{"id":"f78bf8cf-aac9-4ae5-9397-bef293718849","arxiv_id":"2411.15569","paper_version":1,"verdict":"REJECT","confidence":"HIGH","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"The paper derives a complete G-algebra description of the Hochschild cohomology of the first Frobenius kernel of SL2, but a key multiplicity statement contradicts the paper's own appendix.","lead":"The paper computes the full structure of a cohomology ring attached to the first Frobenius kernel of SL2, a basic object in modular representation theory. The authors present a spectral sequence method and give an explicit description for SL2 over all primes.","discovery_kind":"new_method","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 4.3.2(a) undercounts trivial summands in HH^0(G1)(−1): Prop 4.3.1 plus (4.3.2) give (3p−1)/2, not (p−1)/2; e.g. p=5 gives 7 vs 2. The central G-module description is false as stated.","rationale":"The paper's central claim is a complete G-algebra structure for HH^*(G1). The module-structure theorem (Theorem 4.3.2) is the foundation; the ring structure (Theorem 4.4.2) builds on it. The most load-bearing defect is internal: Theorem 4.3.2(a) undercounts the trivial summands in HH^0(G1)(−1). Combining Proposition 4.3.1 with (4.3.2) gives (3p−1)/2 trivial summands, not (p−1)/2 (e.g., p=5: 7 vs 2; p=7: 10 vs 3). The appendix tables for p=5 and p=7 confirm these counts, so this is not an external convention issue. Because Theorem 4.3.2 is false as stated, the advertised complete description fails. The reader's weakest_assumption targeted the collapse and ungrading of the spectral sequence in Theorem 4.4.2; that is also under-justified ('by considering the differentials' without explicit computation), but the multiplicity error is more decisive because it can be settled by direct summation and contradicts the paper's own tables. No machine-checked proof or independent implementation is provided. In good faith, the approach may be salvageable by correcting Theorem 4.3.2(a) and revisiting the dependent statements, but the current manuscript's central claim is not correct. The appropriate verdict remains REJECT.","tokens_in":13526,"tokens_out":13117,"duration_ms":102608,"concrete_test":"Analytical test: For p=5, compute dim HH^0(G1)(−1) by summing the H^0 contributions from Proposition 4.3.1. The contributing even n are 0, 2, 10, 12 (S^n_0 ≅ k) and 4, 6, 8 (S^n_0 ≅ T(8)); each contributes one k by (4.3.2). Total = 7. Theorem 4.3.2(a) says 2. Repeat for p=7: the contributing even n are 0,2,4,14,16,18 (k summands) and 6,8,10,12 (T(12) summands), totaling 10, versus the claimed 3. If these counts are confirmed, Theorem 4.3.2(a) must be corrected and the dependent ring-structure results re-examined.","verdict_should_be":"REJECT","load_bearing_attack":"The central claim requires the multiplicities from Proposition 4.3.1 to be transferred correctly to H^0(G1, −). Using (4.3.5), HH^0(G1)(−1) = H^0(G1, S(g)_0). For p ≥ 3, Proposition 4.3.1 gives: even n in [0, p−3] have S^n_0 ≅ k (count (p−1)/2); even n in [p−1, 2p−2] have S^n_0 ≅ T(2p−2) (count (p+1)/2); even n in [2p, 3p−3] have S^n_0 ≅ k (count (p−1)/2). By (4.3.2), each T(2p−2) summand contributes one copy of k to H^0, and each k summand contributes one copy. Odd n contribute 0. Therefore dim HH^0(G1)(−1) = (p−1)/2 + (p+1)/2 + (p−1)/2 = (3p−1)/2, not (p−1)/2. For p=5 this is 7 trivial summands (from n = 0, 2, 4, 6, 8, 10, 12), whereas Theorem 4.3.2(a) predicts 2; the appendix table for p=5 confirms exactly these seven n have H^0 = k. For p=7 the same count gives 10 vs 3, also matching the appendix. This is an internal inconsistency, not a convention difference, so the theorem as stated is false and the advertised complete G-module description of HH^*(G1) fails.