{"id":"d40fe19d-9273-4a12-81eb-c535a3294d79","arxiv_id":"2601.22911","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"In semiadditive CD categories, the abstract Metropolis–Hastings kernel is reversible exactly when the balancing condition α=(α∘φ)∗r holds μ-almost everywhere, recovering and conversing the involutive-MH theorem.","lead":"This paper recasts the Metropolis–Hastings algorithm in categorical probability, introducing commutative-monoid-enriched 'semiadditive CD categories' and proving an abstract if-and-only-if reversibility condition for an involutive MH kernel. It matters because it gives a unified algebraic language for MCMC correctness that also extends to skew-reversible samplers and adds a converse to the standard theorem.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 1.1 is only recovered in the absolutely continuous case; the acceptance condition α=0 on the singular component S^c is never derived.","rationale":"The reader's weakest_assumption correctly identified that Theorem 4.27 is conditional on the existence of the Radon–Nikodym derivative. My concern sharpens this: even granting that Lebesgue decompositions exist (Corollary 4.44), the paper never proves that the acceptance probability must vanish on the singular component. This is load-bearing because the advertised contribution is a synthetic necessary-and-sufficient condition recovering Theorem 1.1, including its converse. The gap is not an internal inconsistency of the absolutely continuous case—Theorem 4.27 appears sound under its hypotheses—but it means the paper's abstract and conclusion overstate the scope. A conditional acceptance is appropriate: the main theorem can stand as a result about the a.c. case, but the claim to have recovered the full Theorem 1.1 should either be weakened or completed with a singular-component acceptance lemma. The reader's verdict of ACCEPT is therefore too strong as stated.","tokens_in":27822,"tokens_out":12000,"duration_ms":115779,"concrete_test":"In sfKern, take E={0,1}, φ(0)=1, φ(1)=0, and μ=δ_0. Determine whether the paper's categorical machinery (Theorems 4.27, 4.43, 4.44, Propositions 4.25–4.26, Definition 4.38) can be used to prove that any μ-reversible P_MH satisfies α(0)=0. If this singular necessity cannot be derived from the stated results, the full converse of Theorem 1.1 remains unproven; one would expect to find an explicit lemma stating that μ_s⊥φ∘μ_s implies α·μ_s=0 for reversibility, which is absent from Sections 4.5–4.6.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The paper's central claim to recover Theorem 1.1 (and supply a converse) outruns what is proven. Theorem 4.27 gives an iff only under the hypothesis that r=d(φ∘μ)/dμ exists; in sfKern this is the assumption μ^φ≪μ (Corollary 4.28). The genuinely general statement of Theorem 1.1 also treats the singular component, where the acceptance probability must vanish on S^c. Section 4.6.1 does not establish this: Corollary 4.44 constructs the measurable set S such that μ|_S and its pushforward are equivalent and μ|_{S^c} and its pushforward are singular, but no categorical lemma states or proves that μ-reversibility of P_MH forces α=0 on the singular part. The sufficiency of imposing α=0 is trivial, but the necessity direction—part of any full converse—is absent. Thus the 'if and only if' of Theorem 4.27 applies only when the Radon–Nikodym derivative exists globally, and the full balancing condition (2), including the '0 otherwise' clause, is not recovered. This is not a merely technical caveat: for E={0,1}, φ the swap, and μ=δ_0, the classical theorem forces α(0)=0, while the stated categorical theorems do not apply because r does not exist, and no alternative categorical argument is supplied.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops a categorical framework for the Metropolis–Hastings kernel. It first formulates invariance, reversibility, and skew-reversibility in Markov categories, then moves to CD categories for the unnormalised first summand, and finally to CD categories enriched over commutative monoids ('semiadditive CD categories'). In this setting it defines substochastic morphisms, probabilities, cancellative/finite morphisms, absolute continuity via a preorder, singular morphisms via meets, and abstract Lebesgue decompositions. The main result, Theorem 4.27, gives an 'if and only if' for reversibility of the abstract MH kernel P_MH = α·φ + α^c·id: the balancing condition α = (α∘φ)∗r holds μ-a.e., assuming a Radon–Nikodym derivative r = d(φ∘μ)/dμ exists. Instantiating in sfKern gives a converse to the absolutely continuous (quasi-invariant) case of the Andrieu–Lee–Livingstone balancing condition, and Corollary 