{"id":"929e99e8-12d0-4dae-bfec-8c96aad45393","arxiv_id":"2607.16968","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":2,"one_line_summary":"Under three stated assumptions, a resource-free static surface-code patch cannot sharply measure the magic axis at polynomial acceptance; it must pay with a resource, leave the dilute regime, or accept exponentially rarely.","lead":"A theory paper argues that a plain surface-code patch with no fold or self-dual structure cannot perform the magic-axis measurement that magic-state cultivation needs while accepting runs often, unless it adds a charge-converting resource or accepts exponentially rarely. The result is conditional on a statistical-physics assumption, subcriticality, which the authors isolate as an open conjecture.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The no-go hinges on unproven subcriticality (Conjecture 1); the only numerical support is a 2D spatial proxy, while the operative spacetime connective constant μ_G is uncomputed. If μ_G z_A ≥ 1, the claimed exponential suppression fails.","rationale":"The reader's weakest-assumption identification (Condition 5, subcriticality) is exactly the point on which the central practical claim rests. Theorem 3's dichotomy is valid under Assumptions 1, 2, and Condition 5, but the advertised impossibility—that a resource-free static patch cannot perform the H_XY check at p_acc ≥ 1/poly(d)—requires the exponential suppression of Theorem 2, which is supplied by subcriticality alone. The paper is honest about this, isolating Conjecture 1 as the one open input, but the numerical basis for it is a 2D spatial proxy rather than the operative spacetime μ_G, and Section F shows acceptance can in principle drive the activity toward criticality. Thus the concern is real and load-bearing. However, the paper's own careful hedging and explicit conditionality mean this is not a fatal flaw; the verdict CONDITIONAL remains appropriate. My stress test does not find an internal inconsistency or a disguised circularity, so no verdict change is warranted. I also note that Assumptions 1 and 2 are similarly unproven for arbitrary post-selected effects, but the single most load-bearing uncertainty for the headline claim is Conjecture 1.","tokens_in":36125,"tokens_out":12125,"duration_ms":119406,"concrete_test":"On a spacetime slab of linear size up to d=9 (spatial axes) and time extent T up to 8, build the accepted spacetime cell graph G_d for a concrete resource-free cultivation-style decoder (e.g., the complementary-gap decoder of Section F). Compute the connective constant μ_G directly by transfer-matrix or Monte Carlo enumeration of self-avoiding spines, and independently compute the accepted conversion activity z_A from the cluster expansion of the effect. Then evaluate the line tension τ = -log(μ_G z_A). If τ ≤ 0 for any such decoder at operating acceptance p_acc ≥ 1/poly(d), Conjecture 1 is false and the exponential suppression in Theorem 2 fails, directly refuting the no-go's central claim for that protocol. If τ > 0 with a margin stable in d, the subcriticality assumption gains the missing spacetime evidence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Condition 5 (subcriticality, μ_G z < 1) is the step in Theorem 2 that converts a polynomial prefactor into exponential suppression. The paper does not prove it; Conjecture 1 is explicitly open. The numerical evidence (Numerical estimate 1, Section D) is a 2D spatial self-avoiding-walk enumeration on strips, not a computation of the operative (2+1)-dimensional spacetime connective constant μ_G; the paper admits μ_G is larger, so the proxy overestimates the true critical activity. Moreover Section F shows that a decoder which biases acceptance toward conversion can enhance the per-cell activity from p to O(1), so subcriticality is not automatic from resource-freeness, protection, or acceptance. If μ_G z ≥ 1, Theorem 2's bound does not decay; the no-go then only says a sharp, high-acceptance, resource-free protected check must be critical/supercritical, and Section C explicitly leaves open whether a critical/supercritical sector can achieve coherent alignment. Thus the central practical claim—that a useful check must pay for the magic axis—rests entirely on the truth of Conjecture 1. A secondary but related gap: Assumptions 1 and 2 are also not established for the full resource-free class of arbitrary post-selected instruments; Section F derives them only for local-noise stabilizer families, so the theorem's scope is narrower than the headline 'resource-free' suggests.