{"id":"96c33548-ed25-4720-adfa-7316daa48557","arxiv_id":"2607.24693","paper_version":2,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"A concrete qutrit channel has P=Q=0 at every blocklength while being neither PPT nor antidegradable, via a variance-dominating signed lift weaker than antidegradability.","lead":"An explicit qutrit channel has zero private and quantum capacity yet is neither PPT nor antidegradable. It answers two long-open questions by introducing a weaker, observable-level substitute for no-cloning.","discovery_kind":"new_method","skeptic_critique":{"model":"moonshotai/kimi-k3","headline":"No significant objection identified: the analytic chain (Prop 3.4 → Lemma 3.6 → Thm 3.5 → Lemma 3.2 → Cor 3.1) checks out on independent re-derivation; the only load-bearing residual risk is the finite entrywise algebra certifying the CP defect (Prop 4.2 / Appendix A).","rationale":"The reader identified the variance-to-BKM conversion (Thm 3.5 / Lemma 3.6) as the weakest assumption; on independent re-derivation that bridge is airtight — it uses only adjoint-preservation, trace cyclicity, an exact variational square-completion, and a standard logarithmic-mean integral, with no step requiring positivity of J. So I disagree that this is the soft spot. The reader also acknowledged \"residual algebraic risk\" in the appendices, and that is where I locate the genuine (but small and fully checkable) load-bearing item: the CP-defect Kraus factorization of Prop 4.2 is the single certificate on which complete variance domination, and hence the entire capacity conclusion, rests. My spot check of the scalar witness example (4.24)/(4.25) is consistent with the claimed factorization, and the NPT spectrum and two-extension witness objective also re-derive exactly, which raises confidence that the appendices were computed carefully. Because every remaining risk is a finite symbolic computation that either passes or fails outright, and the proof architecture plus all cited external reductions (HRSF22 tensorization, BKM Hessian identity, Myhr–Lütkenhaus symmetric-extension criterion) are standard and correctly oriented, the concern does not warrant moving the verdict. ACCEPT stands; the proposed Choi-eigenvalue computation would convert the residual algebraic risk into a mechanically verified fact.","tokens_in":20043,"tokens_out":9300,"duration_ms":318456,"concrete_test":"Symbolically (e.g., Mathematica/SymPy exact arithmetic) construct Λ, Λᶜ (via K₀=1₃/√2, K_{µ+1}=A_µ/2) and J from (4.13), form the Choi operator of the defect map ∆(τ)=L_{Λ(τ)}−J†∘L_{Λᶜ(τ)}∘J on the 9-dimensional input space, and check (a) Choi(∆)⪰0, (b) rank Choi(∆)=3 with nonzero eigenvalues all equal to 8/9, matching (A.6). Independently recompute the J=1 block (B.6) from the 6j matrices (B.4) and confirm H(W)⪰0 at (w₀,w₁) in (5.13). If Choi(∆) has any negative eigenvalue, Theorem 4.3 collapses; if it is PSD rank-3, the capacity proof is fully certified.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I re-derived the steps the reader flagged as weakest and found them sound. (1) Lemma 3.6: the right-weight inequality (3.34) follows from (3.32) applied to Y† using only adjoint-preservation of J and cyclicity of trace (Tr(A†Aω)=Tr(AωA†)); the variational identity (3.39) is exact completion of the square for strictly positive A_{θ,t}; the t-integration (3.37) matches the BKM Γ⁻¹ because ∫₀¹dt/((1−t)λᵢ+tλⱼ)=(lnλᵢ−lnλⱼ)/(λᵢ−λⱼ). No hidden boundedness or positivity assumption: J need not be positive anywhere in the argument, since only the superoperator Löwner order J†∘A_{ω,t}∘J⪯A_{σ,t} is used. (2) Prop 3.4's qubit-extension trick (3.27)–(3.28) is correct: means vanish on the |0⟩⟨0| block and the variances reduce to the claimed cross-term expressions. (3) The bridge Lemma 3.2's ε-regularization is legitimate because Φ_R(ρ_{ε,t})⪰(ε/d_Rd_A)1⊗Φ(1_A)≻0, and joint convexity plus lower semicontinuity give convergence to the support-compatible limit. (4) Orientation of the less-noisy order in Cor 3.1 is correct (environment dominates receiver), and I(U;E)−I(U;B)=Σp[D(Λᶜρᵤ‖Λᶜρ̄)−D(Λρᵤ‖Λρ̄)]≥0. (5) Spot checks of the channel-specific data all pass: the Kraus normalization is trace-preserving (½1+¼·2·1=1); the complementary-channel block (4.10) follows from (A_νA_µ)_{jk}=δ_{νk}δ_{jµ}−δ_{νµ}δ_{jk}; the scalar example (4.25) is internally consistent with the variance identity (4.24): r₀=|0⟩, r₁=−½|1⟩, r₂=−½|2⟩ give (2/9)(1/3)(3/2)=1/9=2/9−1/9; the witness minimization in §5/Appendix B (f'(x), minimizers w₀=(2√3−4)/5, w₁=(2√3−3)/5, Tr(Wω)=(4√3−7)/10<0) re-derives exactly. The one place the entire capacity conclusion rests on unverified arithmetic is the CP-defect identity ∆(τ)=(2/9)ΣG_µτG_µ† (Prop 4.2, table A.4): if a sign or coefficient were wrong there, Choi(∆) could acquire a negative eigenvalue and complete variance domination — the sole certificate