{"id":"3fc4b6aa-d64b-4363-b6bc-b8eb2b23164c","arxiv_id":"2412.08904","paper_version":2,"verdict":"REJECT","confidence":"MODERATE","novelty_score":3.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"Fei gives two-line proofs of Asai-Iyama's main results using his earlier theorem on tropical F-polynomials, and contends these results were already essentially known.","lead":"This note claims that the main theorems of a 2021/2024 paper by Asai and Iyama on semistable torsion classes are quick consequences of results the author published two years earlier. It provides terse 'two-line proofs' and argues that little essential new mathematics remains in the Asai-Iyama paper.","discovery_kind":"replication","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The promised reduction is not self-contained: Theorem 3.3 is asserted without proof, and Theorem 3.4 invokes [AI, Lemma 2.10] from the target paper, so the central claim that [AI] contains no essential content beyond [Ft] is not established.","rationale":"I agree with the reader's REJECT verdict, but the most load-bearing weakness differs from the reader's stated weakest assumption. The bridge between F(δ)/T-tilde(δ) in (2.1)-(2.2) and the BKT torsion pairs is asserted without proof, but it appears routine and likely fixable via Theorem 2.2, so it is not the decisive gap. The decisive problem is internal to the note's argument for its own central claim: Theorem 3.3 is unproved, and Theorem 3.4's proof explicitly depends on [AI, Lemma 2.10] from the very paper being reduced. This is a circular reliance relative to the meta-claim, and it is not remedied by the note's remark that it checked only the introduction of [AI]. The issue is not that the quoted theorems of [AI] are false, nor that [Ft, Theorem 2.2] is irrelevant; it is that the note does not actually demonstrate that those theorems follow from [Ft] alone. If the missing proofs were supplied and Lemma 2.10 were re-derived independently, the central claim could be rehabilitated; as written, the rejection stands.","tokens_in":4168,"tokens_out":14871,"duration_ms":156537,"concrete_test":"Extract the statement of [AI, Lemma 2.10] and attempt a full derivation using only [Ft, Theorem 2.2], the definitions in §2, and Lemma 2.3, without citing [AI]. Also write out the proof of Theorem 3.3 in full from Theorem 2.2. If deriving Lemma 2.10 requires importing an additional nontrivial statement from [AI], or if the proof of Theorem 3.3 needs a property not stated in §2 (such as additivity of hom over canonical decompositions), then the claimed reduction is not independent and the central claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The note's central claim is that the main results of Asai-Iyama are immediate consequences of [Ft, Theorem 2.2], leaving \"not much essential mathematical content\" in [AI]. This requires every reduction step to be derived from [Ft]/[Fc] or standard material without importing results from [AI]. The note fails this in two explicit places. First, Theorem 3.3 ([AI, Theorem 1.5]) is stated with no proof; the text says only that it is a \"simple consequence\" of Theorem 2.2 and the comment after Lemma 2.3. Since Theorem 3.3 is later used in the proof of Theorem 3.4, the two-line proof is not actually supplied. Second, the proof of Theorem 3.4 states: \"This is a straightforward consequence of Theorem 3.3 and [AI, Lemma 2.10]\", explicitly invoking a lemma from the target paper. Unless that lemma is independently reproved from [Ft, Theorem 2.2] or shown to be a trivial reformulation of it, the argument is circular with respect to the meta-claim: it uses part of [AI] to establish that modulo [Ft] there is nothing essential left in [AI]. The note's own Remark 4 further limits the audit to the introduction, so the scope of the priority claim is narrower than the abstract suggests.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The note claims that the main theorems of Asai-Iyama's paper [AI] are direct consequences of Fei's earlier theorem [Ft, Theorem 3.6] and that modulo that theorem 'there is not much essential mathematical content left in [AI].' It reviews definitions from [DF] and [BKT], recalls the tropical F-polynomial theorem from [Ft] and torsion-free/torsion classes from [Fc], then sketches proofs of [AI, Theorems 1.1, 1.3, 1.4, 1.5], appending remarks on priority and on the introduction of [AI].","tokens_in":4332,"tokens_out":7914,"duration_ms":66909,"significance":"If the reduction were fully proved, the note would establish that parts of [AI] are not independent of [Ft] and would provide a shorter route to several results. The paper also points to earlier work ([HKM], [DW1]) on the torsion theory. However, the note currently does not supply complete reductions for the key theorems; in one place it invokes a lemma from the very paper being