{"id":"429ca43d-6698-4f17-b4b3-d80af62bceb1","arxiv_id":"2507.08781","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Every N-party quantum circuit with quantum control of causal order can be represented as a routed quantum circuit built from one fixed routed graph G_QC-QC(N).","lead":"This paper proves that every quantum circuit with quantum control of causal order (QC-QC) can be written as a routed quantum circuit, using a single universal routed graph for any number of parties. The authors give an explicit construction and show that the two frameworks describe the same processes for all N.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the proof of Theorem 4.1 is internally consistent, with the main residual risk being the unproved external validity theorem (Theorem 2.23 of Ref. [39]).","rationale":"The reader's weakest assumption identifies the same external-theorem dependency. My stress-test found no additional load-bearing concern: the bi-univocality computation, branch-graph loop analysis, and process-vector equality are all explicit and internally consistent. The mixed-case extension by tracing out alpha_F is a standard final trace and does not affect the routed graph's validity. Thus the reader's ACCEPT with moderate confidence is appropriate; I would not adjust the verdict.","tokens_in":62257,"tokens_out":31809,"duration_ms":375166,"concrete_test":"For the smallest nontrivial case N=2, compute the skeletal supermap of G_QC-QC(2) directly from Definition 2.21 and verify by hand or symbolically that plugging arbitrary routed unitaries into the V2 and A1/A2 slots yields a unitary map, without invoking Theorem 2.23; compare the resulting process to the known quantum switch decomposition of Ref. [39]. If unitarity fails, Theorem 2.23 is being misapplied.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I read the construction in Sections 4.1-4.3 and Appendix B as a direct, checkable proof: the generic graph is shown to be bi-univocal, its branch graph has no loops, the V-node and A-node fleshings follow the claimed routes, and the link-product computation in Eq. (4.34) reproduces the QC-QC process vector. I found no internal algebraic or graphical gap. The single load-bearing assumption is the external Theorem 2.23, which certifies that a valid routed graph yields a routed superunitary. This paper restates the routed-circuit framework in a slightly simplified form (Definitions 2.4-2.22) and applies Theorem 2.23 to G_QC-QC(N) without re-proving it or checking that every hypothesis of the original theorem is satisfied under the restatement. If that theorem were false or inapplicable, the skeletal supermap would not be guaranteed to define a valid quantum process. This is an inherited, reasonable prior-result risk rather than an internal defect; the explicit equality to |w_QC-QC> provides independent evidence for the concrete construction.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The manuscript connects the routed quantum circuit framework of Vanrietvelde et al. with the QC-QC framework of Wechs et al. It proposes, for each N, a generic routed graph G_QC-QC(N) and proves that every N-slot QC-QC, pure or mixed, can be obtained by fleshing out the associated skeletal supermap. Sections 2 and 3 review both frameworks; Section 4 gives the construction, proves bi-univocality, checks that the branch graph is loop-free, and computes the Choi vector of the fleshed-out supermap, showing that it equals the QC-QC process vector. Section 5 discusses alternative routed graph descriptions, including recovery of earlier decompositions of the quantum switch and the Grenoble process. Appendices A-D contain the detailed calculations behind the main construction.","tokens_in":62483,"tokens_out":40533,"duration_ms":457720,"significance":"If the central result holds, this is a substantial step: it upgrades the routed circuit framework from a handful of examples to the entire QC-QC class, and it provides a concrete template for studying subclasses and open problems. The proof is constructive and largely self-contained, with explicit derivations of the choice function (Appendix A), branch graph acyclicity (Section 4.2), and the equivalence of the fleshed-out circuit to |w_QC-QC> (Appendix B). No fitted parameters or circular assumptions are involved. The main residual risk is the reliance on the external validity theorem (Theorem 2.23 of Ref. [39]) for the routed-graph-to-routed-superunitary step; the direct equality in Eq. (4.34) provides independent evidence for the concrete construction. Overall, this is a strong and useful contribution to the literature on indefinite causal order.","major_comments":[],"minor_comments":[{"comment":"The paper relies on Theorem 2.23 of Ref. [39] to certify that a valid routed graph yields a routed superunitary, but the framework presented here is a simplified variant of the original