feeding Thm 3.5 — would fail. This is a finite, machine-checkable identity, not a structural gap.","agreement_with_reader":"partial"},"referee_report":{"model":"moonshotai/kimi-k3","summary":"The manuscript exhibits the x=1/2 member of the qutrit noisy Werner–Holevo family, Λ(X)=½X+¼(Tr(X)1−Xᵀ), and proves that its optimized private and coherent information vanish at every blocklength (P⁽¹⁾(Λ⊗ⁿ)=Q⁽¹⁾(Λ⊗ⁿ)=0 for all n), hence P(Λ)=Q(Λ)=0, while its Choi state is NPT (indeed one-copy distillable) and the channel is not antidegradable. The proof introduces a \"complete variance-dominating signed lift\": a complex-linear, adjoint-preserving, generally nonpositive map J with J†∘Λᶜ=Λ whose reference-stable variance contraction is certified by complete positivity of an outer defect superoperator Δ_J (Prop. 3.4). Complete variance domination is converted to complete BKM-metric domination (Thm 3.5 via Lemma 3.6), bridged to the complete less-noisy order in relative entropy (Lemma 3.2), which tensorizes (citing [HRSF22]) and forces the one-shot informations to vanish (Cor. 3.1). For the concrete channel, J is given in closed form (Eq. 4.13), the lift identity is verified by direct algebra (Lemma 4.1), and the defect admits an explicit rank-three Kraus factorization Δ(τ)=(2/9)∑G_μτG_μ† (Prop. 4.2, Appendix A). The NPT and non-antidegradability statements are proved by exact finite calculations (spectrum of the partial transpose, Eq. 5.4; an exact symmetric two-extension witness, Prop. 5.3, Appendix B).","tokens_in":20687,"tokens_out":3419,"duration_ms":133677,"significance":"If the result holds, it settles two explicitly stated open problems: Smith and Smolin's question whether zero quantum capacity can occur outside the PPT and antidegradable classes [SS12], and Buscemi–Datta–Strelchuk's question whether antidegradability is the only nontrivial mechanism forcing P=0 [BDS14]. To my knowledge this would be the first unconditional, explicit, finite-dimensional example of this kind; prior candidate constructions in the nondegradable regime remained conditional on additivity-type assumptions. The capacity conclusion is not defined into existence: it follows from an independently checkable order relation (cRE of the complement) via cited tensorization theorems, and the channel-specific certificates — the closed-form lift J, the exact rank-three CP-defect factorization with explicit Kraus operators G_μ, the partial-transpose spectrum, and the analytic two-extension witness with closed-form optimizer — are all finite and fully displayed. The variance-domination/signed-lift mechanism (Thm 3.5, Prop. 3.4, and the exact variance identity (4.24)) is a conceptual contribution of independent interest: a strictly weaker, observable-level relaxation of antidegradabil","major_comments":[],"minor_comments":[{"comment":"Abstract and §1: the phrase 'neither antidegradable nor positive under partial transposition (PPT)' applies PPT to the channel without definition; only Choi-state PPT is standard. Suggest 'whose Choi state is not PPT' at first occurrence, matching the usage in §5.","section":"Abstract / §1"},{"comment":"Eq. (4.14): the equality (Λᶜ)†∘J = Λ† = Λ uses self-adjointness of Λ, which is not explicitly stated before this point (it is clear from Eq. (4.1) but worth one clause, since the analogous identity J†∘Λᶜ=Λ in Eq. (4.15) is the form used in Definition 3.3).","section":"§4.2, Lemma 4.1"},{"comment":"Prop. 4.2 is certified by the entrywise expansion in Appendix A, and Prop. 5.3 by the 6j-symbol block calculation in Appendix B. Both are finite and displayed, but a short supplementary script (e.g., symbolic or exact-arithmetic verification of Eqs. (A.2)–(A.4) and the block matrices (B.6)) would materially lower the barrier to independent checking; this is a service to the reader, not a gap in the argument.","section":"Appendices A–B"},{"comment":"Eq. (6.1): the decorated implication arrow '⇓ ̸⇑' is informal for a display in the Discussion; suggest spelling out 'the converse fails (shown by Λ)'. Also in the same paragraph, 'by Lemma 4.1 Propositions 3.4 and 4.2' is missing a comma/conjunction.","section":"§6, Eq. (6.1)"},{"comment":"Corollary 3.1 is described as standard and attributed to [HRSF22, Proposition 3.2 and Theorem 4.13]; since the orientation (environment dominates receiver) is the crux of the capacity conclusion, it would help to state at first use that Nᶜ ⪰_cRE N is the direction that makes private/coherent information nonpositive, as the proof in fact does — a one-line signpost in the statement would suffice.","section":"§3.1, Cor. 3.1"},{"comment":"Remark 5.2: the chain D→(ω_Λ) ≤ Q→(Λ) = Q(Λ) = 0 invoking [BKN00] is correct but compressed; one sentence recalling that forward classical communication does not increase quantum capacity would save readers a trip to the reference.","section":"§5, Remark 