reduced. Therefore the significance, as a proof of redundancy, is not currently realized. The strengths are the precise references and the observation that Theorem 2.2 may imply the torsion-pair identifications, but these need full derivations.","major_comments":[{"comment":"Theorem 3.3 is stated without a proof; the sentence that it is 'a simple consequence of Theorem 2.2 (see the comment after Lemma 2.3)' is not a derivation. Since Theorem 3.3 is used in the proof of Theorem 3.4, the reduction of [AI, Theorem 1.4] is incomplete. Please provide the promised two-line proof or explicitly state that the result is taken from [Ft].","section":"§3, Theorem 3.3"},{"comment":"The proof of Theorem 3.4 says it is 'a straightforward consequence of Theorem 3.3 and [AI, Lemma 2.10]'. Invoking a lemma from [AI], the target paper, makes the argument circular with respect to the note's meta-claim that modulo [Ft] there is little content left in [AI]. The phrase 'another easy observation made from the torsion theory of [Fc, Section 3.2]' does not supply the proof. The lemma must be stated and proved independently, or the claim about [AI] must be weakened.","section":"§3, Theorem 3.4"},{"comment":"The identification of the pairs (T(δ), F(δ)) and (ˇT(δ), ˇF(δ)) from (2.1)-(2.2) with the BKT torsion pairs (T_δ, F^δ) and (T^δ, F_δ) is asserted without proof. This identification is load-bearing for Theorems 3.2 and 3.3. In particular, the 'for some n' quantifier in (2.1) needs to be reconciled with the n-dependent statements in Theorem 2.2. A precise proof from Theorem 2.2 is required.","section":"§2, after Lemma 2.3"},{"comment":"The note states that the author does not check statements in [AI] beyond the introduction, while the abstract claims that the main results of [AI] are proved. This is a scope mismatch. The verification must either be extended to all main results or the abstract should be narrowed to the statements actually proved.","section":"§4, Remark 4"},{"comment":"For the hereditary case, the proof says it is 'essentially an easy consequence of results in [DW]' without further detail. Since Theorem 3.5 is among the main results listed in the abstract, this is another missing step. Provide the argument or state explicitly that the hereditary case is quoted from [DW] and not part of the reduction.","section":"§3, Theorem 3.5"}],"minor_comments":[{"comment":"There are several typos ('SEMIST ABLE' in the title, 'fro m' on page 1, 's left' in the abstract, 'central rule' for 'central role'). Please proofread.","section":"Title and throughout"},{"comment":"The display in Theorem 3.3 is garbled (e.g., 'T h /ℓθ' and 'W h /ℓθ'); the notation should be typeset properly so the reader can see which torsion classes are being intersected and unioned.","section":"§3, Theorem 3.3"},{"comment":"The notation K0(proj-A)_R in Definition 3.1 is not defined; specify that it is the real Grothendieck group of projective modules.","section":"§3, Definition 3.1"},{"comment":"The personal remarks in Section 4 about conference talks (Sanya 2019, Morningside 2021) are outside the mathematical scope of the note and could be removed or moved to a footnote; they do not affect the correctness of the claimed reductions.","section":"§4, personal remarks"}],"recommendation":"major_revision","confidential_remarks":"The core problem is that the current version does not prove the reductions it promises, and the circular use of [AI, Lemma 2.10] is serious. In my view the note is fixable: the author can add the missing proofs and explicitly delimit the claim. However, if the author declines to do so, the paper should not be published as a research note. The tone of the remarks about the [AI] paper may also need to be adjusted to a neutral scholarly discussion."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nYou should know this: Fei's note has a real point about priority, but the advertised \"two-line proofs\" don't stand up to scrutiny. The reduction is not self-contained, and the stress-test's circularity objection is accurate.\n\nWhat's genuinely useful: the note draws a clear link between the torsion pairs in Asai-Iyama and the hom-vanishing classes defined in Fei's earlier work. That connection is worth having explicitly, even if it's only asserted. The observation that Theorem 3.2 follows quickly from the definitions is fine. And the note is honest about not checking [AI] beyond the introduction, which actually narrows its own claim.\n\nThe soft spots are exactly where the reader puts them. Theorem 3.3 is central and is simply stated as a \"simple consequence\" with no proof; it's then used inside Theorem 3.4. More importantly, the proof of Theorem 3.4 invokes [AI, Lemma 2.10] directly. Even if that lemma is \"obvious from [Fc]\" as Fei says, a reduction that imports a lemma from the paper it claims to eliminate cannot support the statement that \"not much essential mathematical content is left in [AI].