{"id":"684fb9d2-8bc8-4d58-990f-f49d6bdacccc","arxiv_id":"1908.03859","paper_version":2,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Over characteristic 0 fields, the unit map from the motivic sphere spectrum to the special linear cobordism spectrum MSL is an isomorphism on homotopy modules, proven by comparing framed and SL-oriented framed correspondences.","lead":"This paper proves that the unit map from the motivic sphere spectrum to the special linear cobordism spectrum is an isomorphism on homotopy modules over fields of characteristic 0. The proof works by comparing both sides through explicit geometric generators and relations, using a new SL-oriented version of framed correspondences.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The unit-map comparison is internally consistent; its main risk is inherited from the imported infinite loop-space identification Theorem 2.2.5, so the conclusion is only as secure as that external input.","rationale":"The paper's own proof of Theorem 3.6.1 is careful and internally consistent. Surjectivity is proven by explicit A^1-homotopies that move an SL-oriented framing into the fiber over the distinguished point, and injectivity is proven by constructing a left inverse α_SL through oriented Thom classes and finite Milnor-Witt correspondences. The algebra in Lemma 4.3.3 is checked on ring generators, and the definitions of ZFSL and the comparison map are coherent. The main weakness is not in the internal argument but in the imported Theorem 2.2.5, which is what connects the framed-correspondence groups to actual homotopy groups of MSL. The reader identified this same external dependency as the weakest assumption; I agree that it is the most load-bearing point, though I would sharpen it to the specific colimit passage in Corollary 3.2.2, where the loop-space equivalence for each finite Thom spectrum is assembled into the equivalence for MSL. This is a standard type of check and not evidence of a flaw. If the colimit square is verified, the central claim follows from the given internal arguments. I therefore see no reason to change the ACCEPT verdict.","tokens_in":24192,"tokens_out":26694,"duration_ms":290352,"concrete_test":"Write out the n→n+1 compatibility square for the equivalences Θ_{~T_n} of Theorem 2.2.5 and prove that its colimit is the equivalence Θ_~T of Corollary 3.2.2, explicitly justifying commutation of group completion and sheafification with the filtered colimit over n for the presheaves Fr_{~T_n}(-,~Gr_n); if this square commutes up to equivalence, the identification of ε_* with the unit map on π0 is established.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The single load-bearing point is the imported identification [EHK+19b, Cor. 3.2.4] / [Yak19, Thm 2.2.2] used as Theorem 2.2.5: the infinite P^1-loop space of the Thom spectrum of a rank-r vector bundle is equivalent to the group-completed A^1-localized E-framed correspondence presheaf. Proposition 3.4.8 and Corollary 3.4.2 convert the unit map of MSL into the map ε_* on H0(ZF)-type groups precisely through this identification; if it failed for the bundles ~T_n→~Gr_n, the map ε_* compared in Theorem 3.6.1 would not be the unit map on homotopy groups. The paper includes a proof sketch of Theorem 2.2.5, but the argument is delegated to external references. The internal proof of Theorem 3.6.1—surjectivity via explicit A^1-deformations in Proposition 4.1.5 and injectivity via the left inverse α_SL in Section 4.3—is coherent; the only sub-step that deserves close checking is the passage to the colimit over n in Corollary 3.2.2, where the loop-space equivalence is applied to each Thom spectrum before taking MSL as a colimit. This is not an internal inconsistency, but it is the least secure link in the chain.