{"id":"c353175f-046e-408c-ba6b-a39ba47e7261","arxiv_id":"2512.16979","paper_version":3,"verdict":"REJECT","confidence":"HIGH","novelty_score":7.0,"correctness_risk":"high","formal_verification":"none","parameter_count":4,"one_line_summary":"A proposed condition for equal entanglement spectra across bipartitions is invalid as stated; crossed quotient equality provides a counterexample, though an ordered version might repair the approach.","lead":"This paper proposes a quotient-set condition under which different bipartitions of a quantum state in a constrained subspace would have identical entanglement spectra, plus a polynomial-time check for parity-encoded qubits. The central theorem as stated is false: crossed quotient equality can hold while the spectra differ, so the bundling proof fails.","discovery_kind":"first_principles","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Mixed-state extension in Theorem 3 is false; crossed ∼R cases fail for mixed states, though pure Theorem 1 may be salvageable.","rationale":"The reader correctly identifies the unordered-pair/WLOG step in the proof of Theorem 3(i) as the weak point, but the specific H_3 counterexample they give (A1={1}, A2={2}) does not satisfy A1∼R A2: the quotient partitions induced by qubit 1 and qubit 2 are different partitions, and the unordered pairs are not equal. The genuine crossed failure occurs when A2 is the complement of A1 (up to quotient equality), e.g. R=H_3, A1={1}, A2={2,3}. In that case A1∼R A2 holds, and the mixed state ρ=(|000⟩⟨000|+|100⟩⟨100|)/2 has Spec(ρ_{1})={1/2,1/2} but Spec(ρ_{23})={1,0}. This is a clean, explicit contradiction to Theorem 3. The proof's WLOG cannot be justified for mixed states because the pure-state relation Spec(ρ_A)=Spec(ρ_A^c) does not extend to mixed states. For pure states, the crossed case can be reduced to the ordered case by replacing A1 with its complement and using Schmidt symmetry, so Theorem 1 may survive after revision. The numerical bundling observation and the polynomial-time parity-embedding verification appear independent of the false mixed-state extension and should not be discarded. The paper should therefore not be rejected outright if the authors are willing to restrict the main statement to pure states, remove or correct the mixed-state claim, and repair the proof of Theorem 3. This makes the appropriate verdict conditional rather than a full reject.","tokens_in":26132,"tokens_out":28320,"duration_ms":239395,"concrete_test":"Verify the counterexample numerically: set n=3, R=H_3, A1={1}, A2={2,3}, and ρ=(|000⟩⟨000|+|100⟩⟨100|)/2. Compute Spec(tr_{A2}ρ) and Spec(tr_{A1}ρ); they are {1/2,1/2} and {1,0}, contradicting Theorem 3. Independently check A1∼R A2 by enumerating R/∼A1, R/∼A1^c, R/∼A2, R/∼A2^c. Then rerun the same check with a pure state, e.g. |Ψ⟩=(|000⟩+|100⟩)/√2, to confirm Theorem 1 still holds in this crossed case; if it does, the required revision is to restrict the mixed-state claim or introduce an explicitly ordered equivalence relation.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The proof of Theorem 3(i) reduces the unordered-pair condition A1∼R A2 to the ordered case R/∼A1=R/∼A2 'without loss of generality.' This reduction is valid for pure states only, because for pure states Spec(ρ_A)=Spec(ρ_A^c) by Schmidt symmetry. For mixed states it is not valid. The crossed case is realized with R=H_3, A1={1}, A2={2,3}=A1^c: the unordered pairs are both {R/∼{1}, R/∼{2,3}}, so A1∼R A2. But for ρ=(|000⟩⟨000|+|100⟩⟨100|)/2 ∈ Mix(Q_R), Spec(ρ_{A1})={1/2,1/2} while Spec(ρ_{A2})={1,0}. Thus Theorem 3 as stated is false. The reader's illustration with A1={1},A2={2} in H_3 is not actually a ∼R pair; the quotient partitions differ. The correct crossed counterexample is A2=A1^c. The pure-state claim, Theorem 1, can likely be repaired by applying the same-order argument to A1 and A2^c and then using Schmidt symmetry, so the central bundling result for pure states is not necessarily dead.