{"id":"0f1f875c-8747-4d2e-a46b-d9c680b3e961","arxiv_id":"2511.10758","paper_version":3,"verdict":"REJECT","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"Any non-k-Schmidt-number-breaking quantum channel can, in principle, be certified via a semiquantum signaling game using only trusted inputs.","lead":"This paper proposes a method to certify that a quantum channel preserves entanglement of dimension greater than k, using only trusted state preparation and untrusted measurements. If correct, it would give a general test for high-dimensional entanglement-preserving channels, important for quantum communication and memory verification.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Eq. (21) uses the witness beyond its proven domain; the gap is real but repairable via a strong Schmidt-number witness, so the theorem likely survives with a corrected proof.","rationale":"The reader identified the same load-bearing concern: the soundness step applies the channel-set witness to arbitrary k-SNB CP maps, which is not justified by Eq. (7). I agree that the proof as written is invalid. However, the concern does not force outright rejection because the missing property can be supplied by a standard Hahn-Banach separation in state space. The theorem's claim is existential, and the CJ state of any non-k-SNB channel has Schmidt number strictly greater than k; hence there always exists a strong Schmidt-number witness that is nonnegative on all states with SN <= k and negative on J_N. Such a witness automatically satisfies Eq. (7) and also gives the needed cone inequality for J_{F*∘E}. The concrete examples already use optimal Schmidt-number witnesses of this strong type, which is why they work. Thus the central claim is likely correct, but the proof must be revised to explicitly choose a state-space witness and extend the inequality to the cone of k-SNB CP maps. I would therefore make acceptance conditional on that revision rather than rejecting the paper outright.","tokens_in":14721,"tokens_out":12100,"duration_ms":118817,"concrete_test":"Re-derive Eq. (21) using the strong witness W' separating J_N from S_k = {ρ : SN(ρ) <= k} rather than W from (7). Specifically, write the soundness condition as Tr[W' J_{F*∘E}] >= 0 for every POVM effect Pi_0 (equivalently, every CP map F with 0 <= J_F <= I) and every E in k-SNBC, and verify that it follows from W' being nonnegative on the SN <= k cone plus Lemma 2. If the derivation succeeds, the central theorem is valid after a proof revision; if it fails on a concrete example (e.g. the qutrit depolarizing channel with W'_opt = I⊗I − 3/2 P_3 and random Pi_0), the theorem would need to be weakened.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"Eq. (21) is not implied by Eq. (7). Eq. (7) gives Tr[W J_E] >= 0 only for CJ operators of k-SNB channels, i.e. for PSD operators with the fixed marginal Tr_B[J] = I/d. In the soundness argument, Bob's POVM effect Pi_0 is identified with the CJ operator of a CP map F, and the inequality is applied to J_{F*∘E}. Since F*∘E is CP but generally not TP, its CJ operator need not lie on the channel slice; Lemma 2 only certifies SN(J_{F*∘E}) <= k. Nonnegativity of W on that larger cone does not follow from (7). This is the load-bearing gap identified by the reader.\n\nThe gap is real but repairable. Because N ∉ k-SNBC, Lemma 1 implies SN(J_N) > k. The set S_k of states with Schmidt number at most k is convex and compact, so there exists a strong Schmidt-number witness W' with Tr[W' J_N] < 0 and Tr[W' ρ] >= 0 for all ρ in S_k. Such W' satisfies (7) and, by normalization, extends to all PSD operators with SN <= k, covering J_{F*∘E}. The proof should be amended to construct W' from state-space separation rather than from separation against k-SNBC. With that amendment, Theorem 1 (and the NPT analogue using PPT witnesses) goes through.