{"id":"aa289908-9391-45c1-bbfa-5cd3e8e9041a","arxiv_id":"2607.15096","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Isomorphism of uniform Roe algebras over uniformly locally finite coarse spaces forces bijective coarse equivalence of the underlying spaces.","lead":"The paper proves that if the uniform Roe algebras of two uniformly locally finite coarse spaces are isomorphic as C*-algebras, then the spaces are bijectively coarsely equivalent. This settles the isomorphism rigidity problem for arbitrary uniformly locally finite coarse spaces, with no property A or metric assumptions.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified — the flagged infinite-index paving lemma (7.1) survives scrutiny and the main argument has no detectable gap.","rationale":"The reader's ACCEPT verdict is well supported. I independently examined the proof's main moving parts: the socle argument forces spatial implementation; the random diagonal-unitary lemma gives the two-sided coefficient lower bounds without property A or ghost compactness; the finite double paving lemma correctly controls cross terms; and the bijective upgrade uses a valid infinite-index paving lemma. The most plausible weak point, Lemma 7.1, is the external Kadison–Singer input, and the reader singled it out. However, on inspection the extension from finite to arbitrary index sets is sound: the compactness argument is legitimate because every paving condition depends only on finitely many coordinates, and the finite-dimensional MSS theorem provides a uniform number of classes. The subsequent common refinement for non-self-adjoint operators and the density argument for infinite-dimensional compressions are both correct. I found no internal inconsistency, no hidden use of countable generation or separability, and no circularity. The only caveat is that this is a substantial proof with no formal verification, so a full line-by-line independent audit is still desirable; but this is a confidence consideration, not a specific flaw. Therefore the verdict should remain UNCHANGED.","tokens_in":15126,"tokens_out":44673,"duration_ms":405611,"concrete_test":"As a standalone verification, instantiate Lemma 7.1 for a non-self-adjoint zero-diagonal operator T on ℓ2(I) and check the compactness transition explicitly: for any finite family of conditions (Kν,j), set K=∪Kν, apply the finite-dimensional MSS theorem to p_K T p_K with tolerance δ, and verify each block p_{Kν∩c^{-1}(j)}T p_{Kν∩c^{-1}(j)} has norm ≤δ∥T∥. If this FIP step holds, the infinite-index paving lemma is sound; it is the only non-obvious step on which the bijective upgrade depends.","verdict_should_be":"UNCHANGED","load_bearing_attack":"After a careful pass over the proof, I do not find a load-bearing concern. The theorem's validity hinges on four pillars: spatial implementation via the socle (Prop. 3.2), the random diagonal-unitary lower bound (Lemma 4.2), the finite double ℓ1-paving lemma (Lemma 5.1), and the Kadison–Singer-based upgrading (Lemma 7.1 + Prop. 7.3). I checked each. The reader's flagged assumption — extending MSS paving to arbitrary index sets — is the most external step, but the extension is correct: the compactness argument treats conditions that depend on only finitely many coloring coordinates; the FIP is supplied by the dimension-independent finite-dimensional MSS theorem on the union of the relevant finite sets; and the density of finite-support vectors then controls the infinite compression. The non-self-adjoint case via common refinement of the real and imaginary parts is valid, with at most r0(ε/2)^2 blocks. I found no hidden circularity or unstated boundedness/separability assumption. The proof legitimately relies on the external Kadison–Singer theorem, and this reliance is explicitly acknowledged. Honest non-finding.