or equal to that non-minimized sum (cf. (2.40)-(2.42)), the sign of the minimized difference is not established. Thus Lemma 4.2's claim that the graph cannot satisfy the inequality is unsupported, and the contradiction argument for Theorem 4.2 is incomplete at its central step.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper claims to settle the completeness of the contraction map method for holographic entropy inequalities (HEIs). Theorem 4.1 states that a linear entropy inequality with positive integer coefficients is a valid HEI if and only if a contraction map satisfying boundary homology conditions exists. The sufficiency direction is quoted from prior work (Theorem 2.2). The necessity direction is attempted by contradiction: assuming a valid HEI with no contraction map, the authors take a non-contraction map and construct a subgraph of the domain hypercube, introduce infinite-weight \"virtual edges\" (Definition 4.3), and claim that the resulting graph gives a holographic geometry violating the HEI (Lemmas 4.2 and 4.3).","tokens_in":57385,"tokens_out":5234,"duration_ms":57641,"significance":"If the theorem were correct, it would be a substantial advance: it would show that every valid linear holographic entropy inequality in the stated class admits a contraction-map proof, making the contraction map method both necessary and sufficient and completing the program initiated in [5, 15]. The paper is clearly organized, and it is honest about several disclaimers, including the reliance on an unproved graph-realizability assumption for smooth bulk geometries. However, the main theorem is stated unconditionally while the proof depends on this assumption and on a graph construction that is not shown to correspond to a holographic geometry obeying the RT formula. The central contradiction argument therefore does not currently establish the claimed completeness result.","major_comments":[{"comment":"The object ̃G_X constructed in Lemma 4.2 is not shown to be an RT-region graph of a holographic geometry, so it is outside the quantification of Definition 4.1. Definition 4.1 defines a valid HEI over holographic geometries obeying the RT formula, and Theorem 2.1 provides a geometry-graph duality only for partial cubes satisfying the isometric condition. The graph ̃G_X violates the adjacency condition d_H^adj = d_G and is not a partial cube; moreover, Lemma 4.3 itself states that the discrete entropy S*_{L_u} on this graph is divergent and that \"any geometry realized by ̃G_X cannot give the proper holographic entanglement entropy.\" Thus the conclusion of Lemma 4.3 that there exists at least one holographic geometry violating the HEI does not follow from the construction.","section":"§4.1, Definition 4.3 and Lemma 4.2/4.3"},{"comment":"The discrete entropy S*_N in (2.15) is defined as a minimum over cuts, and the chain (2.40)–(2.42) explicitly uses the inequality S*_N ≥ ̄S*_N, where ̄S*_N is the non-minimized sum over edges with weights d_H(f(x),f(x')). In Eq. (4.23), the proof substitutes the non-minimized sum for S*_N and obtains S*_M − S*_N = −∞. Since the actual S*_N is no larger than the substituted expression, the sign of the minimized difference S*_M − S*_N is not established. Additionally, the first sum in (4.23) is asserted to be positive semidefinite, but this is not true for a non-contraction map: other pairs (x,x') may also violate the contraction condition and contribute negatively. This is a load-bearing gap in the proof of Lemma 4.2.","section":"§4.1, Eq. (4.23)"},{"comment":"The proof relies on an unproved graph-realizability assumption: the authors state in the introduction that they expect every sensible holographic geometry to admit an RT-region graph with discrete entropies matching the continuous RT entropies, but they do not prove this. Theorem 2.1, which is the geometry-graph duality underpinning both directions of the proof, is imported without a proof, and its hypotheses exclude the non-isometric graphs constructed later. Because Theorem 4.1 is stated unconditionally for all holographic geometries, this missing assumption is load-bearing; at minimum the theorem should be stated as conditional on the graph-realizability conjecture, and the proof must show that the constructed violating object is an admissible geometry under Definition 4.1.","section":"§1 and Theorem 2.1"}],"minor_comments":[{"comment":"The abstract advertises completeness for linear HEIs with rational coefficients, while Theorem 4.1 is stated for positive integer coefficients. The reduction from rational to integer coefficients should be stated explicitly.","section":"Abstract and Theorem 4.1"},{"comment":"The isometry condition in (2.5) contains a typo: it compares d_G(w,w') with