{"id":"08055400-13a1-43ec-975f-085d408f1521","arxiv_id":"2510.04958","paper_version":3,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":1,"one_line_summary":"This paper defines the ring stacks sRn and sR⊕n from the ring space of sheared Witt vectors and builds several explicit quasi-isomorphic models for them.","lead":"Using a variant of Witt vectors called sheared Witt vectors, this paper defines new objects called ring stacks and constructs several explicit models for them. These stacks are conjectured to give a uniform description of truncated p-divisible groups and related Shimurian moduli spaces, but that connection is not proved here.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Central claim depends on an unpublished identification in Prop. 2.3.4; if it fails, Cor. 2.3.6 and all economic models collapse.","rationale":"The reader's CONDITIONAL verdict is exactly right and already identifies the same load-bearing assumption. The paper's central claim is not the motivating BT-stack conjecture (that is explicitly conjectural) but the well-definedness of sRn/sR⊕n and the validity of the displayed models. Every path to those models passes through the sheaf Q = W/Ŵ and the unpublished [BMVZ]/[BMVZ] results, especially Proposition 2.3.4 and Corollary 2.3.6. I found no internal contradiction in the paper; the arguments after §3 are mostly careful reductions to those facts. Proposition 4.1.4 is stated without proof, but it is a concrete algebraic equivalence and less load-bearing than the sheaf-theoretic input. The appropriate response is to keep the verdict CONDITIONAL: accept the definitions and models only conditionally on an independent verification of the sheared Witt vector foundations. No verdict adjustment is needed.","tokens_in":31838,"tokens_out":41452,"duration_ms":337718,"concrete_test":"Settle this by an independent, self-contained proof of Proposition 2.3.4: from the definitions of W, Ŵ, Q and ˜V, show that for every p-nilpotent ring R, (Q/˜V(Q))(R) is the colimit perfection of R_red, and deduce Corollary 2.3.6 without citing [BMVZ] or [M1]. Pay particular attention to p=2, where ar u ≠ 1 and (2.5)–(2.6) are the only replacements for the p>2 case, and to rings R for which R_red is not perfect. If this derivation cannot be completed, the economic models (1.4)–(1.6) are not established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The weakest link is the imported theory of Q := W/Ŵ. In the proof of Proposition 2.3.4 the paper asserts without derivation that Q/˜V(Q) = W/(V(W)+Ŵ) is the fpqc-sheafification of the presheaf R ↦ R_red. That identification is used to prove Corollary 2.3.6, namely that ˜V is an isomorphism on T_F(Q). Corollary 2.3.6 is then used verbatim in Proposition 5.3.4 to kill the kernel (5.9) and obtain the first economic model; the later models in §7.1 and §8.3.6 are reductions of the same chain. The paper cites [BMVZ]/[M1] for this fact, both unpublished. If the identification is wrong—e.g. because F: Ŵ→Ŵ fails to be fppf-surjective in some case, or because the p=2 correction ar u ≠ 1 changes the argument—then the exact sequences (3.5)–(3.6), (3.13)–(3.14), Lemma 3.5.1, and the quasi-isomorphisms identifying sRn with the economic models are unsupported. This is a verification gap rather than a demonstrated internal contradiction, but it is the single most load-bearing point in the paper.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper defines, for each n ∈ N, ring stacks sRn := Cone(sW --p^n--> sW) and graded analogs sR⊕n := Cone(sW⊕ --p^n--> sW⊕), using the ring space sW of sheared Witt vectors, following constructions from the unpublished works [BMVZ] and [BKMVZ]. It then proposes several 'models' for these ring stacks, i.e. realizations as cones of quasi-ideals, including a mixed-characteristic economic model Cone(Ŵ(F^n) → W_n) for p > 2 and more refined models for general p. The motivating conjecture, not proved here, is that sRn and sR⊕n describe the stacks BT^{G,μ}_n from [GM]. The paper also develops auxiliary structures: the ring space Q = W/Ŵ, the operator V-tilde, the Lau equivalence for Z-graded rings, derived p-completeness of sW, and a self-dual model A-tilde_n. Several proofs are referenced to unpublished works, and one key Proposition 4.1.4 is left to the reader.","tokens_in":32163,"tokens_out":2930,"duration_ms":26447,"significance":"If the constructions and models are correct, the paper provides concrete, computable algebraic models for objects conjecturally related to moduli of Barsotti–Tate groups and their Shimurian analogs. The definitions are explicit, and a substantial number of quasi-isomorphisms between the proposed models are proved within the text (or justifiably delegated to [BKMVZ, BMVZ] modulo the caveats below). The paper is honest about its