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves Theorem 1.2: for uniformly locally finite coarse spaces (X,E) and (Y,F), every C*-algebra isomorphism between the uniform Roe algebras C_u^*(X,E) and C_u^*(Y,F) forces (X,E) and (Y,F) to be bijectively coarsely equivalent. The proof has four main steps: (1) an intrinsic identification of the socle as the c0-direct sum of compact-operator algebras on coarse components, yielding spatial implementation of the isomorphism; (2) a random diagonal-unitary fourth-moment lemma giving uniform lower bounds on the rows and columns of the implementing unitary; (3) a finite double-ℓ1-paving lemma used to show that the resulting finite-to-one maps are controlled and are mutual coarse inverses; and (4) an upgrade of this coarse equivalence to a bijective one using an infinite-index Kadison–Singer/MSS paving lemma and a Hall/Cantor–Bernstein matching argument. I read the proof in detail, including the external lemmas; the argument is internally coherent and the cited theorems are applicable.","tokens_in":15391,"tokens_out":43877,"duration_ms":415302,"significance":"If correct, this completely resolves the isomorphism rigidity problem for uniform Roe algebras over arbitrary uniformly locally finite coarse spaces, removing the property A, ghost-compactness, metrizability, and countable-generation hypotheses that were present in earlier work. The method is genuinely novel: rather than localizing individual rank-one images, it randomizes the whole atomic diagonal, and the fourth-moment estimate in Lemma 4.2 is elegant and robust. The paper is careful with nonseparable index sets and uncountable coarse structures. In particular, the most delicate external step—the infinite-index extension of MSS paving in Lemma 7.1—is valid: the compactness/finite-intersection argument works because each defining condition depends on only finitely many coloring coordinates, and the finite-dimensional MSS theorem is dimension-independent; the non-self-adjoint case via common refinement also checks out. I found no circularity and no unstated boundedness or separability assumptions. The proof is a genuine advance and is written to be checkable; external tools (Kadison–Singer/MSS, Kőnig/Hall, Tonelli) are explicitly identified.","major_comments":[],"minor_comments":[{"comment":"The symbol E is used both for the coarse structure (e.g., Definition 2.1) and for expectation (e.g., Lemma 4.1 and equation (4.5)). This is a readability issue; using \\mathbb{E} for expectation would remove ambiguity.","section":"Throughout, esp. Sections 2 and 4"},{"comment":"The definition of the continuous function h with h(0)=0 and h(t)=t^{-1/2} on [α²/4,1] is terse. It would help to specify the behavior on [0,α²/4] explicitly, for example by linear interpolation, so that continuity is immediate.","section":"Proposition 7.3"},{"comment":"The argument that the closed span of the vectors η_i is all of ℓ2(σ(C)) is compressed. A one-sentence nondegeneracy argument—if ζ were orthogonal to every η_i, then Φ(k)ζ=0 for all k∈K(ℓ2(C)), contradicting essentialness—would make the spatial implementation fully transparent.","section":"Proposition 3.2"},{"comment":"The Cantor–Bernstein matching construction is standard, but the text would be easier to follow if it explicitly stated that I∞ is the union of the I_n and that σ is well defined because I∖I∞⊆ψ(J).","section":"Lemma 7.2"}],"recommendation":"minor_revision","confidential_remarks":"The paper is within the journal's scope and the result is significant. I have no concerns about circularity or novelty; the reliance on Kadison–Singer/MSS is explicit and legitimate. The requested changes are cosmetic, and once the minor notation and exposition points are addressed, the paper should be accepted."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Bottom line: this paper resolves Problem 1.1, and the proof looks right. It removes property A, ghost compactness, and countability assumptions from the metric-space rigidity results and handles arbitrary coarse structures, including nonseparable ℓ2(X). That is a real advance, not an incremental tweak.\n\nWhat's new: the two techniques — the Steinhaus-randomized diagonal unitary fourth-moment estimate (Lemma 4.2) and the finite double ℓ1 paving lemma (Lemma 5.1) — are genuinely original. The fourth-moment argument avoids needing to localize individual ghost projections, and the paving lemma controls cross terms without a countable exhaustion. The upgrade from coarse equivalence to bijective coarse equivalence via Hall's theorem and the Kadison–Singer-based paving is also clean.