{"id":"a73aa838-1dee-4643-9217-5269b4481967","arxiv_id":"2607.14444","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Two free ergodic measure-preserving flows with isomorphic L¹ full groups are conjugate up to a constant time rescaling, which may reverse time.","lead":"This paper proves that the L¹ full group—a subgroup of symmetries of a measure-preserving flow built from its orbit structure—determines the flow up to rescaling time by a constant. It answers an open question in ergodic theory and completes the flow analogue of Belinskaya's classical theorem for measure-preserving transformations.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"No significant objection identified; the internal proof is sound, with the usual caveat that Theorem 3.1 relies on the cited flow Katznelson criterion [12, Thm. 10.9].","rationale":"The paper's new commensuration lemma is proved correctly, and the reduction in Theorem 3.1 is coherent and complete modulo the cited external results. The reader's weakest assumption correctly identifies the flow Katznelson criterion as a key external dependency; I agree that this is the main verification risk, but I do not see it as a demonstrated flaw. The proof's internal steps survive scrutiny, so the verdict need not change. The only minor point is the unstated measurability of the orientation dichotomy, which is immediate from Tonelli and therefore does not affect correctness.","tokens_in":9346,"tokens_out":32399,"duration_ms":304470,"concrete_test":"Retrieve the statement and full proof of [12, Thm. 10.9] and confirm that its hypotheses—free flows, shared orbits, equal orbit measures, and finite λ_x-measure of s1(x)△s2(x)—are exactly those delivered in Theorem 3.1, with no additional hidden condition such as eventual equality of half-orbits or a separate ergodicity assumption.","verdict_should_be":"UNCHANGED","load_bearing_attack":"I checked Lemma 2.1, Corollary 2.2, and the proof of Theorem 3.1 line by line. The bounds in Lemma 2.1 are correct, and Corollary 2.2 follows from them exactly as stated. In Theorem 3.1, the normalization to equal orbit measures, the construction of A_x, the derivation N(t)=∫λ(A_x△(A_x+t)), the subadditivity/Steinhaus integrability argument, and the use of Corollary 2.2 to force half-line commensurability are all valid. The R-invariance of the ray orientation is justified, and the apparent measurability gap in the dichotomy is closed by Tonelli. The final step applies Katznelson's criterion precisely as quoted. The only non-self-contained ingredient is [12, Thm. 10.9] (and, for Theorem 3.3, the spatial reconstruction results [12, Prop. 4.21], [12, Cor. 4.24]), which are cited from an accepted memoir. This is a verification limitation, not a detected mathematical error.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proves a continuous-time analogue of Belinskaya's theorem: if two free ergodic measure-preserving R-flows have L1 full groups that are isomorphic as abstract groups, then the flows are conjugate up to multiplication of time by a nonzero constant (with time reversal allowed). The main new tool is a commensuration criterion (Corollary 2.2) for measurable subsets A of R with ∫_0^1 λ(A△(A+t)) dt < ∞, asserting that A is commensurate with exactly one of ∅, R, [0,∞), and (-∞,0]. The proof of Theorem 3.1 normalizes the orbit measures, considers the fibers A_x = {u : x ≤ T_u x}, uses the L1-norm formula to transfer the bound to the average translated symmetric difference, applies Corollary 2.2, and then invokes the flow version of Katznelson's criterion to obtain conjugacy. Theorem 3.3 derives the abstract group isomorphism conclusion from known reconstruction results.","tokens_in":9505,"tokens_out":16467,"duration_ms":142271,"significance":"The result is significant: it answers two open questions from Le Maître–Slutsky [12] and shows that the L1 full group is a complete invariant for free ergodic flows up to scalar time rescaling, directly paralleling Belinskaya's theorem for Z-actions. The commensuration criterion is a clean, parameter-free statement with independent interest. The proof is largely self-contained: Lemma 2.1 and Corollary 2.2 are proved from scratch, and the application to flows is transparent. The only non-self-contained ingredients are the flow version of Katznelson's criterion ([12, Thm. 10.9]) and spatial reconstruction results ([12, Prop. 4.21], [12, Cor. 4.24]); these are cited from an accepted memoir and are not re-proved here, which is a verification limitation rather than a detected error. No circularity or fitted parameters are present.","major_comments":[],"minor_comments":[{"comment":"The R-invariance of the dichotomy 'A_x commensurate with [0,∞)' needs a measurability justification before ergodicity is applied. This follows by Tonelli's theorem since λ(A_x △ [0,∞)) = ∫ |1_{A_x}(u)-1_{[0,∞)}(u)| du is measurable in x, but the paper should state this explicitly.","section":"§3, proof of Theorem 3.1"},{"comment":"In the R-invariance step, the equality of the measure of {r : T_r x ∈ B} with λ_x(B) is used implicitly; this is exactly where the normalization λ'_x = λ_x is needed. Please make this explicit when saying that s(y) differing from s(x) by finite λ_x-measure implies A_y is commensurate with a translate of A_x.","section":"§3, proof of Theorem 3.1"},{"comment":"In the proof of the variation bound, the step '2∑C_k + 2∑S_k = 2I' uses ∑(S_k + C_k) = I; this is correct but could be stated for readability.","section":"§2, Lemma 2.1"}],"recommendation":"minor_revision","confidential_remarks":"The disclosure that the key results were discovered by an AI system is unusual; the author states that all proofs were independently verified. I see no mathematical issue, but the editor may wish to consider the journal's policy on AI-assisted research. The paper is a good fit for the journal's scope."