conjectural status and about which statements are imported from unpublished sources. The Lau equivalence, though not proved here, is a natural algebraic tool that may have independent value. The main significance is conditional: the central claims hinge on a small number of externally supplied facts, especially the analysis of Q/˜V(Q) in §2.3.4, whose proof is not actually supplied in the manuscript.","major_comments":[{"comment":"The proof of Proposition 2.3.4 contains a load-bearing unproved assertion: “Therefore Q/˜V(Q) is the fpqc-sheafification of the presheaf R ↦ R_red.” No derivation is given. This statement is used to prove Corollary 2.3.6, which is then used verbatim in Lemma 3.2.5 and in Proposition 5.3.4, and the later economic models of §8.3.6 all depend on that chain. Since the cited sources [BMVZ]/[M1] are unpublished, the reader cannot verify the key identification. Please provide a complete proof, or a precise theorem statement with a proof sketch sufficient to check the fpqc-sheafification claim and the behavior of ˜V.","section":"§2.3.4, proof of Proposition 2.3.4"},{"comment":"The Lau equivalence L: C_ec → C is stated as a proposition but its proof is “left to the reader.” This is not a cosmetic omission: the definition of sW⊕ in §4.2.1 and hence of sR⊕n in (5.2) depends on the equivalence being an equivalence of categories. The remarks in §4.1.5 illustrate the formula but do not establish the inverse functor or the isomorphisms of graded rings. Please provide a complete proof or a precise reference to a published version of [L21] that contains this statement.","section":"§4.1.4, Proposition 4.1.4"},{"comment":"Derived p-completeness of sW is proved by saying “We follow [BMVZ].” This property is used in Proposition 5.2.3 to identify lim_n sRn with sW, and in Remark 3.5.2 to reduce to pointwise derived p-completeness. The proof given is only a sketch: it reduces to derived p-completeness of W(R) and (T_F(Q))(R), but the step “p^n Q(F^n) = 0” uses (2.7) and the assertion that Q(F^n) is killed by p^n. If that step is correct, it should be expanded; if it relies on a property only available in [BMVZ], the dependence should be made explicit and the missing argument supplied.","section":"§3.5.3, proof of Lemma 3.5.1"}],"minor_comments":[{"comment":"Typo: “regared” should be “regarded.”","section":"§1.3.1"},{"comment":"Grammar: “The ring spaces Wis called” should be “The ring space sW is called.”","section":"§3.1.1"},{"comment":"Typo in heading: “descirbed” should be “described.”","section":"§4"},{"comment":"The formula “u := V^{-1}(p-[p])” uses an inverse of V that is not literally defined on W(Z_p) as a two-sided inverse; the notation is informal. A parenthetical clarification would help.","section":"§3.2.1"},{"comment":"The proof of Proposition 5.2.2(i) is deferred to [BMVZ] with a note that §8 will provide a direct description. Since §8 itself relies on §§2–3, the statement of Proposition 5.2.2 is not independently justified in this paper; a cross-reference to the exact place in [BMVZ] would be useful.","section":"§5.2.2(i)"},{"comment":"The references [BMVZ], [BKMVZ], [M1], [M2], [Vo], and [L25] are listed as unpublished or lecture recordings; the manuscript would be easier to evaluate if the author indicated which of these are expected to appear in published form and to what extent the present paper depends on them.","section":"References"}],"recommendation":"major_revision","confidential_remarks":"The manuscript is an exposition of constructions largely developed in unpublished joint works, and several of the most consequential statements are cited to those works rather than proved or even fully sketched here. The paper is self-contained enough that a motivated reader can see the main architecture, but the external dependence is unusually heavy for a standalone paper. If the companion works are expected to be posted soon, a revised version should either include the missing proofs or state precise theorem numbers from the companions. The author's integrity in flagging dependencies is appreciated; the main risk is the unproved identification in §2.3.4, which is the true bottleneck."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here is where I land after reading it. The ring stacks sRn and sR⊕n are genuinely new objects, and the paper is honest about what it takes from [BKMVZ] and [BMVZ]. The definitions are explicit, most of the quasi-isomorphisms between models are actually proved, and the Lau equivalence in §4 is a useful self-contained retelling. The self-dual model in §6 is a real addition, not just a repackaging. The author’s own §1.7 is unusually candid: most of the content is contained in or a straightforward modification of unpublished work. That candor is accurate, and it sets the right frame for the reader.