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper develops spectral sequence methods for computing the Hochschild cohomology of Frobenius kernels and applies them to G = SL2. The main advertised result is a complete description of the G-algebra structure of HH^*(G1) for all primes p, obtained by analyzing the adjoint action on Dist(G1). The paper also contains computations for the Borel subgroup B1 and the unipotent subgroup U1, together with appendix tables for p = 2, 3, 5, 7.","tokens_in":13880,"tokens_out":10346,"duration_ms":71758,"significance":"If correct, the paper would provide the first complete computation of the G-algebra structure of HH^*(G1) for SL2, a substantial result in the representation theory of infinitesimal group schemes. The spectral sequence setup (Theorems 2.2.1–2.2.3) and the use of Kostant's theorem, tilting modules, and good filtrations are appropriate, and the appendix tables are useful data. However, the main theorem contains a clear internal inconsistency: Theorem 4.3.2(a) gives a multiplicity for HH^0(G1)(-1) that contradicts Proposition 4.3.1, and the proof of the ring structure in Theorem 4.4.2 relies on an unproved spectral sequence collapse. The advertised 'complete description' is therefore not established as stated.","major_comments":[{"comment":"Theorem 4.3.2(a) states HH^0(G1)(-1) is isomorphic to k^{⊕(p-1)/2}. However, combining Proposition 4.3.1 with (4.3.2) gives multiplicity (3p-1)/2. Indeed, even n in the first range 0 ≤ n < p-1 contribute (p-1)/2 copies of k; even n in the range p-1 ≤ n ≤ 2(p-1) contribute (p+1)/2 copies of T(2p-2), each contributing one copy of k by (4.3.2); and even n in 2(p-1) < n ≤ 3(p-1) contribute another (p-1)/2 copies of k. For p=5 this gives 7, not 2, and the appendix table for p=5 (rows n = 0, 2, 4, 6, 8, 10, 12) confirms exactly those seven trivial summands in H^0(G1, S^n(g))(-1). The theorem as stated is false.","section":"Theorem 4.3.2(a), Proposition 4.3.1, (4.3.2)"},{"comment":"The proof of Theorem 4.4.2 asserts that the spectral sequence of (4.4.1) 'must stop at E3' for the summand N = T(2p-2), but no differential computation is shown. Since the definition of I_p depends on exactly which differentials kill which terms, the reader cannot verify the claimed B-algebra isomorphism. The later 'ungrading' step, from the associated graded ring to an isomorphism of rings, is justified only by reference to (4.4.4) and two sentences on weight considerations. This is load-bearing because Theorem 4.4.2 is used in Theorem 4.4.3 to obtain the ring structure of HH^*(G1).","section":"Theorem 4.4.2, proof"},{"comment":"The proof states: 'From Theorem 4.3.3, it follows that as a B-module, H^•(B1, \\bar S(g)_0)(-1) is a direct sum of B-module factors isomorphic to k, and S^•(u*), ω, and -ω (when p=2).' For p ≥ 3 the relevant statement is Theorem 4.3.2, not 4.3.3, and Theorem 4.4.2(a) describes H^•(B1, \\bar S(g)_0) as a quotient of S^•(u*)(1) ⊗ H^•(u, \\bar S(g)_0)^{T1} by I_p, a structure that includes summands such as ind_G^B[S^•(u*)⊗ω]⊗L(1). Those are not of the simple listed form. Therefore the collapse of the spectral sequence in Theorem 2.2.1 is not established for p ≥ 3.","section":"Theorem 4.4.3, proof"},{"comment":"The definition of I_p is not mathematically well-formed: the notation I_p = ⟨ \\bigoplus_{i=0}^{(p-1)/2} S^•(u*)(1) ⊗ f^{p-1}x^i, \\bigoplus_{j=(p-1)/2}^{p-1} S^•(u*)(1)_+ ⊗ x^j ⟩ mixes direct sums with products inside an ideal, and the ranges and tensor structures of the generators are ambiguous. This obscures the claimed isomorphism and prevents the reader from verifying the ring structure.","section":"Theorem 4.4.2(a), definition of I_p"}],"minor_comments":[{"comment":"A literal '[cite]' placeholder appears in the text: 'prior calculations of this type for the Taft algebra used techniques involving resolutions. [cite]' This must be replaced with a proper reference.","section":"Section 3"},{"comment":"The phrase 'still an open for problem' should be corrected to 'still an open problem'.","section":"Section 1.2"},{"comment":"Part (b) says x is represented by 1 ⊗ [ef − h], while the proof says x is represented by 1 ⊗ [ef − h2] and later 1⊗[ef − h^2]. The exponent on h should be 2 consistently.","section":"Proposition 4.4.1(b) and its proof"},{"comment":"The reference [BNP04] is listed in the bibliography but does not appear to be cited in the text.","section":"References"},{"comment":"The paper switches between S(g) and \\bar S(g) for the truncated symmetric algebra without a clear definition of \\bar S(g); the reader must infer