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies a fixed planar rotated surface-code patch and asks whether a 'resource-free' adaptive/post-selected protocol can perform a sharp Lüders measurement of the magic axis H_XY=(X+Y)/√2 with non-negligible acceptance. It defines a magic witness as the ℓ1 distance of the normalized accepted effect from the stabilizer-effect octahedron. It proves two unconditional statements: (i) a single fine transcript built from resource-free Pauli/stabilizer measurements has isotropic annular charge and cannot implement the H_XY check (Theorem 1), and (ii) the coefficient-norm distance of the accepted effect from the native cone equals p_acc Δ_stab (Lemma 4). It then states a conditional no-go for decoded families: under Assumptions 1–2 (cone-compatible charged-polymer decomposition and minimum-length cleaning) and Condition 5 (subcriticality μ_G z<1), Theorem 2 bounds Δ_stab p_acc ≤ C |V(G_d)| e^{-τ L(d)}; hence sharp magic and p_acc≥1/poly(d) are incompatible for bounded polynomial-volume spacetime graphs (Theorem 3). The remaining input is Conjecture 1 (subcriticality follows from resource-freeness/protection/acceptance), supported only by a two-dimensional strip numerical proxy.","tokens_in":36499,"tokens_out":9144,"duration_ms":97399,"significance":"If Conjecture 1 and the structural assumptions were established, the result would be a valuable resource-necessity theorem for magic-state cultivation and a useful template for measurement no-goes in topological codes. The paper's strengths are the clean effect-level witness, the explicit separation of proved and assumed statements, the honest isolation of the open statistical-mechanics conjecture, and the released deterministic enumeration code. However, the advertised practical claim is conditional on an open conjecture, and the numerical evidence does not address the operative spacetime constant; moreover the structural assumptions are not derived for the full resource-free class defined in Section 2.4. The paper is best read as a rigorous reduction of the no-go to a specific conjecture, not as a settled no-go.","major_comments":[{"comment":"The exponential suppression in Theorem 2 requires Condition 5, μ_G z < 1. The paper does not prove this and states Conjecture 1 as open. The numerical support in Numerical estimate 1 is a two-dimensional strip connective constant μ_⊥(d), not the spacetime μ_G that appears in the theorem; §D itself notes that μ_G is larger. Hence the estimates do not constrain the sign of log(μ_G z). If μ_G z ≥ 1, the geometric sum in eq. (21) does not decay and the advertised trade-off disappears. Section C leaves open whether a critical/supercritical sector can establish coherent alignment. The paper's practical conclusion—that a useful check must pay for the magic axis—is therefore not established; it is a reduction to Conjecture 1. The abstract and Section 6 should state this even more prominently, or the conjecture should be proved at least for the local-noise family of §F.","section":"§5, Conjecture 1; §D, Numerical estimate 1; Theorem 2, eq. (6)"},{"comment":"Definition 2 defines resource-free protocols to include arbitrary charge-preserving CP maps. Assumptions 1 and 2 (cone-compatible charged decomposition and effect-cleaning/minimum-length) are assumed for the accepted effect, and §F derives them only under local stochastic Pauli noise with stabilizer operations. Theorems 2–3 therefore cover a strictly narrower class than the term 'resource-free' in Condition 2 and Corollary 1 suggests. The paper acknowledges this in §6, but the abstract and corollary state the result for 'resource-free' without this qualifier. This is load-bearing because a protocol outside the derived class could evade the no-go without violating any of the five stated conditions. Please either restrict the theorem statements to the polymer-expandable local-stabilizer class or supply the missing derivation (or a counterexample) for the full class.","section":"§2.4 vs §6/F; Assumptions 1–2"},{"comment":"The fine-grained isotropy theorem is proved only for transcripts built from signed Pauli projectors (Lemma 1). Definition 2(iii), however, allows arbitrary charge-preserving non-Clifford CP maps in each cell. Thus the 'unconditional core' does not cover all resource-free single transcripts under the paper's own definition. Section 5 uses the fine-grained theorem to argue that a counterexample must be a coherent alignment of exponentially many stabilizer transcripts; that constraint is weaker than stated for protocols containing non-Clifford charge-preserving filters. Please state Theorem 1 as a stabilizer-transcript result, as the abstract does, and either extend it or explicitly limit its use in the conjecture discussion.","section":"Theorem 1 / Appendix A"}],"minor_comments":[{"comment":"The cancellation P_{X,+} Y P_{X,+} = 0 is correct but may look like a typo; a one-line derivation using {X,Y}=0 would improve readability.","section":"§E, eq. (31)"},{"comment":"The table would benefit from explicit error bars or a statement that the residual finite-size bias is an estimate, not a rigorous bound. The current caption already warns that the proxy is not μ_G, but the table itself invites overreading.","section":"§D, Numerical estimate 1"},{"comment":"The percolation-family row uses mixed symbols (?, ✗, ✓, ✓, ✗) without an explicit legend. The '?' is explained in the caption, but a clearer notation would help distinguish 'not established' from 'does not attempt'.","section":"Table 2"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is unusually honest: the conditional nature of the result, the open conjecture, and the limitations of the numerical proxy are all stated. My main concern is that the central practical claim is a reduction to Conjecture 1 rather than a proven resource-necessity statement, and the structural assumptions are not established for the full resource-free class defined in the paper. These are fixable by reframing and by tightening the theorem statements to the class for which the assumptions are derived. If the journal is willing to publish a conditional no-go with an explicitly isolated conjecture, this could become acceptable after revision; otherwise the contribution may be seen as too far from a settled result."