\n\nThe soft spots are not manufactured. The largest is Proposition 2.3.4: the paper asserts, with a proof outline citing [M1] and [BMVZ], that Q/˜V(Q) is the fpqc sheafification of R ↦ R_red. That identification is load-bearing. Corollary 2.3.6 follows from it, and Corollary 2.3.6 is used verbatim in Proposition 5.3.4 to get the first economic model; the later models in §7 and §8 are reductions of the same chain. If that identification is wrong, the exact sequences (3.5)–(3.6) and (3.13)–(3.14), Lemma 3.5.1, and the economic models all lose support. I see no sign that it is wrong, but I also cannot verify it from this paper, and neither can a referee without access to the companion works. That is a verification gap, not a demonstrated error. The stress-test note is right to single it out.\n\nSmaller soft spots: Proposition 4.1.4 (the Lau equivalence) is stated without proof, and Conjecture 3.7.3 is open. Both are flagged by the author. Neither undermines the paper’s main purpose, but they do reinforce that this is not a self-contained foundation—it is a precise scaffold that will become fully load-bearing when the unpublished companions appear.\n\nWho is it for? People working on sheared prismatization, displays, or the Drinfeld–Gardner–Madapusi conjectures. They will want these definitions and models on record. It deserves a serious referee, with the condition that the referee either has access to [BMVZ] or is asked to verify Proposition 2.3.4 independently. I would send it to peer review and I would bring it to a reading group if anyone in the group works on BT stacks or Witt vector technology.","headline":"A precise, clearly written construction paper whose central definitions and models are new, but whose foundations rest on unpublished work and one unproved identification a referee must chase down.","tokens_in":32712,"tokens_out":2094,"would_cite":true,"duration_ms":18471,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14L05","14F30"],"pacs":[],"model":"deepseek-v4-flash","headline":"A new family of ring stacks built from sheared Witt vectors is conjecturally the natural coefficient object for truncated p-divisible groups and their related generalizations.","keywords":["ring stacks","sheared Witt vectors","Barsotti-Tate groups","p-divisible groups","truncated displays","Witt frame","prismatization","Cartier duality"],"falsifier":"Compute $sR_n(R)$ for a concrete $p$-nilpotent ring such as $R=F_p[\\epsilon]/(\\epsilon^2)$ using the definition $sR_n=\\mathrm{Cone}(sW \\xrightarrow{p^n} sW)(R)$ and using the claimed model $\\mathrm{Cone}(\\hat W(F^n) \\rightarrow W_n)(R)$; any disagreement disproves the model for $p>2$. Equally, test (3.16)-(3.17) by searching for an element of $sW(R)$ not in the image of $1-\\tilde V$, or by checking that the kernel is not exactly $Z_p(1)$.","tokens_in":31623,"feed_emoji":"🧮","tokens_out":5915,"duration_ms":47353,"temperature":0.7,"texified_at":"2026-08-05T20:33:14.258376+00:00","pith_summary":"The paper defines two ring stacks, $sR_n$ and its graded companion $sR^\\oplus_n$, using the sheared Witt vector ring space $sW$, and proves they are well-defined $Z/p^nZ$-algebra stacks with several equivalent models. The motivating claim is that these stacks describe the stacks $BT^{G,\\mu}_n$ of n-truncated Barsotti-Tate groups, in the sense that the relevant bundles are encoded by maps into or out of $sR_n$. If the conjecture holds, the complicated moduli of p-divisible groups would be governed by explicit, Witt-vector-built ring objects whose structural maps $F$ and $\\tilde V$ carry all the group law data. The paper's own contribution is the definitions and the proof that the various models are quasi-isomorphic, not the conjecture itself.","texify_model":"deepseek-v4-flash","texify_usage":{"total_tokens":8170,"prompt_tokens":884,"completion_tokens":7286,"prompt_tokens_details":{"cached_tokens":0},"prompt_cache_hit_tokens":0,"prompt_cache_miss_tokens":884,"completion_tokens_details":{"reasoning_tokens":6381}},"feed_headline":"Sheared Witt vectors build rings that may encode p-divisible groups","feed_subtitle":"New ring stacks come with several equivalent models, opening a conjectural dictionary to truncated Barsotti-Tate groups.","key_machinery":"The sheared Witt vector ring space $sW = W \\times_Q Q^{perf}$, equipped with Frobenius $F$ and a twisted Verschiebung $\\tilde V$ (which equals $V$ in characteristic $p>2$ but must be corrected by a unit $u$ when $p=2$), together with the ideal $\\hat W$ of nilpotent, finite-support Witt vectors. The cone construction for ring groupoids turns a quasi-ideal $d: I \\to A$ into a ring stack; iterating it on $sW$ (and on the $Z$-graded $sW^\\oplus$ obtained via the graded-ring equivalence) produces $sR_n$ and $sR^\\oplus_n$. Exact