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies bipartite entanglement in constrained subspaces spanned by subsets R of the computational basis. It defines an equivalence relation A1 ∼R A2 via equality of the unordered pair of quotient sets {R/∼A1, R/∼A1c} and {R/∼A2, R/∼A2c}, and claims (Theorem 1) that for every pure superposition in Q_R the reduced density matrices ρ_A1 and ρ_A2 have identical spectra, so all spectral entanglement measures are equal. Theorem 3 extends the claim to all mixed states in Mix(Q_R). The authors also provide an operator-based reformulation (Theorem 2), apply it to the parity and minor embeddings, and give a polynomial-time verification algorithm for the parity embedding. Numerical simulations of annealing processes are presented as evidence of the 'bundling' phenomenon.","tokens_in":26549,"tokens_out":16497,"duration_ms":150074,"significance":"If the pure-state part were proved correctly, the paper would offer a useful structural criterion for when entanglement spectra are identical across bipartitions in constrained Hilbert spaces, with a genuinely non-trivial algorithmic result for the parity embedding. The operator-based framework and the explicit verification algorithm are valuable and appear to be independent of the mixed-state issue. However, the advertised mixed-state generalization is false, and the proof of the pure-state theorem contains a gap that must be repaired. The paper therefore cannot be accepted in its present form; the central claim is defensible only after substantial correction and restriction.","major_comments":[{"comment":"Theorem 3 is false as stated. Take n=3, R=H_3, A1={1}, A2={2,3}=A1^c. Then A1∼R A2 because the unordered pair {R/∼A1, R/∼A1^c} equals {R/∼A2, R/∼A2^c} by complementation. For ρ=(|000⟩⟨000|+|100⟩⟨100|)/2 ∈ Mix(Q_R), ρ_A1=(|0⟩⟨0|+|1⟩⟨1|)/2 has spectrum {1/2,1/2}, while ρ_A2=|00⟩⟨00| has spectrum {1,0}. Thus Spec(ρ_A1)≠Spec(ρ_A2). The proof's 'without loss of generality' reduction to the ordered case is invalid for mixed states, since pure-state Schmidt symmetry cannot be invoked. This is load-bearing: the abstract and conclusion advertise equal spectra for all mixed states in the subspace.","section":"Sec. V A, Theorem 3"},{"comment":"Even for the pure-state Theorem 1, the opening 'Without loss of generality we may assume R/∼A1=R/∼A2' is not justified by Def. III.2. From A1∼R A2 one may be in the crossed case R/∼A1=R/∼A2^c and R/∼A1^c=R/∼A2. For pure states the crossed case can be handled by applying the ordered argument to A1 and A2^c and then using Spec(ρ_A)=Spec(ρ_A^c), but this step is absent. The manuscript should either supply this argument or explicitly restrict the theorem to the ordered-equality condition.","section":"Sec. V A, proof of Theorem 3(i)"}],"minor_comments":[{"comment":"The text says 'quotient groups' but should read 'quotient sets'; the same typo appears in the table below the definition.","section":"Sec. III B, after Def. III.2"},{"comment":"In the final paragraph, the sentence 'If H|B is disconnected, it is sufficient to consider the case where the vertex set of H|B equals V...' appears twice in slightly different forms; the second occurrence should presumably refer to the connected case.","section":"Sec. V B, Example 1 proof"},{"comment":"The phrase 'such subspaces are typical exponentially large' is loose: R is a set of computational basis states, not itself a subspace. Rewording would avoid confusion.","section":"Sec. III B"},{"comment":"The complexity statements would be clearer if the paper distinguished between the general verification cost O(n|R|^2) and the parity-embedding cost O(|V||E|min(|V|,|E|)), and stated precisely which operations are counted.","section":"Sec. IV B"}],"recommendation":"major_revision","confidential_remarks":"The pure-state bundling result and the parity-embedding algorithm have real merit, but the false mixed-state theorem is a serious error that must be addressed. In my view the paper is not publishable as is; it would require restricting all claims to pure states (or adding an explicit ordered-equality condition), correcting the WLOG step in the proof of Theorem 1, and adjusting the abstract and conclusions accordingly. If the authors are unwilling to make that restriction, the manuscript should be rejected."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper's real contributions are the quotient-set equivalence relation on bipartitions, the operator-based reformulation, and the polynomial-time verification algorithm for parity embeddings. The algorithm for checking whether two bipartitions are equivalent without enumerating the exponentially large parity subspace is a concrete, useful tool. The numerical bundling plots are also honest examples of the phenomenon. That part deserves a serious referee.