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves that for a field k of characteristic 0, the unit map e: 1_k -> MSL induces an isomorphism of graded homotopy modules pi_0(1_k)_* -> pi_0(MSL)_* (Theorem 3.6.1 together with Proposition 3.6.3). The strategy is to use E-framed correspondences to present pi_0(1_k)_l(k) and pi_0(MSL)_l(k) as zeroth homology groups of complexes of linear framed and SL-oriented framed correspondences, respectively, and then to compare these presentations directly. Surjectivity of the comparison map epsilon_* is proved by explicit A^1-deformations that move any SL-oriented framed correspondence into the image of the usual framed correspondences (Proposition 4.1.5); injectivity is proved by constructing a left inverse alpha_SL using oriented Thom classes and finite Milnor-Witt correspondences (Section 4.3). As applications, the paper derives uniqueness of special linear orientations for Chow-Witt groups and for MW-motivic cohomology (Corollaries 3.6.5 and 3.6.7).","tokens_in":24477,"tokens_out":12330,"duration_ms":140762,"significance":"If correct, the result gives a geometric, generator-and-relations proof of an isomorphism that is foundational for the structure of the algebraic special linear cobordism spectrum. The main strengths of the paper are its concreteness and internal coherence: surjectivity is established by explicit A^1-homotopies, and injectivity by a transparent left inverse rather than by an indirect comparison. The paper is also honest about its reliance on substantial external machinery, above all Theorem 2.2.5 imported from [EHK+19b] and the computations of Neshitov and of Calmes-Fasel. Within those dependencies, the logical chain from the unit map of MSL to the map epsilon_* is carefully laid out. The applications to unique SL-orientations are natural and clearly derived from the main theorem.","major_comments":[],"minor_comments":[{"comment":"The passage from the colimit over n of the equivalences of Theorem 2.2.5 to an equivalence with Maps(Sigma_T^infty(-)_+, MSL) is stated in one sentence; please spell out that compactness of Sigma_T^infty X_+ is used to commute the colimit over n with the mapping space, just as it is used for the colimit over p in Section 3.2.1. This is the precise point at which the map epsilon_* is identified with the actual unit map of MSL, so making the compactness argument explicit would remove a potentially delicate step.","section":"Section 3.2, Corollary 3.2.2"},{"comment":"The functor alpha_SL is defined on SL-oriented framed correspondences and it is asserted that it factors through stabilization and descends to H_0(ZF^SL(Delta^bullet_k, G_m^{wedge l})); please add a sentence explaining compatibility with the disjoint-union relation and with the A^1-homotopy relation, since this descent is what makes alpha_SL a left inverse on homology rather than only on generators.","section":"Section 4.3, proof of Theorem 3.6.1"},{"comment":"The proof begins by assuming that the correspondence has level n > 0, but the reduction from level 0 is only implicit; state explicitly that every class can be represented by a suspension, so that the n > 0 case implies surjectivity for all levels.","section":"Section 4.1, Proposition 4.1.5"},{"comment":"There is a typo in 'a a framed correspondence' in the paragraph about Voevodsky's framed correspondences; it should read 'a framed correspondence'.","section":"Introduction, page 2"},{"comment":"The ring structure on H_0(ZF^SL(Delta^bullet_k, G_m^{wedge *})) is said to be constructed by the same argument as in [Nes18, Section 3]; a brief indication of the product formula for SL-oriented framed correspondences, analogous to the product of framed correspondences, would improve readability.","section":"Section 3.5.2"}],"recommendation":"minor_revision","confidential_remarks":"The main theorem is conditional on the imported infinite loop-space identification Theorem 2.2.5, whose proof is delegated to [EHK+19b] and [Yak19]. I do not see circularity or an internal inconsistency, but it may be worth confirming the publication status of those references before acceptance, since they carry a load-bearing part of the argument."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Colleague,\n\nThe thing to know: the headline result—ε_* iso—was already stated in BH18 Example 16.34, but the paper's actual contribution is the explicit geometric presentation: SL-oriented framed correspondences, the unit map written as a map on H0 of ZF complexes, and a direct proof by deforming SL-oriented framings into ordinary ones. That is genuinely new and is the part worth reading carefully.