{"id":"72469ff2-2177-4c13-ab98-7d65e6d16e52","arxiv_id":"2601.12016","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For essentially finite type algebras over a field, the ind-pro completion of a ring along a flag of prime ideals has dimension ht(p0)+ht(pn/p0)−n and is semilocal exactly when the flag is saturated.","lead":"This paper proves formulas for the size and shape of certain 'completed' rings built from a Noetherian ring and a chain of prime ideals. A generalist might read it because these rings appear in arithmetic geometry and higher local field theory.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 3.3's reduction relies on an unproved flag-adapted Noether normalization; the cited [Bo72, V, §3.1, Thm 1] does not supply it.","rationale":"The reader's weakest-assumption analysis pinpoints the exact soft spot: the coordinate-flag Noether normalization is asserted but not established. My independent reading of the proof confirms that Theorem 3.3 (and Theorem 3.7 through Remark 3.4) depends on this unproved lemma. The rest of the argument — the polynomial case, the dimension computation via the generic formal fiber, the going-up/down argument, and the semilocality criterion — appears coherent, and I do not see a more serious internal inconsistency. The missing lemma is likely true and could be supplied by a standard relative Noether normalization argument, so the appropriate verdict remains CONDITIONAL rather than REJECT or ACCEPT. I therefore see no reason to change the reader's verdict.","tokens_in":80,"tokens_out":36579,"duration_ms":699761,"concrete_test":"Look up the exact statement of [Bo72, V, §3.1, Theorem 1]. If it does not contain the coordinate-contraction condition, attempt to prove the flag-adapted Noether normalization lemma: for any finitely generated domain R over a field and any chain p0⊂...⊂pn, there exists a polynomial subring A over which R is finite and with pi∩A=(x1,...,x_{ht pi}). If a proof cannot be supplied, search for a counterexample—e.g., a chain in R=k[x,y,z] where an intermediate prime is a height-2 ideal that is not a complete intersection, such as the ideal of a monomial curve (t^4,t^5,t^6), and check whether any coordinate subring A satisfies the required contractions. A successful proof would support the reduction; an explicit counterexample would invalidate Theorem 3.3.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The load-bearing step is in the proof of Theorem 3.3. After reducing to the case where R is a domain and p0=(0), the proof invokes [Bo72, V, §3.1, Theorem 1] to assert the existence of a polynomial subring A=k[x0,...,xm] such that R is finite over A and pi∩A=(x1,...,x_{ki}) for a chain of primes. The cited Bourbaki theorem is the standard Noether normalization theorem; it guarantees finiteness over some polynomial subring but says nothing about prescribing the contractions of a finite chain of prime ideals. This stronger 'coordinate flag' lemma is not proved in the paper, and it is the sole justification for reducing the general case to the polynomial/coordinate-hyperplane case treated in Section 2. The dimension formula of Theorem 3.3 and the semilocality criterion of Theorem 3.7 (via Remark 3.4) both depend on this reduction. If the lemma is false, the central claims fail; if it is true, the proof still needs a missing argument. The lemma is plausible — for a single prime a relative Noether normalization with a principal contraction appears to hold — but the paper neither proves it nor cites a source that does. This is therefore a genuine gap, not a mere cosmetic omission.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies ind-pro completions C_ΔR of a Noetherian ring R along a flag Δ=(p_0,…,p_n) of prime ideals. Section 1 establishes general permanence properties: flatness for inclusions of flags, excellence of C_ΔR, inheritance of normality/regularity/CM/reducedness, and local equidimensionality. Section 2 analyzes in detail the case of a polynomial ring R=k[x_1,…,x_m] with the flag given by coordinate ideals; the main tools are the structural isomorphism C_ΔR ≅ C_{\\tilde Δ}(R/p_0)[[x_1,…,x_{k_0}]], a computation of the generic formal fiber dimension, and an induction giving dim C_ΔR = ht p_n − n together with a semilocality criterion. Section 3 claims to reduce the general case of an essentially finite type algebra over a field to the polynomial case by Noether normalization, obtaining the dimension formula dim C_ΔR = ht p_0 + ht(p_n/p_0) − n and the criterion that C_ΔR is semilocal iff Δ is saturated.","tokens_in":12442,"tokens_out":22699,"duration_ms":216246,"significance":"If the main theorems are correct, the paper substantially generalizes the previously known saturated-flag case of Yekutieli and the formal-fiber dimension result of Matsumura, giving a complete dimension formula and a semilocality criterion for ind-pro completions in the geometric setting. The polynomial-ring case is worked out in real detail, including explicit constructions of prime ideals of the desired height and an infinite family of maximal ideals in the non-saturated case. The general permanence results (excellence, local equidimensionality, preservation of regularity/normality) are also valuable and mostly self-contained. However, the key reduction in Section 3 rests on a flag-adapted Noether normalization that is asserted with an inadequate citation and not proved; this is a genuine gap in the derivation of the central claims.","major_comments":[{"comment":"The reduction to the polynomial case invokes [Bo72, V, §3.1, Theorem 1] for the existence of a polynomial subring A=k[x_0,…,x_m] such that R is finite