\" Also, the identification of the (T(δ), F(δ)) pairs with the [BKT] torsion pairs is just asserted; if that bridge has any subtlety, the rest of the note depends on it. These aren't manufactured flaws—they're load-bearing gaps.\n\nThat said, I don't think this note is worthless. It's a legitimate challenge to the historical narrative, and the underlying mathematics is presumably sound. But as a piece of research it doesn't deliver on its headline. It reads like a comment that belongs in a correspondence or a footnote, not a standalone paper.\n\nFor peer review: I'd send it to a competent referee who knows the torsion theory literature. The priority question matters, and a referee could quickly determine whether the gaps are fillable and whether the claims about [AI] are accurate. I wouldn't desk-reject it, but I'd expect heavy revision or conversion to a comment format.\n\nSo: worth a referee's time, but not something you'd cite for its math.","headline":"A sharp but under-supported priority note: the reduction to [Ft] is plausible yet circular in places, so the sweeping claim about [AI] doesn't land as written.","tokens_in":4975,"tokens_out":2559,"would_cite":false,"duration_ms":26406,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["16G20","18E40"],"pacs":[],"model":"deepseek-v4-flash","headline":"This note claims that the main theorems of a 2024 paper on semistable torsion classes and canonical decompositions are direct corollaries of an earlier theorem relating tropical $F$-polynomials to hom-spaces.","keywords":["semistable torsion classes","canonical decomposition","tropical F-polynomials","general presentations","weight vectors","torsion pairs","E-tame algebras","quiver representations"],"falsifier":"Take a small quiver algebra (for instance, the 3-Kronecker quiver) and a weight vector $\\delta$ with one positive and one negative coordinate, compute the class $\\{N \\mid \\hom(n\\delta,N)=0 \\text{ for some } n\\}$ and the class $\\{N \\mid \\delta(\\dim L)\\le 0 \\text{ for all subrepresentations } L\\}$, and check whether they coincide. Any module that lies in one class but not the other would break the central identification and with it every proof in the note.","tokens_in":3836,"feed_emoji":"🧮","tokens_out":11285,"duration_ms":100548,"temperature":0.7,"pith_summary":"This short note claims that the main theorems of a 2024 paper on semistable torsion classes and canonical decompositions are not independent results but direct corollaries of a theorem the author published two years earlier. The earlier theorem says that, for any representation and any weight vector, tropical $F$-polynomials evaluated along a ray eventually coincide with ordinary hom-space dimensions. Granting that, the 2024 paper's description of torsion classes, its canonical-decomposition theorem, and its criterion for $E$-tame equivalence all reduce to two-line arguments. The note also asserts that little mathematical content in the 2024 paper goes beyond that earlier theorem, and that some historical attributions in its introduction are inaccurate.","feed_headline":"Two-line proofs trace 2024 theorems to one 2023 identity","feed_subtitle":"A short note shows the 2024 torsion-class theorems follow directly from an earlier tropical F-polynomial identity.","key_machinery":"The machine is the asymptotic identity [Ft, Theorem 3.6]: for any representation $M$ and any weight vector $\\delta$, there is an $n$ such that $f_M(n\\delta)=\\hom(n\\delta,M)$ and $\\check f_M(-n\\delta)=e(n\\delta,M)$, and the same holds for every multiple $kn$. Here $f_M(\\delta)=\\max_{L\\hookrightarrow M}(\\dim L)\\cdot\\delta$ is the tropical $F$-polynomial, while $\\hom(n\\delta,M)$ is the dimension of the kernel of the map induced by a general presentation of weight $n\\delta$. The identity converts a maximum over subrepresentations into a hom-vanishing condition, which is exactly the kind of condition that defines the semistable torsion classes. The note also relies on the unproved identification of the resulting classes with [BKT]'s weight-vector torsion pairs.","core_discovery":"On the paper's own terms, the central claim is that [AI]'s main results are consequences of [Ft, Theorem 3.6]. The bridge is the assertion, made after Lemma 2.3, that the torsion-free class $\\mathcal F(\\delta)=\\{N \\mid \\hom(n\\delta,N)=0 \\text{ for some } n\\}$ and the torsion class $\\check{\\mathcal T}(\\delta)=\\{L \\mid e(n\\delta,L)=0 \\text{ for some } n\\}$ are exactly the weight-vector torsion pairs $(\\mathcal T_\\delta, \\mathcal F^\\delta)$ and $(\\mathcal T^\\delta, \\mathcal F_\\delta)$ introduced in [BKT]. Once that identification and the asymptotic identity of Theorem 2.2 are in place, each theorem in [AI]—the containment $[\\theta]_{TF}\\supseteq \\mathrm{cone}^\\circ(\\mathrm{Ind}\\,\\theta)$, the descriptions of $\\mathcal T_\\theta$ and $\\mathcal F_\\theta$ as intersections or unions over $\\ell$, and the $E$-tame equivalence criterion—is proved in a few lines. The note states this reduction explicitly: modulo [Ft], there is not much essential mathematical content left in [AI].","pith_inferences":["If the reduction is sound, the theory of semistable torsion classes could be rebuilt on tropical $F$-polynomials, making canonical decompositions of weight vectors a by-product of computing $\\hom(n\\delta,-)$ asymptotics.","The same asymptotic identity might extend to other stability frameworks wherever a hom-vanishing criterion determines the boundary of a torsion class.","A direct check of the torsion-pair identification on small quiver examples would settle whether the bridge is genuine or needs an additional argument; the note does not supply that check.","The note's priority remarks imply that the historical record in the 2024 paper's introduction should be revised if the two-line proofs are accepted."],"forward_implications":["The semistable torsion-class theorems of [AI] are corollaries of [Ft, Theorem 3.6], so any future proof or application can cite the one identity instead of the later paper's machinery.","If the torsion-pair identification is correct, the classes $\\mathcal T_\\delta$ and $\\mathcal F^\\delta$ are determined by the asymptotic vanishing of $\\hom(n\\delta,-)$ and $e(n\\delta,-)$, giving a uniform recipe for computing them.","The $E$-tame equivalence criterion—$\\eta$ and $\\theta$ are TF equivalent exactly when their indecomposable canonical summands agree—follows from the same identity.","For hereditary algebras the equality $[\\theta]_{TF}=\\mathrm{cone}^\\circ(\\mathrm{Ind}\\,\\theta)$ is traced to earlier work, not to [AI]."],"supporting_citations":[{"why":"The asymptotic identity between tropical F-polynomials and hom/e values that the note uses as its main tool.","marker":"[Ft, Theorem 3.6]"},{"why":"Supplies the fact that F(δ) is torsion-free and Ť(δ) is a torsion class, the basis for the torsion-pair identification.","marker":"[Fc, Lemma 3.8]"},{"why":"Introduced the weight-vector torsion pairs (T_δ, F^δ) and (T^δ, F_δ) that the note identifies with its hom-vanishing classes.","marker":"[BKT]"},{"why":"Provides the canonical-decomposition criterion via vanishing of e(δ_i, δ_j), used in the definitions and arguments.","marker":"[DF, Theorem 4.4]"},{"why":"The target paper whose main theorems are restated and then derived in a few lines.","marker":"[AI]"},{"why":"Introduced the homogeneity/ray condition that the note says was not new in [AI].","marker":"[DW1]"},{"why":"Earlier results invoked for the hereditary case of [AI, Theorem 1.3].","marker":"[DW]"}],"fun_headline_variants":["2024 torsion theorems are two-line corollaries of a 2023 identity","Two-line proofs: AI's torsion theorems follow from one tropical identity","AI's main results are corollaries of a 2023 F-polynomial identity","Short note: AI's torsion-class theorems reduce to one earlier identity","2023 identity explains AI's torsion results in two lines"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"Everything rests on the unproved claim that the torsion classes defined by vanishing hom-spaces after scaling a weight are the same as the torsion classes defined by sign conditions on every subrepresentation; if that identification fails, the two-line proofs of the 2024 theorems collapse.","fun_headline_variants_meta":{"raw":{"variants":["2024 torsion theorems are two-line corollaries of a 2023 identity","Two-line proofs: AI's torsion theorems follow from one tropical identity","AI's main results are corollaries of a 2023 F-polynomial identity","Short note: AI's torsion-class theorems reduce to one earlier identity","2023 identity explains AI's torsion results in two lines"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000959,"raw_usage":{"total_tokens":4029,"prompt_tokens":835,"completion_tokens":3194,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":451,"completion_tokens_details":{"reasoning_tokens":3098}},"tokens_in":451,"tokens_out":3194,"duration_ms":23374,"temperature":1.0,"reasoning_tokens":3098,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-11T17:27:12.284665+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take a small quiver algebra (for instance, the 3-Kronecker quiver) and a weight vector $\\delta$ with one positive and one negative coordinate, compute the class $\\{N \\mid \\hom(n\\delta,N)=0 \\text{ for some } n\\}$ and the class $\\{N \\mid \\delta(\\dim L)\\le 0 \\text{ for all subrepresentations } L\\}$, and check whether they coincide. Any module that lies in one class but not the other would break the central identification and with it every proof in the note.","supporting_citations":[],"review_version":1}