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper introduces a measurement-device-independent method for certifying that a quantum channel is not k-Schmidt-number-breaking (k-SNB). The proposed semiquantum signaling game uses trusted quantum input states and an untrusted joint measurement on the output of the channel and a second prepared state. The central result, Theorem 1, claims that for every channel outside k-SNBC there is such a game whose average payoff is negative, while every k-SNB channel yields nonnegative payoff for all measurements. The proof constructs a witness operator by Hahn-Banach separation from the set of k-SNB channels and decomposes it into local density matrices to define the game. The method is also extended to non-NPT-breaking channels in Appendix A, and experimental implementations are discussed with explicit examples.","tokens_in":15018,"tokens_out":9748,"duration_ms":106352,"significance":"If the central claim is correct, the paper provides a substantial advance: a faithful (i.e., applicable to all channels) certification of effective entanglement dimension under minimal assumptions, avoiding the need for trusted measurements, entangled inputs, or auxiliary side channels. The paper also demonstrates the reach of the method with explicit examples and a proposed optical implementation, and it extends the framework to NPT-breaking channels. The proof strategy, based on Lemma 2 and the Choi-Jamiołkowski isomorphism, is elegant. However, the soundness proof of Theorem 1 contains a significant gap: the witness from Eq. (7) is used beyond its proven domain. The gap is repairable by using a Schmidt-number witness constructed instead from state-space separation, and with that amendment the central claim is likely sound.","major_comments":[{"comment":"The soundness condition (ii) of Theorem 1 is not proven. The witness W is obtained by Hahn-Banach separation from the convex set k-SNBC, so Eq. (7) guarantees Tr[W J_E] ≥ 0 only when J_E is the Choi operator of a k-SNB channel, i.e., a PSD operator with the fixed marginal Tr_B[J_E]=I/d (up to normalization). In the soundness argument, Π0^AB is identified with the Choi operator of an arbitrary CP map F, and Eq. (21) applies the witness to J_{F*∘E}. Lemma 2 only shows that this Choi operator has Schmidt number at most k; it does not put it on the channel slice, and Eq. (7) gives no control there. The sentence 'the final inequality follows from the properties of W as a k-SNB witness' is therefore unsupported. This is load-bearing because condition (ii) must hold for every measurement, including those whose associated F is not trace-preserving. The gap is repairable: choose W as a Schmidt-nu","section":"Sec. III, Theorem 1, Eq. (21)"},{"comment":"The same gap appears in the NPT-breaking extension. The witness W̃ from Eq. (A3) is only known to be nonnegative on Choi operators of NPT-breaking channels. The proof applies it to J_{F*∘S}, which by Lemma 4 is PPT but is not necessarily the Choi operator of a channel. Nonnegativity of W̃ on all PPT operators does not follow from separation from the set of NPT-breaking channels. The proof should either construct W̃ as a PPT witness, nonnegative on all PPT operators, or explicitly justify that the separation in (A2)-(A3) can be chosen to have this stronger property. Without this, the soundness of Theorem 2 is also incomplete.","section":"Appendix A, Theorem 2"}],"minor_comments":[{"comment":"The notation for the inverse Choi isomorphism is hard to follow, especially the double transpose ([ξ^x_{A'}]^⊺)^⊺. Please clarify the convention and the subsystem labels so the identity N(ρ)=Tr_{A'}[J^N_{A'A}(ρ^T⊗I)] is unambiguous.","section":"Sec. III, Eq. (13)"},{"comment":"The matrix in Eq. (27) is poorly formatted and difficult to read; a properly typeset matrix would help. Also, the relation between the states ξ^x, the unitary gates U_x, and the input states ψ^x, φ^y should be stated before Eq. (28) to avoid confusion.","section":"Sec. IV, Example 1"},{"comment":"The term 'k-SNB CP map' is used before being defined. Please define it explicitly, e.g., as a CP map whose Choi operator has Schmidt number at most k, and state how the CJ isomorphism is normalized for non-trace-preserving maps.","section":"Sec. II, Lemma 2"},{"comment":"The proof of Eq. (16) is somewhat terse; a few extra lines showing the substitution of Eq. (13) and the contraction with W would improve readability. This is a presentation issue only.","section":"General"}],"recommendation":"major_revision","confidential_remarks":"The paper's main idea is promising and the central claim is probably correct after a standard repair of the soundness proof. The identified gap in Eq. (21) is real and load-bearing, but it can be fixed by constructing the witness from state-space separation rather than channel-space separation. I therefore recommend major revision rather than rejection. The authors should also update the analogous proof in Appendix A. No issues with circularity or free parameters were found."