d_G(φ(v),φ(v')) for w,w' ∈ V_J and v,v' ∈ V_J; the right-hand side should be the Hamming distance in the hypercube H_J after applying the isometry φ: V → V_{H_J}. The current notation obscures the intended statement.","section":"Definition 2.3"},{"comment":"The inequality S*_N ≥ ̄S*_N is stated to be provable geometrically from entanglement wedge nesting, but no proof or precise citation is given. Since this inequality is reused in the main proof, it should be proved or referenced to a specific statement in the literature.","section":"§2.4, inequality (2.40)"},{"comment":"The maps ι_M, ι_N, and the graph ̃G are introduced informally with notation that is not fully consistent with Definitions 2.7, 2.8, and 2.10. Clarifying the domains and the role of the chosen connected subset X would improve readability.","section":"§3.3, around Eq. (3.9)"}],"recommendation":"reject","confidential_remarks":"The authors are transparent about the conditional nature of their argument, but the central proof step constructs an object that is explicitly not a proper holographic geometry and then uses it to conclude a contradiction. In my view this is not a local gap but a failure of the main necessity argument as written. I would encourage the authors to either prove the required graph realization and construct a genuine RT-region graph violation, or restate the theorem as a conditional statement about graph models."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The main theorem is new and the paper is honest about what it assumes, but the proof of the necessity direction has a hole at the central step. The iff statement (Theorem 4.1) is what people have been waiting for: sufficiency was known from earlier work, but the converse was open. Section 3's worked examples, especially the MMI deformation, are helpful, and the graph intuition (non-contraction leaves a proper subgraph of the hypercube and breaks adjacency) is clearly explained. Lemma 4.1 is straightforward.\n\nThe soft spot is Lemma 4.2/4.3, which bears the weight of the contradiction argument. The construction replaces null edges with virtual edges of infinite weight (Definition 4.3). But Definition 4.1 quantifies over holographic geometries obeying the RT formula, and a graph in which non-adjacent vertices are connected by an infinite-weight edge is not an RT-region graph of a smooth geometry, as the authors themselves acknowledge when they say smoothness breaks down. Invoking a finite but arbitrarily large weight does not fix it: the discrete entropy in (2.15) is a minimum over cuts, and a min-cut will route around a huge edge. That would be acceptable if the proof used the true S*_N, but (4.23) instead substitutes the non-minimized RHS sum from (2.39). Since S*_N is less than or equal to that sum, the negative sign in (4.23) is not established. The claimed violation is therefore unsupported at the exact step where the contradiction is supposed to bite.\n\nTo the paper's credit, Section 1 lists three disclaimers, including the graph-realizability assumption and the treatment of an adjacency condition as a physical requirement. These are real assumptions, not hidden ones, and the authors flag them plainly. But they are load-bearing: if graph realizability fails, the contrapositive does not go through.\n\nWho this is for: people actively working on the holographic entropy cone and contraction-map methods. The paper is readable and the examples are instructive. It deserves a serious referee, not a desk reject, but the referee should be asked to focus on Lemma 4.2/4.3 and whether the virtual-edge construction can be replaced by a real RT-region graph with finite weights that genuinely violates the inequality. My own verdict is conditional: the result is plausible and new, but the proof is not yet complete.","headline":"A plausible and clearly written attack on the completeness question, but the central contradiction argument rests on a graph construction that is not yet shown to be a holographic geometry.","tokens_in":57882,"tokens_out":2325,"would_cite":false,"duration_ms":23863,"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":"The paper proves that every valid holographic entropy inequality has a contraction map, making the standard proof method complete for this class.","keywords":["holographic entropy inequalities","contraction map","holographic entropy cone","Ryu-Takayanagi formula","partial cubes","RT-region graph","geometry-graph duality","entanglement entropy"],"falsifier":"Take the set of all candidate linear inequalities with positive integer coefficients and a fixed party number, and for each one without a boundary-compatible contraction map search for a positive edge-weight assignment on a connected bitstring