over A and p_i∩A=(x_1,…,x_{k_i}) for a prescribed chain of primes p_i. The cited theorem is the standard Noether normalization theorem; it guarantees finiteness over some polynomial subring but does not by itself prescribe the contractions of a finite chain of prime ideals as coordinate ideals. This is a stronger 'flag-adapted' normalization statement, and it is load-bearing: without it the reduction to the coordinate case of Section 2 fails, and hence Theorems 3.3 and 3.7 are not proved. Please provide a complete proof of this lemma (for example, by iterating a relative Noether normalization for each quotient) or a precise reference that contains the flag-adapted version.","section":"Theorem 3.3, proof, first paragraph"},{"comment":"The equality ht p_i = ht r_i is attributed to [Ma87, Theorem 9.4]. That theorem is the going-up theorem; height preservation in a finite extension of domains does not follow from going-up alone. In the present situation A is a polynomial ring and hence normal, so going-down holds for the finite extension A⊂R, and the height equality is valid; but the justification should be stated explicitly. As written, the proof relies on a theorem that does not directly supply the asserted equality.","section":"Theorem 3.3, proof, second paragraph"}],"minor_comments":[{"comment":"The definition of ht(q/p) has the chain direction reversed: it should read p=p_0 ⊊ p_1 ⊊ … ⊊ p_n = q. The present wording q=p_0 ⊊ … ⊊ p_n=p is inconsistent with the use of ht(p_n/p_0) later.","section":"Definition 1.6"},{"comment":"Typo: 'Jackobson radical' should be 'Jacobson radical'.","section":"Introduction"},{"comment":"Typo: 'filed' should be 'field'.","section":"Notation 3.1"},{"comment":"The kernel of the third map is denoted 'rr'; this appears to be a typographical artifact. Use a consistent notation such as \\mathfrak r_r or P_r.","section":"Proposition 2.5, proof"},{"comment":"The equality Frac(A)⊗_A C_ΓA = C_ΓA is true because r_0=(0), so C_ΓA is already a localization at (0); a one-line justification would improve readability.","section":"Theorem 3.3, proof"}],"recommendation":"major_revision","confidential_remarks":"The paper is mathematically substantial and the polynomial-ring computations are convincing. The main barrier is the unproved flag-adapted Noether normalization in Section 3; this is not a matter of style but a missing argument at the core of the reduction. The author should either prove the lemma or cite a source that contains it. If the lemma is supplied, the remaining issues are local and the central claims appear sound. The self-citation [Ba25] is not a concern because the needed flatness proposition is proved in the text."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The reader is right about the substance: this is a real generalization. Ye92 handled saturated flags; this paper handles arbitrary flags for essentially finite type algebras over a field and also proves excellence for all Noetherian rings. The dimension formula dim C_∆R = ht p0 + ht(pn/p0) − n and the semilocality iff saturated criterion are the main results, and they look correct. The excellence result (Theorem 1.14) follows from known results about completions of quasi-excellent rings and is a reasonable addition. The paper is honestly written, the proofs are mostly detailed, and the self-citation to [Ba25] is not load-bearing because the flatness proposition used from it is proved here. No fitting or circularity concerns.\n\nWhere the soft spot is: the reduction in Theorem 3.3. The proof asserts, citing Bourbaki, that for a finitely generated k-algebra domain R and flag ∆, there exists a polynomial subring A = k[x0,...,xm] with R finite over A and p_i ∩ A = (x1,...,x_{k_i}) for the same flag. Standard Noether normalization only gives finiteness over some polynomial subring; it does not prescribe contractions of a chain of primes. This is a genuine gap, not a cosmetic omission, and the dimension formula and semilocality criterion both rely on the reduction. The lemma is plausible — I expect it can be proved by a relative normalization argument using that k is a field and the heights are finite — but the paper does not supply the proof or a reference that contains it. This is the difference between ACCEPT and CONDITIONAL for a referee.\n\nMinor points: Remark 3.4 and the example there are helpful but the example is a bit dense; Proposition 2.3 is the technical heart and the proof is terse in places (the height computation depends on catenarity and local equidimensionality already established, which is fine). These are minor.\n\nThe paper deserves a serious referee. The gap is fixable and the claimed results are likely true; the referee should ask for a proof or citation of the flag-adapted Noether normalization lemma. I would send this to review, and I would not cite it until the gap is closed, but I would definitely want to see the revised version.","headline":"A useful generalization of the dimension and semilocality results for ind-pro completions to arbitrary flags, with one genuine unproved normalization step in the main reduction.","tokens_in":12807,"tokens_out":1139,"would_cite":false,"duration_ms":11064,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["13A15","13B35","13C15"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper claims that ind-pro completions of essentially finite type algebras over a field have Krull dimension htp0 + ht(pn/p0) − n, and are semilocal exactly when the flag is saturated.","keywords":["ind-pro completion","flag of prime ideals","Krull dimension","semilocal ring","formal fibers","excellent ring","higher local fields","Noether