\n\nIt does it well. Proposition 4.1.5 is a concrete A^1-deformation argument for surjectivity; injectivity comes from a left inverse constructed from oriented Thom classes and finite MW-correspondences. The comparison with Neshitov and Calmès-Fasel is handled cleanly, and the diagram chase in 4.3.2 is legitimate. I checked the generator-by-generator identification in Lemma 4.3.3 and it works. The circularity burden is low: the proof constructs α_SL explicitly and never assumes the desired isomorphism.\n\nThe soft spot is not internal; it is the heavy dependence on Theorem 2.2.5, imported from EHK+19b and the author's thesis. That theorem identifies infinite P^1-loop spaces of Thom spectra with group-completed A^1-localized E-framed correspondence presheaves. Everything in Section 3.4 routes through it. The paper gives only a proof sketch and delegates. If that theorem has a hidden flaw, this paper's conclusion about the actual unit map does not follow. The stress-test note is right to point at the colimit over n in Corollary 3.2.2—passing the loop-space equivalence through a colimit of Thom spectra deserves a careful check, though I don't see an actual error. I would flag it as a place to ask the author for more detail, not as a known gap.\n\nThe char 0 assumption is another caveat, inherited from Neshitov. The remark about Druzhinin-Kylling extending to odd characteristic is honest and correct.\n\nBottom line: for someone working in motivic homotopy theory or with framed correspondences, this is a useful and trustworthy paper. It does not pretend to prove more than it does, and it credits the external inputs clearly. I would send it to a serious referee—the kind of referee who will actually check Theorem 2.2.5's applicability rather than just the internal algebra. I would not desk reject it.","headline":"A careful and explicit framed-correspondence proof that the MSL unit map is an isomorphism on homotopy modules in characteristic zero; the main risk is inherited from an imported infinite-loop-space theorem.","tokens_in":25059,"tokens_out":2141,"would_cite":true,"duration_ms":24450,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14F42","14F99"],"pacs":[],"model":"deepseek-v4-flash","headline":"Over characteristic-zero fields the unit map of MSL is an isomorphism on zeroth homotopy modules.","keywords":["framed correspondences","special linear cobordism","MSL","Milnor-Witt K-theory","homotopy modules","motivic Thom spectra","MW-motivic cohomology","oriented Grassmannian"],"falsifier":"Find a single explicit SL-oriented framed correspondence over a characteristic-zero field that is provably not $\\mathbb{A}^1$-homotopic to a standard framed correspondence, for example by computing a nontrivial invariant in the cokernel of the natural inclusion $Fr_n \\to Fr^{SL}_n$; if such a class exists, surjectivity of $\\varepsilon_*$ fails. A more numerical check: for $k = \\mathbb{Q}$ and $l=1$ the theorem predicts $\\pi_0(\\mathrm{MSL})_1(\\mathbb{Q}) \\cong \\mathbb{Q}^\\times$, and the explicit presentation via SL-oriented framed correspondences should have no torsion; a direct computation exhibiting an element of order $2$ in that presentation would falsify the theorem.","tokens_in":23941,"feed_emoji":"","tokens_out":15587,"duration_ms":145833,"temperature":0.7,"pith_summary":"This paper establishes that, over a field of characteristic zero, the algebraic special linear cobordism spectrum MSL has the same zeroth stable homotopy sheaf as the motivic sphere spectrum. The unit map $1_k \\to \\mathrm{MSL}$ induces an isomorphism of homotopy modules $\\pi_0(1_k)_* \\cong \\pi_0(\\mathrm{MSL})_*$. To reach this conclusion, the paper rewrites both sides in geometric generators and relations: the zeroth homotopy group is expressed through framed correspondences for the sphere spectrum and through special-linear-oriented framed correspondences for MSL. It then proves the induced map between the two zeroth homology groups is a graded ring isomorphism, using explicit $\\mathbb{A}^1$-homotopies for surjectivity and finite Milnor-Witt correspondences for injectivity. This makes the comparison between the sphere spectrum and MSL explicit rather than formal.","feed_headline":"MSL's unit map is