4.44 constructs the usual absolutely continuous/singular decomposition of μ under an involution. The paper claims to recover Theorem 1.1 and to supply a converse.","tokens_in":28187,"tokens_out":14525,"duration_ms":144734,"significance":"If completed, the paper would make a valuable contribution: it gives a clean synthetic proof of the balancing condition, a converse in the quasi-invariant case, a skew-reversible extension, and a reusable toolkit (CMon-enriched CD categories) for substochastic kernels, absolute continuity, and Lebesgue decompositions. The proofs are detailed and the hypotheses are explicit; the paper also honestly acknowledges in Corollary 4.44 that the existence of Lebesgue decompositions is imported from the classical theorem. No circularity is apparent: the balancing condition is derived from the categorical axioms rather than assumed. The main caveat is that the advertised recovery of Theorem 1.1 is incomplete for the singular component; this is fixable by qualification or by adding a missing argument.","major_comments":[{"comment":"Corollary 4.44 is said to give 'the remaining part of Theorem 1.1', but it only constructs S. It does not prove that μ-reversibility of P_MH forces α=0 on S^c. Theorem 4.27 is an iff only when r=d(φ∘μ)/dμ exists; in sfKern this means μ^φ≪μ, which for an involution φ implies μ≡μ^φ, so S is μ-null. Thus Corollary 4.28 covers only the quasi-invariant case, and the '0 otherwise' clause of (2) is not derived. The sufficiency of α=0 is trivial; the necessity—needed for a full converse—is missing. Example: E={0,1}, φ swap, μ=δ_0; the classical condition requires α(0)=0, but no categorical theorem applies because r does not exist. Please add such an argument or explicitly qualify the claimed recovery of Theorem 1.1.","section":"§4.6.1, Corollaries 4.28 and 4.44"},{"comment":"The conclusion states that the paper gives a 'purely algebraic derivation of their balancing condition in Theorem 1.1'. This is stronger than what is established, since the singular-part clause of the balancing condition is not derived categorically. The theorem statement of Theorem 4.27 and the surrounding discussion should either be expanded to handle the singular component or restricted to the absolutely continuous / quasi-invariant case.","section":"Section 5 / conclusion"}],"minor_comments":[{"comment":"There are several typos: 'already lead to' should be 'already led to'; Corollary 4.28 has 'determinsitic' for 'deterministic'; Proposition 4.25 reads 'If both P and Q and are μ-reversible'.","section":"Introduction / general"},{"comment":"The cancelled summand is described only verbally as 'a composition of a finite morphism and a substochastic morphism'. A displayed string diagram or an explicit reference to Proposition 4.22 would improve readability.","section":"Proof of Proposition 4.26"},{"comment":"The phrase 'restricting to a cancellative μ' is initially opaque before the reader recalls Proposition 4.17. Consider writing 'finite (equivalently, σ-finite) μ' when instantiating in sfKern.","section":"Remark 4.23"}],"recommendation":"major_revision","confidential_remarks":"The core categorical derivation appears sound, and the paper is likely to be a solid contribution after the scope of the main theorem is clarified. The key issue is whether the advertised recovery of Theorem 1.1 is too strong: the singular-component acceptance condition α=0 on S^c is not derived. I would ask the authors either to prove this missing necessity direction or to state explicitly that Theorem 4.27 and its corollaries apply only in the absolutely continuous (quasi-invariant) case."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Worth reading, but say the acceptance condition α=0 on the singular component is not actually proven. The paper builds a genuinely useful new structure—semiadditive CD categories—and gives a clean string-diagram proof of an iff reversibility condition in the absolutely continuous case, plus a skew-reversible extension. That is real progress, and the categorical treatment of absolute continuity, meets, and Lebesgue decompositions is thoughtful and well-integrated with existing literature. The proof of Theorem 4.27 is coherent and the “balancing condition” converse is novel.\n\nThe soft spot is the paper’s claim to recover all of Theorem 1.1 (Andrieu–Lee–Livingstone). Theorem 4.27 and Corollary 4.28 require the Radon–Nikodym derivative r = d(φ∘μ)/dμ to exist, so in sfKern they only cover μ^φ ≪ μ. The paper then uses Section 4.6 to construct the measurable set S via the classical Lebesgue decomposition, and Corollary 4.44 establishes the equivalence/singularity decomposition of μ. But nowhere does it show that μ-reversibility forces α=0 on S^c. That missing necessity direction is part of any full converse. For a concrete example, take E={0,1}, φ the swap, μ=δ_0; the classical condition forces α(0)=0, but the categorical theorems do not apply because r does not exist, and no alternative argument is supplied. This is not a minute technicality—it delimits the main advertised result.