subset that satisfies the inequality; the theorem predicts the search always finds a violation, so a single no-contraction-map inequality that survives all such assignments would refute it. Equivalently, exhibiting a smooth holographic geometry whose continuous RT entropies cannot be matched by any discrete RT-region graph would break the geometry-graph duality on which the contradiction argument rests.","tokens_in":56813,"feed_emoji":"🌀","tokens_out":6934,"duration_ms":70289,"temperature":0.7,"pith_summary":"The paper answers, in the affirmative, the open question of whether the contraction map proof method is complete for linear holographic entropy inequalities with rational coefficients. Its central theorem states that a candidate inequality with positive integer coefficients is a valid holographic entropy inequality if and only if there exists a contraction map between two bitstring hypercubes that respects boundary homology conditions. The proof works by showing that any non-contraction map leaves a proper cubical subgraph in its preimage, and that replacing the missing edges by virtual edges of infinite weight produces a graph whose discrete entropies violate the inequality. Since by the geometry-graph duality such a graph corresponds to a holographic geometry obeying the RT formula, an inequality without a contraction map cannot be valid. If correct, the result upgrades the contraction map method from a proof technique to a complete characterization of this class of inequalities.","feed_headline":"Every valid holographic entropy inequality has a contraction map","feed_subtitle":"The standard proof method is complete: any non-contraction map yields a holographic geometry that violates the inequality.","key_machinery":"The load-bearing object is the contraction map $f:\\{0,1\\}^M \\to \\{0,1\\}^N$, defined by the condition that Hamming distance never increases, together with its graph-theoretic realization as a graph map between hypercubes. The proof centers on the nontrivial preimage $P(\\Phi_f(H^M))$: for a contraction map this preimage is the full $M$-dimensional hypercube, while for a non-contraction map it is a proper cubical subgraph with missing edges. The missing edges, called virtual edges, are assigned infinite weight in the RT-region graph, a graph whose vertices are bulk regions and whose weighted edges record crossings of RT surfaces. These virtual edges make the discrete entropy of the LHS divergent while the RHS stays finite, which is the mechanism that produces a violating holographic geometry.","core_discovery":"The paper's central claim is that validity of a linear holographic entropy inequality and existence of a contraction map are equivalent. Given an inequality $\\sum_i \\alpha_i S_{P_i} \\ge \\sum_j \\beta_j S_{Q_j}$ with positive integer coefficients $\\alpha_i,\\beta_j$, set $M=\\sum_i\\alpha_i$ and $N=\\sum_j\\beta_j$; the paper claims there is a contraction map $f:\\{0,1\\}^M \\to \\{0,1\\}^N$, mapping the occurrence bitstrings of the LHS subregions to those of the RHS, if and only if every holographic geometry satisfying the RT formula obeys the inequality. The converse is established by contradiction: assuming a valid inequality has no contraction map, every candidate map is non-contractive, which forces the preimage of the image graph to be a proper subgraph of the hypercube. In that subgraph, the edges that a non-contractive map collapses are treated as virtual edges with infinite weight; the resulting RT-region graph gives a holographic geometry for which the LHS discrete entropy is infinite while the RHS is finite, violating the inequality. Hence the supposed valid inequality fails, so no valid inequality can lack a contraction map.","pith_inferences":["Beyond the paper: if the graph-realizability assumption holds, the same iff statement should transfer to any entropy functional defined by min-cuts on graphs, so tensor-network or quantum error-correcting models with graph entropies would inherit the completeness result.","The virtual-edge argument suggests a practical diagnostic for invalid candidate inequalities: check whether the forced map deletes an edge between single-bit neighbors; such an edge deletion is the signature of a violating bulk chambering.","A natural next step, not taken here, is to make the proof constructive and effective: given a valid inequality, extract an explicit contraction map; the examples indicate deterministic filling rules may already do this.","The theorem shifts attention from proving individual inequalities to classifying contraction maps, so the main bottleneck for higher-party cones becomes combinatorial enumeration rather than geometric construction."],"forward_implications":["Every valid linear