an isomorphism over characteristic 0","feed_subtitle":"Over characteristic-zero fields the sphere and MSL share zeroth homotopy sheaves, uniquely orienting Chow-Witt cohomology.","key_machinery":"The central object is the category of special-linear-oriented framed correspondences $Fr^{SL}_*(k)$. An ordinary framed correspondence of level $n$ from $X$ to $Y$ cuts out its support as the zero locus of a map $\\varphi : U \\to \\mathbb{A}^n$; an SL-oriented correspondence instead uses a map $\\varphi : U \\to \\tilde{T}_n$ into the tautological bundle over the oriented Grassmannian $\\tilde{Gr}_n$, cutting out the support as the preimage of the zero section. A recognition theorem for infinite loop spaces of motivic Thom spectra identifies $\\pi_0(\\mathrm{MSL})_l(k)$ with $H_0(ZF^{SL}(\\Delta^\\bullet_k, \\mathbb{G}_m^{\\wedge l}))$, and the corresponding recognition theorem for suspension spectra identifies $\\pi_0(1_k)_l(k)$ with $H_0(ZF(\\Delta^\\bullet_k, \\mathbb{G}_m^{\\wedge l}))$. The proof that the induced map $\\varepsilon_*$ is an isomorphism has two mechanisms: explicit $\\mathbb{A}^1$-homotopies deform any SL-oriented framing into the distinguished affine fiber, proving surjectivity; and a functor from SL-oriented framed correspondences to finite Milnor-Witt correspondences, built from oriented Thom classes, composes with the comparison between Milnor-Witt K-theory and MW-motivic cohomology to give a left inverse, proving injectivity.","core_discovery":"Let $k$ be a field of characteristic $0$. The paper's main theorem (3.6.1) states that the unit map at the level of framed correspondences is a graded ring isomorphism $$\\varepsilon_* : H_0(ZF(\\$\\Delta$^\\bullet_k, \\mathbb{G}$_m^{{\\wedge *}}$)) \\xrightarrow{\\sim} H_0($ZF^{{SL}}$(\\$\\Delta$^\\bullet_k, \\mathbb{G}$_m^{{\\wedge *}}$)).$$ Here $ZF$ is the stabilized free abelian group on framed correspondences and $ZF^{SL}$ its special-linear-oriented variant; the upper index records the power of the multiplicative group used as the target. Proposition 3.6.3 converts this into a spectrum-level statement: the unit map $e : 1_k \\to \\mathrm{MSL}$ induces an isomorphism of homotopy modules $\\pi_0(1_k)_* \\cong \\pi_0(\\mathrm{MSL})_*$, where a homotopy module is the motivic analogue of the sequence of stable homotopy groups of a spectrum. The paper then draws two consequences: the Chow-Witt cohomology theory $H^*(-, K^{MW}_*)$ and MW-motivic cohomology $H^{*,*}_{MW}(-, \\mathbb{Z})$ each carry a unique special linear orientation.","pith_inferences":["We infer that the same correspondence-level strategy should compute the unit map for other Thom spectra built from oriented vector bundles, provided the relevant complement is $\\mathbb{A}^1$-chain connected as it is for the oriented Grassmannian.","We infer that the characteristic-zero assumption enters through the cited identification of framed-correspondence homology with Milnor-Witt K-theory, since the paper's own deformation argument does not invoke characteristic zero; a preprint extension noted by the paper would then carry the theorem to odd characteristic after inverting the characteristic.","We infer that the explicit deformation proof could be turned into a practical test for equality of low-degree classes in $\\pi_0(\\mathrm{MSL})$: reduce to Milnor-Witt K-theory via the left inverse, then check the difference against the known relations.","We infer that the uniqueness of the special linear orientation may extend to any effective cohomology theory represented by a homotopy module in the heart of the homotopy $t$-structure, since only the isomorphism $\\pi_0(\\mathrm{MSL})_* \\cong \\pi_0(1)_*$ is used."],"forward_implications":["Over characteristic-zero fields the zeroth homotopy sheaves of MSL are canonically the sheaves of Milnor-Witt K-theory, since they agree with those of the sphere spectrum.","The Chow-Witt cohomology theory $H^*(-, K^{MW}_*)$ carries a unique special linear orientation.","The MW-motivic cohomology spectrum carries a unique special linear orientation.","Unlike the algebraic cobordism spectrum MGL, whose unit map kills the motivic Hopf element and factors through $1/\\eta$, MSL needs no $\\eta$-quotient for its unit