\n\nTo be fair, the paper is explicitly honest that it imports the classical Lebesgue decomposition theorem for existence, so there is no circularity or hidden dependence. But the text in the introduction and around Corollary 4.28 overstates what has been proven. The “if and only if” is only for the mutually absolutely continuous case; the singular component remains classical and is only used to show sufficiency, not necessity.\n\nWho gets value: categorical probabilists wanting a working theory of unnormalised kernels and absolute continuity, and MCMC theorists interested in synthetic proofs of reversibility. It deserves a serious referee—the framework is valuable and the main derivation is sharp, even if the singular-part claim needs repair or the framing needs to be scaled back. I would engage with it and cite it, with a caveat about the gap.","headline":"Good categorical framework, but the advertised full recovery of Theorem 1.1 overstates what is proven: the singular-part acceptance condition is never derived.","tokens_in":28586,"tokens_out":2025,"would_cite":true,"duration_ms":22689,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["18M05","60J22"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper establishes that the reversibility of an abstract Metropolis–Hastings kernel is equivalent, in a wide class of categorical probability settings, to one balancing equation holding almost everywhere.","keywords":["Metropolis-Hastings","Markov categories","CD categories","semiadditive categories","reversibility","balancing condition","Radon-Nikodym derivative","Lebesgue decomposition"],"falsifier":"In the category of s-finite kernels, take a finite measure μ and a deterministic involution φ with μ∘φ^{-1} absolutely continuous with respect to μ, choose an acceptance function α satisfying α(ξ)=α(φ(ξ))r(ξ) for μ-almost every ξ, and check detailed balance of the kernel P_MH directly on a generating algebra. If any such α fails to give a μ-reversible kernel, Theorem 4.27 is false; if a reversible kernel exists whose α violates the balancing equation on a set of positive μ×μ measure, the converse is false. A sharper structural test is to construct a finitely cancellative semiadditive CD catego","tokens_in":27757,"feed_emoji":"🎲","tokens_out":8980,"duration_ms":92317,"temperature":0.7,"pith_summary":"The paper asks whether the correctness condition at the heart of Metropolis–Hastings—reversibility with respect to a target distribution—can be proved and characterised using categorical probability instead of measure-theoretic calculation. It shows that in a finitely cancellative semiadditive CD category (a setting with unnormalised kernels, addition of morphisms, and a workable notion of Radon–Nikodym derivative), the involutive MH kernel P_MH = α·φ + (1−α)·id is target-reversible if and only if the acceptance probability satisfies α = (α∘φ) ∗ r almost everywhere, where r is the Radon–Nikodym derivative of the target pushed forward through the involution. This recovers the classical involutive-MH balancing condition and supplies the previously missing converse. The same condition extends to skew-reversible samplers. A sympathetic reader should care because the paper reduces a correctness property of a widely used algorithm to one algebraic identity, showing that categorical tools can carry genuine statistical content.","feed_headline":"A single balancing equation decides Metropolis-Hastings reversibility","feed_subtitle":"A category-theoretic proof recovers the classical acceptance condition and adds the missing converse, for reversible and skew-reversible sam","key_machinery":"The load-bearing construction is the enrichment of CD categories over commutative monoids: hom-sets carry an addition operation, so the full MH kernel can be written as a convex combination of two morphisms. Two further notions carry the proof: finite morphisms, which behave cancellatively under addition, and Radon–Nikodym derivatives as effects between morphisms. Reversibility of the first summand α·φ is proved equivalent to the balancing condition by string-diagram manipulation, while the second summand is trivially reversible and substochastic; finite cancellativity lets the proof cancel it and conclude the equivalence for the whole kernel. Absolute continuity, singular measures, and