\n\nThe main theorem, however, is stated too broadly. Theorem 3 claims that if A1 ~R A2 then the reduced spectra are equal for all mixed states in Mix(Q_R). That is false. Take R = H_3, A1 = {1}, A2 = {2,3} = A1^c. Then A1 ~R A2 by definition, but the mixed state (|000><000| + |100><100|)/2 has Spec(rho_{A1}) = {1/2,1/2} and Spec(rho_{A2}) = {1,0}. The proof's \"without loss of generality\" step, where the unordered pair equality is read as same-order equality R/~A1 = R/~A2, is exactly where this breaks: it is not valid for mixed states. The stress-test note is right that the reader's own counterexample with A1={1}, A2={2} does not actually satisfy A1 ~R A2, so that specific complaint is off.\n\nThe pure-state version, Theorem 1, is more likely to be true. If the unordered pair matches in the crossed order, then R/~A1 = R/~A2^c and R/~A1^c = R/~A2; applying the same-order argument to A1 and A2^c and then using Schmidt symmetry should go through. But the proof as written does not make this case, so the gap is real even if the result is repairable.\n\nThere are also smaller soft spots: the minor-embedding section is more sketchy, and the converse direction in Theorem 1 is interesting but not central. None of those are fatal.\n\nBottom line: this is a paper with a new framework and a useful algorithm that is currently attached to an overclaimed theorem. A serious referee would catch the mixed-state failure, but the pure-state bundling result and the verification algorithm are worth engaging with. I would send it to peer review with a request to either restrict the theorem to pure states or fix the mixed-state argument, and to repair the WLOG in the proof. The framework deserves another round.\n\nReading group: maybe. Cite: only after the repair is published.","headline":"Interesting framework and a genuinely useful polynomial-time verification algorithm, but the mixed-state theorem is false as stated; the pure-state result may survive a repair.","tokens_in":26923,"tokens_out":3097,"would_cite":false,"duration_ms":31227,"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":"In constrained energy subspaces, many distinct bipartitions of a quantum system provably share the same entanglement spectrum.","keywords":["entanglement spectrum","bipartite entanglement","constrained subspace","parity embedding","quantum annealing","entanglement bundling","reduced density matrix","equivalence relation"],"falsifier":"Take R to be all 3-qubit computational basis states, A1={1}, A2={2}, and |Ψ> = sqrt(0.6)|000> + sqrt(0.1)|100> + sqrt(0.3)|111>. Then A1 ~R A2 holds (both bipartitions give quotient-class counts 2 and 4), but Spec(ρ_A1) = {0.6, 0.4} while Spec(ρ_A2) = {0.7, 0.3}. Showing this difference would contradict Theorem 1 as stated; resolving it requires the ordered-equality assumption.","tokens_in":26063,"feed_emoji":"🔗","tokens_out":9795,"duration_ms":84056,"temperature":0.7,"pith_summary":"The paper tries to establish a structural fact about bipartite entanglement when a quantum state is confined to a subspace spanned by a fixed set of computational basis states: if two subsystems cut the basis set in the same 'pattern' of restrictions, then the reduced density matrices of the two subsystems have identical eigenvalues for every state in the subspace. This makes entanglement entropies of many different bipartitions 'bundle' to a single value under unitary evolution, which the authors observe in quantum annealing and QAOA simulations and then prove theoretically. The proof rests on an equivalence relation that compares, for each subsystem, how the basis states group when restricted to the subsystem and to its complement. As a practical payoff, for the parity embedding used in quantum optimization the paper derives an operator-based characterisation and a polynomial-time algorithm that decides whether two bipartitions belong to the same bundle.","feed_headline":"Entanglement spectra bundle inside constrained energy subspaces","feed_subtitle":"Many qubit cuts lock to one entropy value in annealing and QAOA; bundle membership can now be checked quickly.","key_machinery":"The load-bearing object is the equivalence relation ~R on subsystems, defined by equality of the unordered pair of quotient sets R/~A and R/~A^c. R is the set of computational basis states generating the constraint subspace; two states in R are in the same class of ~A if their restrictions to A coincide. The relation is sufficient for spectral equality, and Theorem 2 translates it into equality of bipartite operator sets