from context that both denote S^•(g) modulo p-th powers.","section":"Notation"}],"recommendation":"reject","confidential_remarks":"The internal inconsistency in Theorem 4.3.2(a) is decisive: the claimed complete G-module description is false in degree zero. Even if that multiplicity were corrected, the proof of Theorem 4.4.2 lacks the differential computation needed to justify the collapse at E3, and the definition of I_p is too unclear to verify. The presence of a literal '[cite]' placeholder and several typos suggests the manuscript was not fully proofread. The spectral sequence machinery and the appendix tables are potentially valuable, but the paper is not ready for publication in its current form."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Quick take: the main theorem on the G-module structure of HH^*(G1) is wrong as stated. The multiplicity of the trivial module in HH^0(G1)(-1) is (3p-1)/2, not (p-1)/2, by the paper's own Proposition 4.3.1 and (4.3.2). The appendix tables for p=5 and p=7 confirm the larger number. This is an internal inconsistency, not a convention difference.\n\nThat said, the paper is not without merit. The spectral sequence framework for Hochschild cohomology of Frobenius kernels is a genuinely useful new angle, and the Taft algebra computation for B1 in Section 3 is clean and convincing. The tables in the appendix, and the systematic use of tilting modules and Kostant's theorem, are helpful for anyone who wants to compute more examples.\n\nThe soft spots are concentrated in Section 4. The multiplicity error in Theorem 4.3.2(a) is load-bearing: it invalidates the advertised complete G-module description. The proof of Theorem 4.4.2 (ring structure) is also underdeveloped. The collapse at E3 is asserted with 'by considering the differentials' and no differential computation is shown, and the ungrading from the spectral sequence to a ring isomorphism relies on a weight argument that is only sketched. These might be fixable, but they are not minor.\n\nWho should read it: people working on cohomology of Frobenius kernels or finite group schemes will want to know the spectral sequence construction, and the Borel computation is a nice model. But nobody should cite the SL2 result in its current form.\n\nMy recommendation: send to referees, not desk reject, because the method is novel and the errors look corrigible. But the referee report should be blunt: the central theorem is numerically false as written, and the ring structure proof needs a real rewrite. A revised version could be a solid paper.","headline":"The SL2 main theorem is internally inconsistent (HH0 multiplicity off by (p+1)/2); the spectral sequence approach is promising but the paper needs major correction.","tokens_in":14443,"tokens_out":5901,"would_cite":false,"duration_ms":48518,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["20G10","16E40","17B50","20G05"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper computes the full $G$-algebra structure of the Hochschild cohomology of the first Frobenius kernel of $SL_2$, for all primes $p$.","keywords":["Hochschild cohomology","Frobenius kernels","SL2","algebraic groups","spectral sequences","restricted enveloping algebra","tilting modules","G-algebra structure"],"falsifier":"Compute the $E_2$-differential in the spectral sequence for $N=T(2p-2)$ at $p=5$ on the class $e^*\\otimes f^{p-1}$: if this class already survives to $E_3$, or lands somewhere other than the predicted place in $S^{(p-1)/2}(\\mathfrak{u}^*)^{(1)}\\otimes x^{(p-1)/2}$, then the ideal $I_p$ in Theorem 4.4.2 is wrong; an explicit resolution of $\\overline{S}(\\mathfrak{g})_0$ as a $B_1$-module in low degrees would settle the question.","tokens_in":13283,"feed_emoji":"🧮","tokens_out":14921,"duration_ms":113083,"temperature":0.7,"pith_summary":"This paper aims to determine the Hochschild cohomology of the Frobenius kernels of an algebraic group, and it gives a complete answer for the first Frobenius kernel $G_1$ of $G=SL_2$ over any algebraically closed field of characteristic $p$. The authors build spectral sequences that compute $HH^*(G_r)$ from the cohomology of a Borel subgroup and the adjoint