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The short version: this paper tests the folklore that a fold or self-dual patch is necessary for the magic-axis check in surface-code cultivation. It proves a conditional no-go: assuming two structural conditions (a cone-compatible charged decomposition and an effect-cleaning minimum-length property) plus subcriticality, a static resource-free patch cannot do a sharp H_XY check with polynomial acceptance. The one open assumption is honestly isolated as Conjecture 1, and the paper gives a spatial-strip numerical proxy for the entropy scale.\n\nWhat is actually new: the effect-level magic witness Δ_stab, the distance of the accepted effect from the stabilizer octahedron, and the polymer-based extension from single stabilizer transcripts to post-selected, coarse-grained families. The unconditional single-transcript isotropy result is close to the textbook statement that stabilizer operations cannot measure a non-Pauli observable, but it is properly formulated and the paper does not oversell it. The normalization identity (lemma 4) is a neat, sound step that turns the witness into a bound on the charge-converting part.\n\nThe soft spot is exactly Condition 5, subcriticality. The exponential suppression in Theorem 2 needs μ_G z < 1, and this is not proved. The numerical evidence in Numerical estimate 1 is a two-dimensional self-avoiding-walk enumeration on strips, not the operative (2+1)-dimensional spacetime connective constant; the paper acknowledges μ_G is larger, so the proxy overestimates the critical activity. If μ_G z ≥ 1, the no-go degenerates to the statement that a successful check must be critical or supercritical, and Section C explicitly leaves open whether that regime can give coherent alignment. Also, Assumptions 1 and 2 are not established for arbitrary post-selected instruments; Section F derives them only for local-noise stabilizer families. So the headline scope is narrower than 'resource-free' might suggest.\n\nThat said, the paper is unusually explicit about what is proved, what is assumed, and what remains open. It does not hide the conjecture inside a definition. The classification of prior constructions is useful, and the distance-3 example is clear. I see no circularity: the magic witness is defined independently.\n\nThis is a competent, carefully limited theory paper. It deserves a serious referee, not a desk reject. I would bring it to a reading group, but I wouldn't cite it as an unconditional no-go.\n\nRecommendation: send to peer review, and the referees should focus on whether Conjecture 1 is the right open problem and whether the proxy is meaningful.","headline":"A careful conditional no-go that isolates one open conjecture; the numerical proxy is 2D, so the main claim rests on unproven subcriticality.","tokens_in":36966,"tokens_out":4005,"would_cite":true,"duration_ms":34622,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P68","82B43"],"pacs":["03.67.Lx","03.67.Pp"],"model":"deepseek-v4-flash","headline":"A static surface-code patch with no added structure cannot perform the magic-axis check that magic-state cultivation needs while still accepting runs often; the price must be paid in an added charge-converting resource, a non-dilute accepte","keywords":["surface code","magic-state cultivation","magic-axis measurement","anyon charge","stabilizer-effect octahedron","subcriticality","polymer expansion","conditional no-go"],"falsifier":"Exhibit a concrete distance-d surface-code protocol on a fixed patch with no fold, self-duality, or anyon-permuting structure, all cells charge-non-converting, spacetime volume polynomial, fault distance Ω(d), acceptance ≥ 1/poly(d), and normalized logical effect at constant octahedron distance; or evaluate the operative spacetime connective constant μ_G and accepted activity z_A numerically and show μ_G z_A ≥ 1 for a genuinely resource-free family.","tokens_in":35949,"feed_emoji":"⚛️","tokens_out":5845,"duration_ms":56026,"temperature":0.7,"pith_summary":"The paper tests the folklore that a plain surface-code patch cannot measure the magic axis H_XY = (X+Y)/√2 that magic-state cultivation relies on. It proves, under two structural assumptions and one subcriticality assumption, that a static, resource-free, protected patch cannot perform this check sharply while accepting runs with probability at least 1/poly(d). If correct, a useful in-place magic measurement must pay for the magic axis somewhere: by adding a charge-converting resource such as a fold or self-dual patch, by letting the accepted defect ensemble leave the dilute regime, or by accepting only exponentially rarely. The unconditional core is that any single stabilizer-measurement transcript has isotropic anyon charge and therefore cannot host the coherent m–ϵ superposition the check needs, together with an identity showing that the magic of an accepted effect equals its acceptance probability times its distance from the stabilizer-effect octahedron.","feed_headline":"Static surface