O_A, O_{A^c} built from a commuting generator set of product operators. The operator formulation enables the polynomial-time parity-embedding algorithm; direct checking of ~R costs O(n|R|^2) and is generally exponential.","core_discovery":"The central claim, Theorem 1, is that for any two non-trivial subsystems A1 and A2 with A1 ~R A2, the entanglement spectra Spec(ρ_A1) and Spec(ρ_A2) coincide for every pure state—and, by Theorem 3, every mixed state—in the subspace spanned by R. The equivalence A1 ~R A2 holds exactly when the pair of quotient sets {R/~A1, R/~A1^c} equals the pair {R/~A2, R/~A2^c}, where two basis states are identified under ~A if their restrictions to A coincide. The theorem extends the elementary fact that A and its complement have identical spectra, and it implies equal values for every spectral entanglement measure. The converse is deliberately one-way: equal spectra for all states does not force A1 ~R A2","pith_inferences":["Editorial inference: the proof of Theorem 3 can be repaired by making the equivalence relation order-sensitive (R/~A1 = R/~A2 and R/~A1^c = R/~A2^c); the current unordered statement overreaches in crossed cases, but the bundle phenomenology and the parity-embedding algorithm do not obviously depend on those cases.","Editorial inference: the same quotient-set criterion applies to any basis-set-restricted subspace, so it should transfer to symmetry-restricted or gauge-invariant sectors of other many-body models, not just optimization embeddings.","Editorial inference: because the entanglement spectrum is experimentally measurable, bundle classes with trivial operator sets could be used as built-in 'spectrometers' to read off the weights of basis states in a superposition—a use the paper mentions only as a side remark."],"forward_implications":["All entropy-based bipartite entanglement measures—von Neumann, Rényi, negativity, and others—take identical values on every bipartition in the same bundle.","For the parity embedding, equivalence of two bipartitions can be verified in polynomial time even though the constrained subspace is exponentially large.","In embedded quantum optimization, entanglement dynamics of different bipartitions coincide throughout the evolution whenever constraints keep the state inside the subspace.","Measuring the entanglement spectrum of one member of a bundle gives the spectrum of every member, potentially reducing experimental measurement effort.","For a 'double spanning tree' bipartition, the reduced density matrix spectrum directly reveals the probabilities of the basis states in the superposition."],"fun_headline_variants":["Bipartite entanglement spectra bundle in energy subspaces","All bipartitions share one entanglement spectrum in constraint subspace","Entanglement bundles: identical spectra across bipartitions in subspace","Polynomial-time check for identical entanglement spectra in parity subspace"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The proof of Theorem 3 assumes, without loss of generality, that when the two unordered pairs of quotient sets coincide, they do so in matching order (R/~A1 = R/~A2 and R/~A1^c = R/~A2^c); crossed cases are not covered and can break the spectral equality.","fun_headline_variants_meta":{"raw":{"variants":["Bipartite entanglement spectra bundle in energy subspaces","All bipartitions share one entanglement spectrum in constraint subspace","Entanglement bundles: identical spectra across bipartitions in subspace","Polynomial-time check for identical entanglement spectra in parity subspace"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000193,"raw_usage":{"total_tokens":1135,"prompt_tokens":638,"completion_tokens":497,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":382,"completion_tokens_details":{"reasoning_tokens":432}},"tokens_in":382,"tokens_out":497,"duration_ms":5262,"temperature":1.0,"reasoning_tokens":432,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T15:25:40.028370+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take R to be all 3-qubit computational basis states, A1={1}, A2={2}, and |Ψ> = sqrt(0.6)|000> + sqrt(0.1)|100> + sqrt(0.3)|111>. Then A1 ~R A2 holds (both bipartitions give quotient-class counts 2 and 4), but Spec(ρ_A1) = {0.6, 0.4} while Spec(ρ_A2) = {0.7, 0.3}. Showing this difference would contradict Theorem 1 as stated; resolving it requires the ordered-equality assumption.","supporting_citations":[],"review_version":1}