one. Please add an explicit remark stating that Theorem 2.23 is quoted from Ref. [39] and confirming that the simplified definitions in Section 2 satisfy the hypotheses of the original theorem, or provide a short proof sketch of the transfer. This would remove a residual unclarity about the central validity certificate.","section":"Section 2.6, Theorem 2.23"},{"comment":"Bi-univocality of the adjoint graph is justified by a compact \"self-adjoint up to relabelling\" argument. Since univocality of the adjoint is load-bearing for bi-univocality, please give the explicit involution on nodes, arrows, and index values that maps G_QC-QC(N) to its adjoint, and indicate why the associated choice relations are isomorphic. The current presentation is plausible but too terse for a formal proof.","section":"Section 4.1.4"},{"comment":"The notation (H_{\\bar\\alpha}\\otimes H_{\\bar C(k)})_{\\mathrm{prac}} and the \"dropping\" of one-dimensional tensor factors in the practical subspaces deserve a clearer formal statement. Please specify explicitly that the identification is via the canonical isomorphism between a Hilbert space and its tensor product with one-dimensional spaces, and that this identification is compatible with the route relations at the corresponding nodes.","section":"Section 4.3.3, Eq. (4.30)"},{"comment":"When direct arrows V_n \\to V_{n+1} are added to obtain G^\\alpha_{\\mathrm{QC-QC}}(N), the routes at the affected V-nodes are not explicitly updated. Please specify how the added index value \\alpha_n enters the domain and codomain of \\lambda_{V_n} and \\lambda_{V_{n+1}}, and why the resulting routed graph remains valid under the stated conventions.","section":"Section 5.1"},{"comment":"The proof of the forward direction of Proposition 5.6, especially the G_{\\mathrm{merge}\\downarrow} case, is sketched rather than fully argued. In particular, the claim that the restriction to a routed unitary at V is \"without loss of generality\" needs a more explicit construction showing how an arbitrary routed isometry at V is recovered from a routed unitary and a state preparation at the merged node, without violating the no-ancilla convention of Definition 2.26.","section":"Section 5.3, Proposition 5.6"},{"comment":"There are several minor typographical and formatting issues, including \"admissable\" for \"admissible\" in Section 6.3 and various LaTeX rendering artifacts in equations. A careful copyedit would improve readability, especially in the appendices and figure captions.","section":"Section 6.3 and general presentation"}],"recommendation":"minor_revision","confidential_remarks":"The paper is a solid contribution and I do not see an internal gap in the proof of Theorem 4.1. The main caveat is the reliance on the external routed-circuit validity theorem of Ref. [39], which I regard as an acceptable prior-result dependency, but the authors should make the transfer from their simplified definitions to that theorem explicit. The minor comments above are local and should be addressable without changing the central claims."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Short version: the paper proves the general theorem it advertises. For any N, the single routed graph G_QC-QC(N) of Definition 4.2, fleshed out with the QC-QC's own internal isometries and a fixed adapter for the agent slots, reproduces the process vector of any N-slot QC-QC. That was an open gap: previously only the quantum switch and the Grenoble process had routed decompositions. This is a genuine structural advance, not a routine extension. The construction is also economical—one graph, one fleshing-out template—and the appendices do the work: Appendix A computes the choice function, Section 4.2 proves the branch graph has no loops, and Appendix B chases the link product to arrive at |w_QC-QC>. I read the central chain of Sections 4.1–4.3 and Appendix B as internally consistent. No fitted parameters, nothing circular.\n\nWhere are the soft spots? The main one is the one the reader flags: the proof leans on Theorem 2.23 of Vanrietvelde et al. [39], which certifies that a valid routed graph yields a routed superunitary. The paper restates the routed-circuit framework in a slightly simplified form, then applies that theorem to G_QC-QC(N) without reproving it or checking each hypothesis in the restatement. That is an inherited, reasonable prior-result risk rather than an internal defect; the explicit link-product equality provides independent evidence for the concrete construction. If someone ever found a flaw in [39]'s main theorem, the present result would need revisiting. I'd call that a moderate caveat, not a red flag.\n\nSection 5 is broader and looser—alternate graphs, node merging, arrow removal—but it is clearly exploratory and is flagged as such. The discussion in Section 6 is appropriately careful: it does not oversell the consequences for causal inequalities or spacetime implementability, and it states what remains open.