5.2"},{"comment":"The comparison with quantum statistical morphisms [Bus16] in §3.3 is useful; consider also remarking explicitly that for a physical (completely positive unital) J the definition reduces to antidegradability via the Kadison–Schwarz inequality (this is said around Eq. (3.17) but is worth repeating where the hierarchy in Eq. (6.1) is introduced).","section":"§3.3 / §6"},{"comment":"Typographical: the displayed spectrum in Eq. (5.4) uses a nonstandard multi-set notation '(1/4)^×5, 0, (−1/12)^×3'; a multiplicity annotation would be clearer. Minor notational collision: 'N' is used both as the generic channel in §3 and inside the Kraus/complementary-channel formulas in §4 with system labels suppressed; the suppression is announced but occasionally requires backtracking.","section":"§5, Eq. (5.4); §4"}],"recommendation":"accept","confidential_remarks":"The channel is the x=1/2 point of the noisy Werner–Holevo family previously studied by Roofeh and Karimipour [RK24, RK25], who left a gap between capacity bounds at exactly this parameter; the present paper closes that gap and adds the PPT/antidegradability separations, so the novelty relative to that literature is real but should be positioned precisely (the authors already do this fairly). The manuscript includes an explicit AI-assistance disclosure; the mathematics is conventional and fully displayed, and I see no integrity concern, but the editor may wish to confirm the journal's policy on such disclosures. Given the prominence of the claimed resolution of the Smith–Smolin and Buscemi–Datta–Strelchuk questions, I would recommend a second referee independently verify the finite algebra in Appendices A and B; my own checks of the load-bearing steps (the defect factorization on matrix units, the 6j-block positivity, and the orientation of the less-noisy order in Corollary 3.1) found them correct."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This one does what it claims. They give an explicit qutrit Werner–Holevo channel at x=1/2, prove that every blocklength has vanishing private and coherent information (hence P=Q=0), and separately show the Choi state is NPT (one-copy distillable) and not two-extendible on the output. That answers the Smith–Smolin existence question and the Buscemi–Datta–Strelchuk question about whether only antidegradability kills private capacity.\n\nWhat is new is the intermediate object: a complete variance-dominating signed lift. J is adjoint-preserving, generally nonpositive, reconstructs every receiver expectation from the environment, and contracts variance even with a reference. They certify that by an exact rank-three CP factorization of the outer defect, then ride the published BKM → relative-entropy → complete less-noisy pipeline to all-blocklength zero information. The hierarchy (antidegradable ⇒ signed lift ⇒ cRE of the complement ⇒ P=Q=0) is clean, and they prove the first arrow is strict for this channel.\n\nThe general lemmas check out on re-derivation: left-weight domination to BKM does not need positivity of J, the qubit trick for the defect is correct, and the capacity orientation is right. Channel-specific spot checks (Kraus TP, complementary block, variance identity example, witness minimizers) pass. The only residual risk is the entrywise identity in Appendix A that makes ∆ CP. That is finite arithmetic, not a structural hole; a referee or a short script can close it.\n\nCitations are appropriate (Watanabe, HRSF, BKM-RE bridges, RK24 gap). AI use is disclosed and limited. No free parameters, no circular capacity definitions.\n\nThis is for people who care about quantum channel capacities, less-noisy orders, and structural mechanisms for incapacity. It deserves a serious referee. I would bring it to reading group and I would cite it. Send it out.","headline":"Explicit qutrit channel with P=Q=0 outside PPT and antidegradability; the signed-lift criterion looks real and the open problems are actually closed.","tokens_in":21823,"tokens_out":508,"would_cite":true,"duration_ms":12600,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.5","headline":"An explicit qutrit channel has zero private and quantum capacity while being neither PPT nor antidegradable.","keywords":["quantum capacity","private capacity","antidegradable channels","PPT channels","less-noisy order","signed lift","Werner-Holevo channel","coherent information"],"falsifier":"Check the claimed Kraus factorization of the outer defect for the constructed lift J; if that defect is not completely positive, or if some reference-amplified input violates the BKM or relative-entropy comparison, the zero-capacity conclusion fails. Alternatively, exhibit strictly positive private or coherent information for any finite number of uses of Λ.","tokens_in":21201,"feed_emoji":"⚛️","tokens_out":867,"duration_ms":34940,"temperature":0.7,"pith_summary":"Two structural