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"The paper delivers what it says: for free ergodic measure-preserving flows, isomorphism of the L¹ full groups implies conjugacy up to constant time rescaling, and the one-sided inclusion version holds as well. This answers two open questions from Le Maître–Slutsky's memoir, so it is a genuine completion of the L¹ full group program for flows, not a repackaging.\n\nThe real new mathematics is in Section 2. Lemma 2.1's decomposition into the S_k and C_k pieces is a nice continuous analogue of the discrete variation argument, and the bound sum |a_{k+1}-a_k| ≤ 2I is correct. Corollary 2.2 is exactly the right commutation criterion: a subset of R with finite average symmetric difference against its translates is commensurate with one of the four canonical sets. I checked the proof line by line and found no gaps. The reduction in Theorem 3.1 — normalizing the orbit measures, introducing A_x, deriving N(t)=∫λ(A_x△(A_x+t)), then using subadditivity and Steinhaus to get the integrability needed for Corollary 2.2 — is sound. The ergodicity step that chooses between [0,∞) and (−∞,0] is justified; the measurability of that dichotomy is not spelled out, but it is immediate from the definition via Tonelli, so this is a minor exposition issue, not a mathematical one.\n\nThe main caveat is that the final conjugacy step leans on the flow version of Katznelson's criterion from the author's own memoir [12, Thm. 10.9], and Theorem 3.3 also depends on the spatial reconstruction results [12, Prop. 4.21] and [12, Cor. 4.24]. Those are black boxes here. They come from an accepted memoir, so this is a verification limitation rather than a detected error, but an editor should make sure a referee has access to those statements and can confirm the quotations. There is no circularity: Corollary 2.2 is proved from scratch and the target results are genuinely open questions being resolved, not assumptions.\n\nThe AI-provenance disclosure is transparent and honest. Since the author says he verified every step and the proofs are in the text, it neither helps nor hurts the mathematics; it is a footnote for the reader's information.\n\nWho is this for? Anyone working in orbit equivalence, full groups, or rigidity of measure-preserving flows. It is a short paper with a clear main theorem and a clever new lemma; it would make a good reading group talk. I would recommend accepting it after a referee checks the citations to [12] and the minor measurability point is cleaned up. Definitely deserves peer review, not a desk reject.","headline":"A clean, correct proof of the flow analogue of Belinskaya's theorem; the new commensuration criterion is real, the dependence on the author's earlier memoir is the only soft spot, and it should be sent to a serious referee.","tokens_in":10148,"tokens_out":1264,"would_cite":true,"duration_ms":15183,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37A10","37A05","37A15","37A20"],"pacs":[],"model":"deepseek-v4-flash","headline":"The paper proves that two free ergodic measure-preserving flows with abstractly isomorphic L¹ full groups are conjugate up to a non-zero scalar time change, making the L¹ full group a complete invariant of the flow up to rescaling and time","keywords":["L1 full groups","Belinskaya's theorem","measure-preserving flows","commensurated sets","Katznelson criterion","orbit equivalence","ergodic theory"],"falsifier":"Compute, for any measurable A ⊂ R, the integral I = ∫_0^1 λ(A △ (A+t)) dt and check whether A is commensurate with exactly one of ∅, R, (−∞,0], [0,∞). If an A with I < ∞ is commensurate with both half-lines or with neither, Corollary 2.2 fails and with it the proof of Theorem 3.1. Equivalently, at the flow level: exhibit two free ergodic flows sharing orbits with one contained in the other's L¹ full group whose half-orbits have infinite symmetric difference on a positive-measure set — that would directly contradict the paper's conclusion.","tokens_in":9095,"feed_emoji":"⏱","tokens_out":12727,"duration_ms":104912,"temperature":0.7,"pith_summary":"Belinskaya's theorem for integer actions says that the integrable full group of an ergodic measure-preserving transformation remembers the transformation up to flip conjugacy. This paper establishes the continuous-time analogue: for free ergodic measure-preserving flows, the L¹ full group — the transformations in the full group that move points a finitely integrable distance along their orbits — remembers the whole flow up to a constant rescaling of time. The proof shows that if two such flows share their orbits and the time maps of one belong to the other's L¹ full group, then their positive half-orbits are commensurate on almost every orbit after possibly reversing time; a flow version of Katznelson's criterion then upgrades commensuration to conjugacy. The new ingredient is a one-dimensional commensuration criterion: any measurable subset of the real line whose average symmetric difference with its unit translates is finite must be commensurate