\n\nWho is this for? Anyone working on indefinite causal order, especially the QC-QC and routed-circuit communities. It is a solid paper for a serious referee, despite my own caveat; the proven theorem is the kind of thing the field needs. I'd send it to review.","headline":"This paper delivers a real structural result—every N-slot QC-QC has a routed circuit decomposition from one generic graph—and the proof is detailed enough to check; the main caveat is inherited reliance on the routed-circuit framework's central theorem.","tokens_in":62986,"tokens_out":2043,"would_cite":true,"duration_ms":22396,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that every quantum circuit with quantum control of causal order can be obtained by fleshing out one generic routed graph per number of parties, giving all QC-QCs a routed circuit decomposition.","keywords":["indefinite causal order","quantum control of causal order","routed quantum circuits","skeletal supermap","branch graph","quantum switch","process matrix"],"falsifier":"Take a specific $N$-slot QC-QC specified by its internal isometries, compute the Choi vector obtained by the paper's fleshing out of $G_{\\mathrm{QC-QC}}(N)$ with those isometries, and compare it with the QC-QC process vector computed directly from the definition; a single mismatch would disprove Theorem 4.1. A more pointed test is the appendix's Zurich process for $N=4$, which the paper shows cannot be represented by the compact internal-node-free graph, so checking whether the full generic graph does represent it exercises exactly the part of the claim that goes beyond earlier examples.","tokens_in":62087,"feed_emoji":"🔀","tokens_out":8349,"duration_ms":91651,"temperature":0.7,"pith_summary":"Quantum processes with indefinite causal order have been studied through two complementary lenses: quantum circuits with quantum control of causal order (QC-QCs), built bottom-up from physically motivated components, and routed quantum circuits, which decompose a process into a graph whose routes track where information can flow. Until now, only a handful of specific processes had been shown to admit a routed circuit decomposition. This paper proves that for any number $N$ of parties, a single routed graph, denoted $G_{\\mathrm{QC-QC}}(N)$, can be fleshed out to produce any $N$-slot QC-QC, pure or mixed. If correct, the result makes routed circuits a general representation language for the entire QC-QC family, and gives a diagrammatic picture of the information flow inside these processes.","feed_headline":"One routed graph captures every N-party QC-QC","feed_subtitle":"A single generic graph per party number can be fleshed out to reproduce any QC-QC process.","key_machinery":"The load-bearing object is the generic routed graph $G_{\\mathrm{QC-QC}}(N)$ of Definition 4.2: a directed graph whose nodes are the $N$ agent slots $A_k$ and the $N+1$ internal-operation slots $V_{n+1}$, with arrows labelled by index values $X^k_n$ that are either a subset $K_n$ of agents containing $k$ or the null value. A route at each node is a relation constraining which index combinations may pass through that node, and the global index constraints require that a consistent assignment encodes one full ordering of the agents. The argument then shows the graph is valid, meaning bi-univocal with no non-weak loops in its branch graph, so the routed-circuit framework's main theorem certifies that its skeletal supermap is a routed superunitary. Fleshing out means filling each slot with a compatible routed operation; the paper's specific fleshing out uses isomorphisms $J^{\\mathrm{in/out}}$ to translate between the graph's sectorised wires and the QC-QC's systems, and then plugs in the internal isometries and one-slot agent supermaps.","core_discovery":"The paper's central claim is Theorem 4.1: any $N$-slot quantum circuit with quantum control of causal order (QC-QC) can be implemented by fleshing out the skeletal supermap associated with the generic routed graph $G_{\\mathrm{QC-QC}}(N)$. The graph has two kinds of nodes, agent slots $A_k$ and internal isometry slots $V_{n+1}$, and its arrows carry index values $X^k_n$ that record which subset of agents has already acted and which agent acts at time $n$. The routes at the nodes enforce that each agent acts exactly once and exactly one agent acts per time slot. The paper shows this graph is bi-univocal and its branch graph has no loops, so the routed-circuit framework's main theorem applies and the skeletal supermap is a routed superunitary. The authors then give explicit isomorphisms that insert the QC-QC's internal isometries and agent operations into the slots and compose the resulting Choi vectors with the link product, recovering exactly the QC-QC process vector; tracing out a final ancillary system extends the argument