action on the distribution algebra, shifting the method away from explicit resolutions. The main result describes $HH^*(G_1)$ as a $G$-algebra: the even cohomology is built from copies of the coordinate ring of the nilpotent cone, the odd cohomology from specific induced modules, and the full ring is an explicit quotient of a symmetric algebra tensor a subalgebra of $\\mathfrak{u}$-cohomology invariants. A reader should care because this is the first complete $G$-algebra structure of Hochschild cohomology for a Frobenius kernel, and it shows precisely where the induction conjecture for $HH^*(G_r)$ is corrected for $SL_2$.","feed_headline":"Full Hochschild cohomology of SL2 Frobenius kernel computed","feed_subtitle":"Spectral-sequence method yields the full G-algebra structure of SL2 Frobenius kernel cohomology for all primes.","key_machinery":"The central object is the restricted enveloping algebra $\\operatorname{Dist}(G_1)_{\\mathrm{ad}}$ equipped with the adjoint action; for $SL_2$ this identifies with the truncated symmetric algebra $\\overline{S}(\\mathfrak{g})$ on $\\mathfrak{g}=\\mathfrak{sl}_2$, and its $G$-module decomposition into tilting modules $T(2n)$, the Steinberg module $L(2p-2)$, and copies of $k$ carries the cohomology computation. The argument is powered by a chain of spectral sequences (Theorems 2.2.1, 2.2.2, 2.2.3) that reduce $HH^*(G_1)$ first to $B_1$-cohomology, then to $T_1$-invariants in $\\mathfrak{u}$-cohomology. The load-bearing computational pieces are the tilting-module decomposition of $\\overline{S}(\\mathfrak{g})$ (Proposition 4.1.1), the explicit basis and multiplication of $H^\\bullet(\\mathfrak{u},\\overline{S}(\\mathfrak{g})_0)^{T_1}$ (Proposition 4.4.1), and the ideal $I_p$ that records the elements killed by the $E_3$ differentials. The technique of Drupieski, Nakano, and Ngo is used to 'ungrade' the collapsed spectral sequence and obtain a genuine ring isomorphism.","core_discovery":"The central claim is that for $G=SL_2$ the Hochschild cohomology of the first Frobenius kernel is completely determined by the $G$-structure of the truncated symmetric algebra $\\overline{S}(\\mathfrak{g})$, where $\\mathfrak{g}=\\mathfrak{sl}_2$. Using the identification $HH^*(G_1)\\cong H^*(G_1,\\operatorname{Dist}(G_1)_{\\mathrm{ad}})$, the authors decompose $\\operatorname{Dist}(G_1)_{\\mathrm{ad}}\\cong \\overline{S}(\\mathfrak{g})$ into $G$-summands $k$, $T(2p-2)$, and $L(2p-2)$ for $p\\ge 3$, and compute the $G_1$-cohomology of each summand. The resulting module structure is $HH^0(G_1)(-1)\\cong k^{\\oplus (p-1)/2}$, $HH^{2\\bullet}(G_1)(-1)\\cong k[\\mathcal{N}]^{\\oplus (p-1)}$ for $\\bullet>0$, and $HH^{2\\bullet+1}(G_1)(-1)\\cong [\\operatorname{ind}_G^B[S^\\bullet(\\mathfrak{u}^*)\\otimes \\omega]\\otimes L(1)]^{\\oplus (p-1)/2}$. For the ring structure, the spectral sequence through the Borel subgroup collapses only at the $E_3$ page, and the associated graded ring is explicitly identified as $[S^\\bullet(\\mathfrak{u}^*)^{(1)}\\otimes H^\\bullet(\\mathfrak{u},\\overline{S}(\\mathfrak{g})_0)^{T_1}]/I_p$; an ungrading argument then promotes this to an isomorphism of $B$-algebras, and induction from $B$ to $G$ gives the full $G$-algebra structure. The case $p=2$ is computed separately as a quotient $[S^\\bullet(\\mathfrak{u}^*)\\otimes \\overline{S}(\\mathfrak{g})]/I_2$.","pith_inferences":["The same spectral-sequence framework should extend to higher Frobenius kernels $G_r$ for $SL_2$; since the authors note the $r>2$ spectral sequence is not first-quadrant, testing $HH^*(G_2)$ for $p=3$ would reveal whether the $I_p$-type quotient pattern persists.","The failure of $E_2$-collapse is tied to the presence of the projective summand $T(2p-2)$ in $\\overline{S}(\\mathfrak{g})$, suggesting that for general reductive $G$ the deviation from the induction conjecture is measured by tilting modules whose highest weights lie in the upper alcove; one could test this by computing $H^\\bullet(B_1,T(\\lambda))$ for such $\\lambda$.","Because the ungrading argument uses the ring structure of $H^\\bullet(U_1,k)$, extending this computation to quantum groups at roots of unity would likely require