with exactly one of the empty set, the whole line, and the two half-lines. If the paper is correct, the L¹ full group is a complete algebraic invariant for free ergodic flows up to constant time rescaling.","feed_headline":"A flow's L1 full group remembers it up to time rescaling","feed_subtitle":"The L1 full group pins down a flow up to time rescaling and reversal, the continuous analogue of Belinskaya's theorem.","key_machinery":"The carrying machinery is the commensuration criterion of Section 2. For a measurable A ⊂ R, the lemma bounds the average displacement of A by sums over unit intervals: if I = ∫_0^1 λ(A △ (A+t)) dt is finite, then the sequence a_k = λ(A ∩ [k,k+1)) has finite total variation and its limits at ±∞ lie in {0,1}; hence A is commensurate with exactly one of ∅, R, (−∞,0], and [0,∞). The proof applies this fiberwise to A_x = {r ∈ R : x ≤ T_r x}, the set of times where the second flow's trajectory along the shared orbit is ahead of the first flow's. The L¹ norm formula — the norm of a full-group element equals the orbit-measure of the symmetric difference of the two half-orbit orderings — converts gl","core_discovery":"The central claim is that the L¹ full group is a complete invariant of a free ergodic measure-preserving flow up to rescaling the time parameter by a non-zero constant. Concretely, two free ergodic flows on possibly different probability spaces with abstractly isomorphic L¹ full groups are conjugate after one flow's time has been multiplied by some α ∈ R\\{0}; negative α amounts to time reversal. The argument combines a one-sided orbit-wise version — shared orbits plus containment of one flow in the other's L¹ full group forces commensurate positive half-orbits — with Katznelson's flow criterion, which says commensurability of right half-orbits yields conjugacy. Behind it all lies a new measu","pith_inferences":["Inference: the core lemma is purely one-dimensional and can be stress-tested independently of flows by computing I for explicit sets A, giving an early check of the proof's pivotal step.","Inference: the one-sided form of the theorem suggests that containment, not equality, is the natural rigidity condition; analogous one-sided rigidity may hold for other group actions equipped with a displacement norm.","Inference: because the proof works orbit by orbit, a natural next step is a rigorous non-ergodic version in which the rescaling factor becomes a measurable invariant function rather than a constant; the paper's remark points to this possibility."],"forward_implications":["The L¹ full group of a free ergodic flow determines the flow up to conjugacy followed by a constant rescaling of time, with time reversal allowed.","Two free ergodic flows with the same orbits where one's time maps belong to the other's L¹ full group are conjugate after a scalar time change — no equality of full groups is needed.","Any measurable subset of the real line with finite average symmetric difference against its unit translates is commensurate with exactly one of the empty set, the whole line, or one of the two half-lines; this criterion is now available for use in other settings.","For ergodic flows, abstract group isomorphism of L¹ full groups upgrades previously known flip Kakutani equivalence to the stronger conclusion of conjugacy up to scalar time change."],"fun_headline_variants":["L1 full group determines flow up to time rescaling","Flows with isomorphic L1 full groups are time-rescaled","Belinskaya's theorem extends to measure-preserving flows","L1 full group remembers flow up to rescaling and reversal","Free ergodic flows: L1 full groups pin down time change"],"cache_read_input_tokens":2304,"weakest_assumption_plain":"The argument's load-bearing premise is the imported flow version of Katznelson's conjugacy criterion — that commensurate right half-orbits force the flows to be conjugate — together with the reconstruction of orbit equivalence from an abstract isomorphism of L¹ full groups; if either imported result fails, the chain from L¹ containment to conjugacy breaks.","fun_headline_variants_meta":{"raw":{"variants":["L1 full group determines flow up to time rescaling","Flows with isomorphic L1 full groups are time-rescaled","Belinskaya's theorem extends to measure-preserving flows","L1 full group remembers flow up to rescaling and reversal","Free ergodic flows: L1 full groups pin down time change"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000615,"raw_usage":{"total_tokens":2685,"prompt_tokens":725,"completion_tokens":1960,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":469,"completion_tokens_details":{"reasoning_tokens":1884}},"tokens_in":469,"tokens_out":1960,"duration_ms":12872,"temperature":1.0,"reasoning_tokens":1884,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-02T02:07:32.577762+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute, for any measurable A ⊂ R, the integral I = ∫_0^1 λ(A △ (A+t)) dt and check whether A is commensurate with exactly one of ∅, R, (−∞,0], [0,∞). If an A with I < ∞ is commensurate with both half-lines or with neither, Corollary 2.2 fails and with it the proof of Theorem 3.1. Equivalently, at the flow level: exhibit two free ergodic flows sharing orbits with one contained in the other's L¹ full group whose half-orbits have infinite symmetric difference on a positive-measure set — that would directly contradict the paper's conclusion.","supporting_citations":[],"review_version":1}