to mixed QC-QCs.","pith_inferences":["If the theorem holds, the subclass structure of QC-QCs, such as superpositions of fixed orders, classical control, and non-influenceable causal orders, could be read off as additional index constraints on the same routed graph, a programme the paper sketches but does not complete.","Combined with the belief that QC-QCs exhaust the processes realisable in a fixed spacetime, the result lends support to the conjecture that every such process has a routed circuit decomposition from a valid routed graph.","The index bookkeeping in $G_{\\mathrm{QC-QC}}(N)$ suggests a design test for proposed causal-control processes: a structure is routable if its causal bookkeeping can be encoded as index constraints on a valid routed graph, which may help in searching for new indefinite-causal-order processes."],"forward_implications":["Every pure $N$-slot QC-QC, not just previously studied examples, acquires a routed circuit decomposition based on the single graph $G_{\\mathrm{QC-QC}}(N)$.","Mixed QC-QCs are covered too: tracing out a final ancillary subsystem of the global future space turns the pure construction into the mixed process matrix.","The validity of $G_{\\mathrm{QC-QC}}(N)$ independently certifies, through the routed-circuit framework, that QC-QCs are legitimate quantum processes.","The alternative graphs obtained by splitting and merging nodes recover the earlier routed decompositions of the quantum switch and the Grenoble process as special cases of the generic construction.","For more than three parties, the fully compact representation without internal V-nodes is generally impossible, so the generic graph is not just a convenience but a necessary level of description."],"supporting_citations":[{"why":"Supplies the main theorem that a valid routed graph yields a skeletal supermap that is a routed superunitary, which the paper's construction invokes to certify validity.","marker":"[39]"},{"why":"Defines quantum circuits with quantum control of causal order and gives the internal-isometry decomposition that the fleshing out reproduces.","marker":"[34]"},{"why":"Introduces routed linear maps and routed circuits, the compositional machinery underlying the skeletal supermap and its fleshing out.","marker":"[43]"},{"why":"Provides the pure link product used to compose Choi vectors when reconstructing the QC-QC process vector from the fleshed-out graph.","marker":"[45]"}],"fun_headline_variants":["One routed graph per N covers every QC-QC","Generic routed graph reproduces any QC-QC","Single skeleton routes all N-party QC-QCs","Every QC-QC arises from one routed graph","All N-party QC-QCs share a routed pattern"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the routed-circuit framework's guarantee that a graph satisfying its validity conditions, namely bi-univocality and only harmless loops, yields a valid quantum process; the paper relies on that guarantee, and on the standard decomposition of a QC-QC into internal isometries, without reproving either.","fun_headline_variants_meta":{"raw":{"variants":["One routed graph per N covers every QC-QC","Generic routed graph reproduces any QC-QC","Single skeleton routes all N-party QC-QCs","Every QC-QC arises from one routed graph","All N-party QC-QCs share a routed pattern"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000944,"raw_usage":{"total_tokens":4058,"prompt_tokens":1000,"completion_tokens":3058,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":616,"completion_tokens_details":{"reasoning_tokens":2985}},"tokens_in":616,"tokens_out":3058,"duration_ms":25546,"temperature":1.0,"reasoning_tokens":2985,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-06T18:09:07.952098+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a specific $N$-slot QC-QC specified by its internal isometries, compute the Choi vector obtained by the paper's fleshing out of $G_{\\mathrm{QC-QC}}(N)$ with those isometries, and compare it with the QC-QC process vector computed directly from the definition; a single mismatch would disprove Theorem 4.1. A more pointed test is the appendix's Zurich process for $N=4$, which the paper shows cannot be represented by the compact internal-node-free graph, so checking whether the full generic graph does represent it exercises exactly the part of the claim that goes beyond earlier examples.","supporting_citations":[{"cited_title":"Quantum cir- cuits with classical versus quantum control of causal order","cited_arxiv_id":null,"evidence_quote":"Defines quantum circuits with quantum control of causal order and gives the internal-isometry decomposition that the fleshing out reproduces."},{"cited_title":"Routed quantum circuits","cited_arxiv_id":null,"evidence_quote":"Introduces routed linear maps and routed circuits, the compositional machinery underlying the skeletal supermap and its fleshing out."}],"review_version":1}