map to be an isomorphism on $\\pi_0$.","The geometric presentation gives explicit generators for $\\pi_0(\\mathrm{MSL})_l(k)$: SL-oriented framed correspondences modulo $\\mathbb{A}^1$-homotopy and suspension, with relations governed by Milnor-Witt K-theory."],"supporting_citations":[{"why":"Supplies the recognition theorem for infinite loop spaces of motivic Thom spectra (Corollary 3.2.4), which converts the zeroth homotopy group of MSL into homology of SL-oriented framed correspondences.","marker":"[EHK+19b]"},{"why":"Provides the alternative proof of the same recognition theorem cited in the proof of Theorem 2.2.5.","marker":"[Yak19]"},{"why":"Proves the framed-correspondence description of zeroth stable homotopy groups of suspension spectra (Corollary 11.3), the model for the sphere side of the comparison.","marker":"[GP18a]"},{"why":"Computes $H_0(ZF(\\Delta^\\bullet, \\mathbb{G}_m^{\\wedge *}))$ as non-negative Milnor-Witt K-theory, the key identification used in the injectivity argument.","marker":"[Nes18]"},{"why":"Provides the comparison theorem identifying MW-motivic cohomology with Milnor-Witt K-theory, used to make the left inverse land in a known ring.","marker":"[CF17b]"},{"why":"Constructs the category of finite Milnor-Witt correspondences and the twisted Chow-Witt machinery used to define the left inverse.","marker":"[CF17a]"},{"why":"Defines the functor from framed correspondences to finite Milnor-Witt correspondences and its properties (Proposition 2.1.12), which the proof extends to the SL-oriented setting.","marker":"[DF17]"},{"why":"Supplies the alternative construction of framed correspondences via oriented Thom classes used to define the SL-oriented functor.","marker":"[EHK+19a]"},{"why":"Constructs MSL as a Thom spectrum over oriented Grassmannians and proves its universal property for special linear orientations, used for the corollaries.","marker":"[PW10]"},{"why":"Provides the theory of strictly $\\mathbb{A}^1$-invariant sheaves and unramified sheaves used to promote field-wise isomorphisms to an isomorphism of homotopy modules.","marker":"[Mor12]"}],"fun_headline_variants":["MSL unit map is an isomorphism over char 0","Over char 0, MSL unit map is an isomorphism","Unit map of MSL is iso on homotopy sheaves in char 0","Unique SL orientation from MSL unit isomorphism in char 0"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The result rests on the imported identification that translates the zeroth homotopy group of a motivic Thom spectrum of a rank-zero vector bundle into the zeroth homology of a group-completed, $\\mathbb{A}^1$-localized framed-correspondence complex; if that identification fails, the map $\\varepsilon_*$ shown to be an isomorphism is not the unit map of MSL on homotopy groups.","fun_headline_variants_meta":{"raw":{"variants":["MSL unit map is an isomorphism over char 0","Over char 0, MSL unit map is an isomorphism","Unit map of MSL is iso on homotopy sheaves in char 0","Unique SL orientation from MSL unit isomorphism in char 0"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000563,"raw_usage":{"total_tokens":2652,"prompt_tokens":905,"completion_tokens":1747,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":521,"completion_tokens_details":{"reasoning_tokens":1684}},"tokens_in":521,"tokens_out":1747,"duration_ms":11734,"temperature":1.0,"reasoning_tokens":1684,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T14:00:21.115009+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Find a single explicit SL-oriented framed correspondence over a characteristic-zero field that is provably not $\\mathbb{A}^1$-homotopic to a standard framed correspondence, for example by computing a nontrivial invariant in the cokernel of the natural inclusion $Fr_n \\to Fr^{SL}_n$; if such a class exists, surjectivity of $\\varepsilon_*$ fails. A more numerical check: for $k = \\mathbb{Q}$ and $l=1$ the theorem predicts $\\pi_0(\\mathrm{MSL})_1(\\mathbb{Q}) \\cong \\mathbb{Q}^\\times$, and the explicit presentation via SL-oriented framed correspondences should have no torsion; a direct computation exhibiting an element of order $2$ in that presentation would falsify the theorem.","supporting_citations":[],"review_version":1}