Lebe","core_discovery":"The central claim, Theorem 4.27, is an if-and-only-if statement: in a finitely cancellative semiadditive CD category, with finite target μ, deterministic involution φ, probability α, and Radon–Nikodym derivative r = d(φ∘μ)/dμ, the abstract Metropolis–Hastings kernel P_MH := α·φ + (1−α)·id is μ-reversible exactly when α = (α∘φ) ∗ r holds μ-almost everywhere. Here CD categories are the categorical framework for unnormalised kernels, and semiadditive means morphisms can be added. In the standard category of s-finite kernels this becomes the familiar statement that the Metropolis–Hastings kernel with involution φ is reversible with respect to μ precisely when its acceptance function obeys α(ξ)=α","pith_inferences":["The result suggests that the balancing equation itself, not any particular balancing function, is the essential design constraint for involutive MH: new acceptance schemes could be obtained by solving α(ξ)=α(φ(ξ))r(ξ) directly.","A natural next step, left open by the paper, is a categorical Radon–Nikodym theorem deriving existence of derivatives and Lebesgue decompositions from order completeness; if such a theorem exists, MH correctness proofs could be fully synthetic.","Because the proof is internal to the category, other concrete finitely cancellative semiadditive CD categories are testable arenas: finding one where the balancing condition fails to characterise reversibility would pinpoint exactly which axioms the classical result depends on.","The preorder-based treatment of absolute continuity might transfer to other MCMC correctness arguments, such as those for continuous-time or Hamiltonian proposals, whenever they can be expressed as involutive kernels."],"forward_implications":["The balancing condition is both necessary and sufficient: any reversible involutive MH kernel in this setting must satisfy it, not merely kernels constructed from a balancing function.","The same single equation characterises skew-reversible MH-type kernels, generalising earlier sufficient conditions and adding a converse.","Substochastic kernels, finiteness, absolute continuity, singularity, and Lebesgue decompositions become algebraic facts about addition and preorders, so the proof transfers to any category with the required structure.","The categorical framework supplies the decomposition and cancellation arguments; for the classical instantiation only the existence of Lebesgue decompositions is imported from measure theory.","Markov-category formulations of invariance, reversibility, and state-space augmentation extend to unnormalised CD settings, offering a unified language for correctness arguments in MCMC."],"fun_headline_variants":["The one equation that makes Metropolis-Hastings reversible","Reversible MH: it all hinges on a single balance equation","Category theory settles the exact condition for MH reversibility","One balance equation decides MH reversibility"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The theorem is conditional on the existence of the Radon–Nikodym derivative r = d(φ∘μ)/dμ in a finitely cancellative semiadditive CD category; the framework itself proves no general existence result, and for the classical instantiation the singular-part statement is imported from the standard Lebesgue decomposition theorem.","fun_headline_variants_meta":{"raw":{"variants":["The one equation that makes Metropolis-Hastings reversible","Reversible MH: it all hinges on a single balance equation","Category theory settles the exact condition for MH reversibility","One balance equation decides MH reversibility"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000827,"raw_usage":{"total_tokens":3436,"prompt_tokens":716,"completion_tokens":2720,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":460,"completion_tokens_details":{"reasoning_tokens":2657}},"tokens_in":460,"tokens_out":2720,"duration_ms":21583,"temperature":1.0,"reasoning_tokens":2657,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T06:18:50.711392+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"In the category of s-finite kernels, take a finite measure μ and a deterministic involution φ with μ∘φ^{-1} absolutely continuous with respect to μ, choose an acceptance function α satisfying α(ξ)=α(φ(ξ))r(ξ) for μ-almost every ξ, and check detailed balance of the kernel P_MH directly on a generating algebra. If any such α fails to give a μ-reversible kernel, Theorem 4.27 is false; if a reversible kernel exists whose α violates the balancing equation on a set of positive μ×μ measure, the converse is false. A sharper structural test is to construct a finitely cancellative semiadditive CD catego","supporting_citations":[],"review_version":1}