normalization"],"falsifier":"Directly compute C_{(0),(x,y)} k[x,y] by Definition 1.3 (localize at (x,y), complete to k[[x,y]], then localize at (0) and complete at (0)). If the resulting ring is the fraction field k((x,y)), its Krull dimension is 0, while the claimed formula gives htp0 + ht(pn/p0) − n = 0 + 2 − 1 = 1, settling the claim either way.","tokens_in":12019,"feed_emoji":"📏","tokens_out":30090,"duration_ms":263507,"temperature":0.7,"pith_summary":"For a Noetherian ring R and a flag of prime ideals Δ=(p0,...,pn), the paper studies the ind-pro completion C_Δ R, built by alternately localizing and completing along the flag. Its main claim is that when R is essentially of finite type over a field, the Krull dimension of C_Δ R is exactly htp0 + ht(pn/p0) − n, and that C_Δ R is semilocal precisely when the flag is saturated. If correct, this gives a complete structural description of the local factors that appear in the adelic geometries of schemes, and it lets one compute the dimension of these rings directly from the height data of the flag. The proof reduces to the polynomial ring case by a coordinate normalization, and the key computation is a generic formal fiber dimension formula in the polynomial case.","feed_headline":"Dimension formula for ind-pro completions of Noetherian rings","feed_subtitle":"Finite-type algebras over a field get a height-only formula; semilocality holds exactly for saturated flags.","key_machinery":"The central object is the ind-pro completion C_Δ R = C_{p0} S^{-1}_{p0} ... C_{pn} S^{-1}_{pn} R, iterating localization and completion along the flag. The dimension proof relies on a reduction to the polynomial ring k[x_1,...,x_m] with a coordinate flag (x_1,...,x_{k_i}); the crucial step is a formula (Proposition 2.3) for the dimension of the localization at the zero ideal, which generalizes Matsumura's theorem on formal fibers. A second load-bearing device is a Noether-normalization step that puts a given flag into coordinate form, and a going-up/going-down argument that transfers dimension from the polynomial ring to the original algebra.","core_discovery":"The central discovery is a dimension and semilocality theorem for ind-pro completions. For any essentially finite type algebra R over a field, and any flag of prime ideals Δ=(p0,...,pn), the Krull dimension of C_Δ R equals htp0 + ht(pn/p0) − n, and C_Δ R is semilocal if and only if the flag is saturated. The paper also establishes that C_Δ R is always excellent, and that it inherits normality, regularity, Cohen–Macaulayness, reducedness, and local equidimensionality from the base ring under mild assumptions.","pith_inferences":["If the dimension formula extends to broader classes of excellent rings, it would tie the generic formal fiber dimension to the flag's height data; the paper's Remark 3.5 shows the formula can fail for general excellent rings, so finite-type hypotheses are essential.","The coordinate-flag normalization, if it can be proved fully, would give a transparent combinatorial model for all ind-pro completions of affine algebras.","A natural testable extension is to non-reduced or analytically ramified bases, where the dimension formula may need correction terms involving embedded primes."],"forward_implications":["For locally equidimensional R, the formula simplifies to dim C_Δ R = ht p_n − n, giving a direct height-only expression.","The semilocality criterion identifies exactly when C_Δ R is a finite product of local fields: the flag must be saturated.","Excellence of C_Δ R holds for every Noetherian R, so the construction preserves excellent behavior even for non-finite-type inputs.","The inheritance properties (regularity, normality, CM, reducedness, local equidimensionality) mean that local factors of adelic groups inherit regularity from the underlying scheme."],"fun_headline_variants":["Ind-pro completions: dimension formula and semilocality test","Semilocality iff saturated flags for ind-pro completions","Krull dimension of ind-pro completions computed","Inherited regularity and dimension for ind-pro completions","A height formula for ind-pro completions"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The reduction to the polynomial case assumes that any finitely generated k-algebra with a flag of primes admits a single polynomial subring over which the algebra is finite and every prime of the flag contracts to a coordinate ideal; this normalization is stronger than the cited theorem and is not proved.","fun_headline_variants_meta":{"raw":{"variants":["Ind-pro completions: dimension formula and semilocality test","Semilocality iff saturated flags for ind-pro completions","Krull dimension of ind-pro completions computed","Inherited regularity and dimension for ind-pro completions","A height formula for ind-pro completions"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000641,"raw_usage":{"total_tokens":2697,"prompt_tokens":566,"completion_tokens":2131,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":310,"completion_tokens_details":{"reasoning_tokens":2052}},"tokens_in":310,"tokens_out":2131,"duration_ms":16892,"temperature":1.0,"reasoning_tokens":2052,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-03T09:57:41.363416+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Directly compute C_{(0),(x,y)} k[x,y] by Definition 1.3 (localize at (x,y), complete to k[[x,y]], then localize at (0) and complete at (0)). If the resulting ring is the fraction field k((x,y)), its Krull dimension is 0, while the claimed formula gives htp0 + ht(pn/p0) − n = 0 + 2 − 1 = 1, settling the claim either way.","supporting_citations":[],"review_version":1}