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"First thing to know: the paper is not ready as written, but the main theorem looks correct and the flaw is a single unjustified step with a straightforward fix. If you're deciding whether to engage, the answer is yes—with the expectation of a revised proof.\n\nWhat's actually new here: this is the first faithful minimal-assumption certification of all non-k-SNB channels, going beyond the steering-based approach [41], which is not faithful. The move of using Lemma 2 (closure of k-SNB under composition with arbitrary CP maps) to lift Rosset et al.'s argument from EB channels to SNB channels is genuine, and the same trick extends cleanly to NPT-breaking channels. The explicit qutrit examples with finite-dimensional witnesses and the circuit in Fig. 2 are concrete and useful.\n\nWhat's soft: the proof of Theorem 1's soundness condition. Eq. (21) asserts Tr[W J_{F*∘E}] ≥ 0 because W is a k-SNB witness, but the witness in Eq. (7) was only separated against the set of k-SNB channels, not against all k-SNB CP maps. J_{F*∘E} need not be a channel, so its CJ operator does not have the fixed marginal that channel CJ operators have. The step does not follow as written. This is not a trivial footnote; it's the part that ensures no false positives.\n\nHowever, the gap is repairable. Since N ∉ k-SNBC, Lemma 1 gives SN(J_N) > k, so J_N can be separated from the compact convex set of states with Schmidt number at most k by a strong Schmidt-number witness W' that is nonnegative on that whole state set, which by scaling extends to all positive semidefinite operators with SN ≤ k. With that W' in place, the contested inequality goes through. The same fix applies to the NPT result using PPT witnesses. The examples in the paper already use strong witnesses (e.g., W_opt^2), so the experimental demonstrations are unaffected.\n\nSo the result likely stands. The paper deserves a serious referee, not a desk reject. The referee should ask for the proof to be recast using the strong witness, and for a brief note on why the strong witness can be expressed in the required separable form (it can, since it's Hermitian).\n\nBottom line: a solid advance with one repairable hole. Worth citing once the fix is in.","headline":"A real proof gap in the soundness argument, but a clean repair exists; the result deserves serious peer review, not a desk reject.","tokens_in":15558,"tokens_out":4118,"would_cite":true,"duration_ms":38893,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["81P45","81P40"],"pacs":["03.67.-a","03.65.Ud"],"model":"deepseek-v4-flash","headline":"This paper proves that any quantum channel capable of preserving entanglement beyond a fixed dimension can be certified with a game that assumes trust only in state preparation.","keywords":["Schmidt-number-breaking channels","semiquantum signaling games","entanglement dimension certification","measurement-device-independent","Choi–Jamiołkowski isomorphism","NPT-breaking channels","quantum channel certification"],"falsifier":"Take any non-k-SNB channel N and its separating witness W from Eq. (7), then find a k-SNB channel E and a CP map F such that Tr[W J_{F*∘E}] < 0. Numerically, one could fix a qutrit depolarizing channel (non-2-SNB for appropriate λ), use the optimal witness from Example 1, and search over simple CP maps F (e.g., constant maps outputting a pure state) to see if the inequality in Eq. (21) fails.","tokens_in":14511,"feed_emoji":"🎲","tokens_out":6227,"duration_ms":56029,"temperature":0.7,"pith_summary":"The paper claims that the property of a quantum channel being non-k-Schmidt-number-breaking—meaning it can preserve entanglement of dimension greater than k—can be faithfully certified using semiquantum signaling games, assuming only trusted preparation of input product states and no trust in measurement devices. The method is faithful: every non-k-SNB channel admits a game with negative average payoff, while every k-SNB channel gives nonnegative payoff for all measurement choices. This extends earlier measurement-device-independent certification of non-entanglement-breaking channels to the whole hierarchy of Schmidt-number-breaking channels. The same framework also