mechanisms have long explained why a quantum channel can have zero quantum capacity: the channel is PPT, so it cannot distill entanglement, or it is antidegradable, so the environment can fully simulate the receiver. This paper gives a concrete three-level channel that escapes both mechanisms yet still cannot send private or quantum information at any asymptotic rate. The channel is a fixed noisy Werner–Holevo map on qutrits. The authors prove that its optimized private and coherent information vanish for every number of uses, so both capacities are exactly zero, while the Choi state is NPT and one-copy distillable and no antidegrading map exists. The work answers two open existence questions and isolates a weaker, observable-level comparison that is still strong enough to force zero capacity.","feed_headline":"Qutrit channel hits zero capacity outside known traps","feed_subtitle":"A weaker observable-level shadow of the receiver still forces private and quantum rates to vanish","key_machinery":"A complete variance-dominating signed lift: an adjoint-preserving, generally nonpositive map J that sends each receiver observable to an environment observable with the same mean and no larger variance, stably under arbitrary reference systems. Complete positivity of its outer defect certifies the comparison, which upgrades through BKM metrics to the complete less-noisy order of the complement and therefore to zero private and quantum information at every blocklength.","core_discovery":"For the explicit qutrit channel Λ(X)=½X+¼(Tr(X)1−Xᵀ), the optimized private and coherent information of every tensor power vanish, so P(Λ)=Q(Λ)=0, even though the Choi state is not PPT and the channel is not antidegradable.","pith_inferences":["The same signed-lift certificate may close other open low-dimensional capacity gaps where PPT and antidegradability both fail.","The stated hierarchy invites a search for channels that satisfy complete less-noisy order of the complement yet fail variance domination, separating the information orders further.","Observable-by-observable shadow reconstruction could serve as a practical zero-private-capacity diagnostic when full environment-to-receiver simulation is unavailable.","One-copy distillability of an NPT Choi state with vanishing one-way distillable entanglement points to a sharp one-way versus two-way separation worth checking on related families."],"forward_implications":["Zero quantum capacity can occur outside the PPT and antidegradable classes.","Antidegradability is not the only nontrivial mechanism that forces private capacity to vanish.","Complete variance-dominating signed lifts strictly contain antidegradable channels and still imply P=Q=0.","The complete less-noisy order of the complement can hold without any physical simulation of the receiver.","At this noisy Werner–Holevo parameter both capacities are settled at zero for all blocklengths."],"fun_headline_variants":["Qutrit channel with P=Q=0 outside PPT and antidegradable classes","Zero private and quantum capacity without PPT or no-cloning","Explicit qutrit map forces vanishing capacities beyond known mechanisms","Observable-level relaxation yields zero-capacity qutrit channel","New zero-capacity class: neither PPT nor antidegradable yet P=Q=0"],"cache_read_input_tokens":16512,"weakest_assumption_plain":"The argument needs that unbiased, reference-stable variance domination by a possibly nonpositive observable map is enough to dominate relative-entropy Hessians and hence relative entropies themselves.","fun_headline_variants_meta":{"raw":{"variants":["Qutrit channel with P=Q=0 outside PPT and antidegradable classes","Zero private and quantum capacity without PPT or no-cloning","Explicit qutrit map forces vanishing capacities beyond known mechanisms","Observable-level relaxation yields zero-capacity qutrit channel","New zero-capacity class: neither PPT nor antidegradable yet P=Q=0"]},"model":"grok-4.5","effort":"low","cost_usd":0.003735,"raw_usage":{"total_tokens":1237,"prompt_tokens":813,"num_sources_used":0,"completion_tokens":98,"cost_in_usd_ticks":37348000,"prompt_tokens_details":{"text_tokens":813,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":326,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":813,"tokens_out":98,"duration_ms":6227,"temperature":1.0,"reasoning_tokens":326,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-31T07:40:04.241587+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Check the claimed Kraus factorization of the outer defect for the constructed lift J; if that defect is not completely positive, or if some reference-amplified input violates the BKM or relative-entropy comparison, the zero-capacity conclusion fails. Alternatively, exhibit strictly positive private or coherent information for any finite number of uses of Λ.","supporting_citations":[],"review_version":1}