sequences (3.5)-(3.6) and (3.13)-(3.14), plus derived $p$-completeness, are what make the comparisons between models valid.","core_discovery":"Taking the ring space $sW := W \\times_Q Q^{perf}$ of sheared Witt vectors, the author sets $sR_n := \\mathrm{Cone}(sW \\xrightarrow{p^n} sW)$ and, after applying a graded-ring reconstruction equivalence to the pair $(sW, F, \\tilde V)$, $sR^\\oplus_n := \\mathrm{Cone}(sW^\\oplus \\xrightarrow{p^n} sW^\\oplus)$, obtaining stacks of $Z/p^nZ$-algebras and $Z$-graded $Z/p^nZ$-algebras, respectively. The paper proves several models for $sR_n$ are canonically quasi-isomorphic to these definitions: for $p>2$, $sR_n \\cong \\mathrm{Cone}(\\hat W(F^n) \\rightarrow W_n)$, and over $F_p$ a similar cone model holds for every $p$, while an additional economic model exists for all $p$ with a small modification when $p=2$. It also proves the projective limit of the $sR_n$ is $sW$, so the family is essentially the decompletion of the u","pith_inferences":["One can read the construction as a proposal that truncated displays can be encoded by ring homomorphisms out of sR_n; if so, deformation theory of p-divisible groups would reduce to deformation theory of ring homomorphisms.","The economic model over F_p for every p suggests an F_p-flavored version of the dictionary that might be testable via explicit computations with perfect and semiperfect rings.","The autoduality conjecture of §3.7, if proved, would give a clean Ext-formulation of Cartier duality on sW and could serve as the algebraic engine behind the conjectural description of BT stacks.","Because the paper stops short of proving the motivating conjecture, the models here are the machinery one would use to attempt a proof; the natural next step is to construct a comparison map from BT^{G,μ}_n to the corresponding bundles on sR_n."],"forward_implications":["If the motivating conjecture holds, the stack of n-truncated Barsotti-Tate groups of height d and dimension d' is isomorphic to the relevant G-bundles over the ring stack sR_n for G=GL(d), so the moduli problem is governed by a single ring object.","The economic model for p>2, sR_n ≅ Cone(\\hat W(F^n)→W_n), gives an explicit, ind-finite description that makes commutation with filtered colimits immediate and may make deformation-theoretic computations tractable.","The equality lim_{←n} sR_n = sW gives a universal sheared-Witt coefficient ring from which all truncated versions are recovered, placing sW on the same footing as the ordinary Witt scheme.","The self-dual model of §6 and the Cartier duality between \\hat W(F^n) and W_n suggest that duality of truncated p-divisible groups is visible at the level of the ring stacks themselves.","The p=2 case requires a modified model, so the conjecture's behavior at the prime 2 is marked by genuinely different algebra."],"fun_headline_variants":["Sheared Witt vectors build ring stacks for p-divisible groups","New ring stacks from sheared Witt vectors","Conjectural ring stacks to truncated Barsotti-Tate groups","Sheared Witt vectors link to truncated p-divisible groups","Ring stacks: a conjectural dictionary to BT groups"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The entire edifice rests on unpublished results on sheared Witt vectors — surjectivity of $F$ on $\\hat W$, fppf/fpqc triviality of $\\hat W$-torsors, Proposition 2.3.4 about $Q/\\tilde V(Q)$, and derived $p$-completeness of $sW$ — so if any of these fails, the exact sequences and hence all models for $sR_n$ collapse.","fun_headline_variants_meta":{"raw":{"variants":["Sheared Witt vectors build ring stacks for p-divisible groups","New ring stacks from sheared Witt vectors","Conjectural ring stacks to truncated Barsotti-Tate groups","Sheared Witt vectors link to truncated p-divisible groups","Ring stacks: a conjectural dictionary to BT groups"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000469,"raw_usage":{"total_tokens":2111,"prompt_tokens":621,"completion_tokens":1490,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":365,"completion_tokens_details":{"reasoning_tokens":1411}},"tokens_in":365,"tokens_out":1490,"duration_ms":9312,"temperature":1.0,"reasoning_tokens":1411,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T11:21:51.101629+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute $sR_n(R)$ for a concrete $p$-nilpotent ring such as $R=F_p[\\epsilon]/(\\epsilon^2)$ using the definition $sR_n=\\mathrm{Cone}(sW \\xrightarrow{p^n} sW)(R)$ and using the claimed model $\\mathrm{Cone}(\\hat W(F^n) \\rightarrow W_n)(R)$; any disagreement disproves the model for $p>2$. Equally, test (3.16)-(3.17) by searching for an element of $sW(R)$ not in the image of $1-\\tilde V$, or by checking that the kernel is not exactly $Z_p(1)$.","supporting_citations":[],"review_version":1}