certifies non-NPT-breaking channels.","feed_headline":"Semiquantum game certifies channels that keep high-dim entanglement","feed_subtitle":"No entangled inputs or trusted measurement devices are needed—only trusted preparation of product states.","key_machinery":"The Choi–Jamiołkowski isomorphism maps each channel to a bipartite state (its CJ operator), allowing resource classes to be seen as convex sets of states. The proof uses Lemma 2, which says that composing a k-SNB channel with any completely positive map yields a k-SNB CP map, to extend the witness inequality from channels to the broader class of CP maps. This extension is the load-bearing step in proving soundness.","core_discovery":"Theorem 1 states that for every channel N not in the set of k-Schmidt-number-breaking channels, there exists a semiquantum signaling game whose average payoff is negative on N and nonnegative on every k-SNB channel. The game is constructed from a witness operator W that separates N from the convex set of k-SNB channels; the payoff is defined so that its average equals Tr[W J_N] for a specific measurement, and the nonnegativity for all k-SNB channels is argued through a structural closure property.","pith_inferences":["If the witness-extension gap is resolved, the theorem would show that semiquantum games are complete for a whole family of channel resource theories; if not, the soundness guarantee might only hold for a restricted subset of witnesses.","Because the soundness proof currently assumes the witness is nonnegative on all k-SNB CP maps, a counterexample would likely come from a k-SNB channel composed with a CP map that stretches the CJ operator outside the witness's positive cone.","The trusted-preparation assumption could be quantified by examining how tolerances in state preparation affect the payoff, parallel to how noise-robustness analyses are done for steering-based certificates."],"forward_implications":["Any non-k-SNB channel can be certified without false positives from k-SNB channels, regardless of Bob's measurement choices, provided the input states are prepared faithfully.","The method removes the need for entangled input states and side channels, making it practical for quantum memory and communication link verification.","The same construction works for non-NPT-breaking channels, and more generally for any resource-breaking class that satisfies analogues of Lemmas 1 and 2.","The proposed optical circuit (inverse controlled-shift plus inverse QFT) gives a concrete path to experimental implementation."],"fun_headline_variants":["Semiquantum game certifies channel dimension with minimal assumptions","Game separates channels that preserve high-dim entanglement","Faithful test for channel entanglement dimension from a game","No entangled inputs: game certifies channel dimensions"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof of Theorem 1's soundness step assumes that the Hahn-Banach witness operator W, which is constructed to be nonnegative on all k-SNB channels, is also nonnegative on the Choi operators of all k-SNB completely positive maps—a strictly larger set that the separation theorem does not guarantee.","fun_headline_variants_meta":{"raw":{"variants":["Semiquantum game certifies channel dimension with minimal assumptions","Game separates channels that preserve high-dim entanglement","Faithful test for channel entanglement dimension from a game","No entangled inputs: game certifies channel dimensions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000337,"raw_usage":{"total_tokens":1663,"prompt_tokens":666,"completion_tokens":997,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":410,"completion_tokens_details":{"reasoning_tokens":936}},"tokens_in":410,"tokens_out":997,"duration_ms":11229,"temperature":1.0,"reasoning_tokens":936,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T22:21:54.213300+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take any non-k-SNB channel N and its separating witness W from Eq. (7), then find a k-SNB channel E and a CP map F such that Tr[W J_{F*∘E}] < 0. Numerically, one could fix a qutrit depolarizing channel (non-2-SNB for appropriate λ), use the optimal witness from Example 1, and search over simple CP maps F (e.g., constant maps outputting a pure state) to see if the inequality in Eq. (21) fails.","supporting_citations":[],"review_version":1}