\n\nI checked the soft spots the reader flagged. The infinite-index extension of the MSS paving theorem in Lemma 7.1 is the most external step, but the compactness argument is sound: the conditions depend on finitely many coloring coordinates, and the FIP comes from the finite-dimensional theorem on finite unions. The non-self-adjoint case via common refinement of real and imaginary parts is fine. The spatial implementation in Prop 3.2 is standard and correct. I did not find circular reasoning; the coefficient bounds are derived, not fitted.\n\nSoft spots: The paper is long and dense; a referee will need real time. Some steps are compressed, e.g. the polar part construction in Prop 7.3 and the measure-theoretic bookkeeping in Lemma 4.2, but they are supplied. The proof leans on Kadison–Singer, which is a heavy hammer, but that is acknowledged and legitimate. No formal verification artifacts, but the argument is structured so the key estimates can be checked independently. The citation pattern looked appropriate to me; prior work is clearly distinguished from the new result.\n\nWho this is for: anyone working on coarse geometry and operator algebras, especially rigidity of Roe algebras. This is a serious paper and deserves a full referee. I would send it to peer review without hesitation and would bring it to a reading group if someone wants a deep dive.","headline":"Settles the isomorphism rigidity problem for uniform Roe algebras over all uniformly locally finite coarse spaces, with a proof that holds up under scrutiny.","tokens_in":15843,"tokens_out":1830,"would_cite":true,"duration_ms":18749,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["46L05","46L85","51F30"],"pacs":[],"model":"deepseek-v4-flash","headline":"Every C*-algebra isomorphism between uniform Roe algebras forces the underlying coarse spaces to be bijectively coarsely equivalent, with no extra geometric assumptions.","keywords":["uniform Roe algebra","coarse space","bijective coarse equivalence","C*-algebra rigidity","random diagonal unitary","finite paving","coarse components","spatial implementation"],"falsifier":"Construct a unitary U:ℓ2(X)→ℓ2(Y) such that U ℓ∞(X) U* is contained in a uniform Roe algebra C_u^*(Y,F) but inf_{y∈Y} sup_{x∈X} |⟨U δ_x, δ_y⟩| = 0; this would contradict Lemma 4.2 and collapse the coefficient estimates. Alternatively, exhibit any pair of uniformly locally finite coarse spaces whose uniform Roe algebras are isomorphic but which are not bijectively coarsely equivalent.","tokens_in":15031,"feed_emoji":"🔗","tokens_out":3783,"duration_ms":41466,"temperature":0.7,"pith_summary":"Uniform Roe algebras are C*-algebras built from the large-scale geometry of a space. This paper proves the rigidity direction: if two such algebras are isomorphic as C*-algebras, the underlying coarse spaces must be bijectively coarsely equivalent. Earlier results needed extra hypotheses such as property A, metrizability, or countably generated coarse structures; this argument removes all of them. The proof does not rely on ghost-operator compactness or property A. Instead, it randomizes diagonal unitaries to force uniform lower bounds on the matrix coefficients of the implementing unitary, then uses finite paving to extract controlled maps and ultimately a bijection.","feed_headline":"Uniform Roe algebras remember all of coarse geometry","feed_subtitle":"Rigidity holds for arbitrary uniformly locally finite coarse spaces, with no property A or metrizability assumptions.","key_machinery":"The two load-bearing tools are (1) a random diagonal-unitary lemma, which states that if a unitary conjugates the atomic diagonal algebra into a uniform Roe algebra, then the maximum coefficient in each row and column is bounded below by a positive constant—proved by fourth-moment estimates and a norm-approximation contradiction; and (2) a finite double l1-paving lemma, which partitions any relation with bounded row and column weights into finitely many classes in which cross terms are uniformly small. This paving step detects entourages from matrix coefficients above a fixed threshold. The final bijection step invokes an infinite-index paving theorem for zero-diagonal operators to make cert","core_discovery":"The paper establishes Theorem 1.2: for uniformly locally finite coarse spaces (X,E) and (Y,F), any C*-algebra isomorphism C_u^*(X,E) ≅ C_u^*(Y,F) implies that (X,E) and (Y,F) are bijectively coarsely equivalent. The proof first identifies the closed socle of a uniform Roe algebra as the c0-direct sum of compact-operator algebras over coarse components, which gives spatial implementation by a unitary that permutes component summands. A random diagonal-unitary argument then produces two-sided uniform lower bounds on the largest matrix entries of that unitary. A finite double l1-paving lemma converts these bounds into controlled maps, and a final paving theorem for zero-diagonal operators upgra","pith_inferences":["The same random diagonal-unitary mechanism may extend to other operator algebras that contain the full atomic diagonal and whose norm closure controls matrix supports, suggesting rigidity could hold for broader classes of Roe-like algebras.","A testable corollary is that every automorphism of a uniform Roe algebra is spatially implemented and determines a bijective coarse self-equivalence of the underlying space; this could be checked by examining whether the resulting bijection can be chosen to be the implementing unitary's canonical permutation.","The quantitative constants in the coefficient bounds could likely be made effective, yielding explicit paving constants in terms of the uniform local finiteness degree and the norm of the isomorphism.","A natural next probe is stable isomorphism or Morita equivalence rigidity in the same full generality, since the present method targets isomorphisms and may require additional coarse-cardinality hypotheses for stable equivalence."],"forward_implications":["Isomorphism rigidity holds unconditionally for uniformly locally finite coarse spaces: no property A, ghost-compactness, metrizability, or countable-generation hypothesis is needed.","The abstract uniform Roe algebra, without its distinguished canonical diagonal, determines the underlying coarse space up to bijective coarse equivalence.","Every C*-algebra isomorphism between such algebras is spatially implemented by a unitary that permutes the coarse-component summands.","The proof yields a concrete coarse equivalence consisting of controlled maps f and g with both composites close to the identity, and then a bijection close to f.","If the two uniform Roe algebras are isomorphic, then the spaces have the same coarse components in a bijective correspondence."],"fun_headline_variants":["Roe algebras pin down coarse geometry exactly","Coarse equivalence forced by Roe algebra isomorphism","Uniform Roe rigidity without extra assumptions","Roe algebras remember every coarse detail","Socle and paving prove full Roe rigidity"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The final upgrade from a coarse equivalence to a bijective coarse equivalence depends on extending the finite-dimensional matrix paving theorem to zero-diagonal operators on arbitrary, possibly uncountable index sets, for both self-adjoint and non-self-adjoint operators; if that infinite-index paving theorem failed, the argument would only produce a coarse equivalence, not a bijective one.","fun_headline_variants_meta":{"raw":{"variants":["Roe algebras pin down coarse geometry exactly","Coarse equivalence forced by Roe algebra isomorphism","Uniform Roe rigidity without extra assumptions","Roe algebras remember every coarse detail","Socle and paving prove full Roe rigidity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000703,"raw_usage":{"total_tokens":2938,"prompt_tokens":604,"completion_tokens":2334,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":348,"completion_tokens_details":{"reasoning_tokens":2280}},"tokens_in":348,"tokens_out":2334,"duration_ms":20187,"temperature":1.0,"reasoning_tokens":2280,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T00:12:11.317182+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Construct a unitary U:ℓ2(X)→ℓ2(Y) such that U ℓ∞(X) U* is contained in a uniform Roe algebra C_u^*(Y,F) but inf_{y∈Y} sup_{x∈X} |⟨U δ_x, δ_y⟩| = 0; this would contradict Lemma 4.2 and collapse the coefficient estimates. Alternatively, exhibit any pair of uniformly locally finite coarse spaces whose uniform Roe algebras are isomorphic but which are not bijectively coarsely equivalent.","supporting_citations":[],"review_version":1}