a new ingredient, and the paper's restriction to Frobenius kernels is motivated by the relative simplicity of the adjoint action there.","The module decomposition implies a closed-form Hilbert series for $HH^*(G_1)$ in terms of the Hilbert series of $k[\\mathcal{N}]$ and of the induced modules, and checking this series against the $p=5$ and $p=7$ tables is a quick consistency test."],"forward_implications":["For $p\\ge 3$, the even-degree part of $HH^*(G_1)(-1)$ is exactly $(p-1)$ copies of $k[\\mathcal{N}]$ in positive degrees, so the Hilbert series of the even subring is known completely.","The odd-degree part is $(p-1)/2$ copies of $\\operatorname{ind}_G^B[S^\\bullet(\\mathfrak{u}^*)\\otimes \\omega]\\otimes L(1)$, giving the complete $G$-module structure of all odd cohomology groups.","The ring isomorphism $HH^\\bullet(G_1)(-1)\\cong \\operatorname{ind}_G^B H^\\bullet(B_1,\\overline{S}(\\mathfrak{g})_0)(-1)$ holds for all $p\\ge 2$; for $p\\ge 3$ the right-hand side is the quotient by $I_p$, and for $p=2$ the quotient by $I_2$.","The spectral sequence in Theorem 2.2.3(b) collapses at $E_3$, not $E_2$, for $N=T(2p-2)$, which is why the naive induction form $HH^*(G_1)\\cong S^\\bullet(\\mathfrak{u}^*)^{(1)}\\otimes H^\\bullet(\\mathfrak{u},\\operatorname{Dist}(G_1)_{\\mathrm{ad}})^{T_1}$ fails for $SL_2$.","The tables for $p=2,3,5,7$ give explicit $G$-module decompositions of every $HH^n(G_1)$ for small primes, providing concrete data that any alternative computation must match."],"supporting_citations":[{"why":"Supplies the cohomology ring $k[\\mathcal{N}]$ for $G_1$, the values of $H^\\bullet(G_1,L(2p-2))$, and the tilting-module and good-filtration facts used to decompose $\\overline{S}(\\mathfrak{g})$.","marker":"[AJ84]"},{"why":"Provides Kostant's theorem for Lie algebra cohomology of $\\mathfrak{sl}_2$, used to compute $H^1(\\mathfrak{u},L(\\lambda))$ in the Borel computation.","marker":"[UGA09]"},{"why":"Supplies the technique for promoting a collapsed spectral sequence to a ring isomorphism, which is what turns the $E_3$ collapse into the explicit quotient description of $H^\\bullet(B_1,\\overline{S}(\\mathfrak{g})_0)$.","marker":"[DNN12]"},{"why":"Provides the standard framework for Frobenius kernels, induction, Kempf vanishing, and $B_1T$-module structure that the spectral-sequence and induction arguments rely on throughout.","marker":"[Jan03]"}],"fun_headline_variants":["Full Hochschild cohomology for SL2 Frobenius kernel","SL2 Frobenius kernel: complete G-algebra structure","Spectral sequences yield full SL2 Frobenius kernel cohomology","Explicit G-algebra for SL2 Frobenius kernel cohomology"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the spectral sequence computing $H^\\bullet(B_1,\\overline{S}(\\mathfrak{g})_0)$ collapses at the $E_3$ page with exactly the differentials the authors describe, so that the quotient they write down is the actual ring and not merely a graded approximation.","fun_headline_variants_meta":{"raw":{"variants":["Full Hochschild cohomology for SL2 Frobenius kernel","SL2 Frobenius kernel: complete G-algebra structure","Spectral sequences yield full SL2 Frobenius kernel cohomology","Explicit G-algebra for SL2 Frobenius kernel cohomology"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000545,"raw_usage":{"total_tokens":2658,"prompt_tokens":1047,"completion_tokens":1611,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":663,"completion_tokens_details":{"reasoning_tokens":1534}},"tokens_in":663,"tokens_out":1611,"duration_ms":11889,"temperature":1.0,"reasoning_tokens":1534,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-12T14:10:28.009630+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the $E_2$-differential in the spectral sequence for $N=T(2p-2)$ at $p=5$ on the class $e^*\\otimes f^{p-1}$: if this class already survives to $E_3$, or lands somewhere other than the predicted place in $S^{(p-1)/2}(\\mathfrak{u}^*)^{(1)}\\otimes x^{(p-1)/2}$, then the ideal $I_p$ in Theorem 4.4.2 is wrong; an explicit resolution of $\\overline{S}(\\mathfrak{g})_0$ as a $B_1$-module in low degrees would settle the question.","supporting_citations":[],"review_version":1}