code can't measure magic axis while accepting often","feed_subtitle":"A useful cultivation check must add charge-converting structure, leave the dilute regime, or accept runs only exponentially rarely.","key_machinery":"The argument is carried by three objects. The magic witness Δ_stab is the ℓ¹ distance of the normalized accepted logical effect from the stabilizer-effect octahedron, the convex hull of Pauli-measurement effects; it separates a coherent magic check from a classical mixture of X and Y readouts. The annular charge content Q_b(A) reads the anyon charge of cleaned, code-preserving Pauli representatives, so a commuting stabilizer projector always has isotropic charge while H_XY requires the non-commuting pair (m, ϵ). The coarse-grained bound decomposes the accepted effect into a native part and a charge-converting part expanded over conversion polymers in the bounded-degree spacetime cell graph,","core_discovery":"The central claim is that five individually natural conditions are jointly incompatible on a fixed distance-d surface-code patch: sharp magic (constant distance from the stabilizer octahedron), resource-freeness (no fold, self-duality, anyon-permuting wall, magic ancilla, or local charge-converting cell), linear protection, inverse-polynomial acceptance, and subcriticality of the accepted charge-conversion sector. The paper proves that a resource-free, protected, sharp check with good acceptance must fail subcriticality; equivalently, under subcriticality the magic–acceptance product is bounded by C|V(G_d)|e^{-τ L(d)}, which forces acceptance to be exponentially small for protected patches.","pith_inferences":["If Conjecture 1 is proved — that resource-freeness and protection force the accepted charge-conversion gas to be subcritical — the no-go becomes unconditional under the two structural assumptions, completing the resource-necessity statement for measurements.","The same method, reading an accepted effect against the free stabilizer polytope and asking which defect sectors the accepted history can support, could plausibly be carried to other non-Clifford measurements and other codes, though the paper does not prove such extensions.","A concrete falsifier of the conditional theorem would be a resource-free, protected protocol with p_acc ≥ 1/poly(d) and Δ_stab ≥ c > 0 on a fixed patch; alternatively, computing the operative spacetime connective constant μ_G and accepted activity z_A and finding μ_G z_A ≥ 1 would place the ensemble in the escape route rather than refute the no-go."],"forward_implications":["If the no-go holds, every known route to a sharp magic-axis check on a surface code must give up resource-freeness; folds, self-dual patches, cross-caps, and code switches are different ways of paying the same charge-conversion price.","A resource-free, protected check that measures the magic axis sharply cannot accept runs with inverse-polynomial probability; the acceptance must decay exponentially in the code distance.","For a single stabilizer-measurement transcript, the obstruction is unconditional: no such transcript can implement the (I ± H_XY)/2 projector, regardless of acceptance.","The magic–acceptance trade-off means cultivation's in-place measurement step is not an artifact that can be optimized away within the resource-free class; effort is better spent making the charge-converting structure cheaper rather than trying to eliminate it."],"fun_headline_variants":["Resource-free magic-axis check on static code is impossible","No free magic-axis measurement on static surface code","Magic-axis check demands extra cost or exponential rarity","Static code can't check magic axis without added structure"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The load-bearing premise is Condition 5: after post-selection, the accepted charge-conversion sector remains subcritical (μ_G z < 1), so the sum over exponentially many patch-spanning conversion strings decays; the paper does not prove this, and if the ensemble is critical or supercritical the no-go collapses.","fun_headline_variants_meta":{"raw":{"variants":["Resource-free magic-axis check on static code is impossible","No free magic-axis measurement on static surface code","Magic-axis check demands extra cost or exponential rarity","Static code can't check magic axis without added structure"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000449,"raw_usage":{"total_tokens":2097,"prompt_tokens":734,"completion_tokens":1363,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":478,"completion_tokens_details":{"reasoning_tokens":1302}},"tokens_in":478,"tokens_out":1363,"duration_ms":9007,"temperature":1.0,"reasoning_tokens":1302,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-01T19:23:17.054567+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Exhibit a concrete distance-d surface-code protocol on a fixed patch with no fold, self-duality, or anyon-permuting structure, all cells charge-non-converting, spacetime volume polynomial, fault distance Ω(d), acceptance ≥ 1/poly(d), and normalized logical effect at constant octahedron distance; or evaluate the operative spacetime connective constant μ_G and accepted activity z_A numerically and show μ_G z_A ≥ 1 for a genuinely resource-free family.","supporting_citations":[],"review_version":1}