holographic entropy inequality with rational coefficients can in principle be proved by the contraction map method; no separate proof technology is needed for this class.","The algorithmic enumeration of holographic entropy inequalities via partial cubes and graph contractions is complete: it generates all valid linear inequalities, not just those already known.","A candidate inequality can be certified false by proving that no boundary-compatible contraction map exists, because the theorem excludes valid inequalities without such a map.","The existence question reduces to a finite combinatorial search for distance-nonincreasing maps between hypercubes with prescribed boundary values, independent of the detailed bulk geometry.","Since the theorem identifies valid inequalities exactly with contraction maps, the holographic entropy cone for any party number is determined by the set of such maps."],"supporting_citations":[{"why":"Supplies the Ryu-Takayanagi formula that defines holographic entanglement entropy as minimal bulk surface area.","marker":"[2]"},{"why":"Provides the original contraction map proof of sufficiency, the geometry-graph duality, and the discrete entropy/min-cut framework used throughout.","marker":"[5]"},{"why":"Establishes the equivalence between contraction maps and graph contraction maps on hypercubes and partial cubes, the basis for the algorithmic enumeration.","marker":"[15]"},{"why":"Supplies deterministic filling rules and structural properties of contraction maps used in the examples and in motivating completeness.","marker":"[21]"},{"why":"Gives the partial cube theory, including isometric embedding, used to identify RT-region graphs with hypercube subgraphs.","marker":"[25]"},{"why":"Introduces the Winkler equivalence relation on edges used to match edge classes to RT surfaces.","marker":"[26]"},{"why":"Contextualizes the relation between entanglement wedge nesting and contraction maps, clarifying what the proof does not rely on.","marker":"[37]"}],"fun_headline_variants":["Contraction map method proven complete for holographic entropy inequalities","Every valid holographic entropy inequality admits a contraction map","Necessity proven: holographic entropy inequalities require contraction maps","Completeness of contraction map proof: no valid inequality lacks one"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The proof assumes that every holographic geometry obeying the RT formula can be replaced, for entropy purposes, by a discrete graph whose edge-weight min-cuts reproduce the continuous RT entropies exactly; if a geometry admits no such graph realization, the constructed counterexample may not be a real holographic geometry.","fun_headline_variants_meta":{"raw":{"variants":["Contraction map method proven complete for holographic entropy inequalities","Every valid holographic entropy inequality admits a contraction map","Necessity proven: holographic entropy inequalities require contraction maps","Completeness of contraction map proof: no valid inequality lacks one"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000699,"raw_usage":{"total_tokens":3131,"prompt_tokens":893,"completion_tokens":2238,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":509,"completion_tokens_details":{"reasoning_tokens":2169}},"tokens_in":509,"tokens_out":2238,"duration_ms":15166,"temperature":1.0,"reasoning_tokens":2169,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-15T18:55:04.610310+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take the set of all candidate linear inequalities with positive integer coefficients and a fixed party number, and for each one without a boundary-compatible contraction map search for a positive edge-weight assignment on a connected bitstring subset that satisfies the inequality; the theorem predicts the search always finds a violation, so a single no-contraction-map inequality that survives all such assignments would refute it. Equivalently, exhibiting a smooth holographic geometry whose continuous RT entropies cannot be matched by any discrete RT-region graph would break the geometry-graph duality on which the contradiction argument rests.","supporting_citations":[{"cited_title":"Ryu and T","cited_arxiv_id":null,"evidence_quote":"Supplies the Ryu-Takayanagi formula that defines holographic entanglement entropy as minimal bulk surface area."},{"cited_title":"Ovchinnikov,Graphs and Cubes","cited_arxiv_id":null,"evidence_quote":"Gives the partial cube theory, including isometric embedding, used to identify RT-region graphs with hypercube subgraphs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Introduces the Winkler equivalence relation on edges used to match edge classes to RT surfaces."}],"review_version":1}