{"id":"64283a47-ad60-42d5-967e-ca16d5636367","arxiv_id":"2607.10289","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Bump–Friedberg-type periods extend continuously beyond the cuspidal spectrum under regularity, and on GL_{2n+1} Eisenstein series they equal a sum of L-values indexed by dual-variety fixed points.","lead":"The paper extends classical Bump–Friedberg periods from cuspidal forms to general automorphic forms of moderate growth, and evaluates a new SL_{n+1}×GL_n period on certain Eisenstein series as a sum of L-values. This matches the fixed-point prediction of the Ben-Zvi–Sakellaridis–Venkatesh numerical conjecture for a candidate dual variety.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader correctly isolates the dual-variety matching as the sole conditional claim and notes that the period evaluation itself stands independently. The analytic core (Theorems 3.1.5.1, 4.1.3, 5.1.5.1 and the evaluation 5.4.1) rests on classical tools whose hypotheses are made explicit by the regularity conditions; those conditions are used precisely to kill residual orbits, so the argument is self-contained. No free parameters or circular steps appear. The concrete check for n=1 is a low-cost verification of the most delicate bookkeeping step and does not alter the overall assessment. Hence the ACCEPT verdict with high confidence remains appropriate.","tokens_in":46207,"tokens_out":527,"duration_ms":6504,"concrete_test":"Independently re-derive the constant-term contribution of a single Weyl element w_i (i∈J_{1}(π)) for the smallest case n=1 (G=GL_{3}, H=SL_{2}\times GL_{1}) and check that the resulting local zeta integral at the unramified places reproduces exactly L(1,η_i^{-1}⊗Π_i)L(1,η_i⊗Π_i^∨)L(1/2,Π_i,η_i^{-1}⊗∧^{2}) after the normalizations of §5.4; agreement confirms the Euler-factor bookkeeping.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The period formula of Theorem 5.4.1 is derived by standard unfolding, constant-term, and Euler-product arguments under the stated Δ_{1}*-regularity; the only conditional step is the dual-variety comparison in §5.5, which the paper already flags as relying on the hypothetical global Langlands correspondence and a candidate ˇX motivated by TWZ26. That comparison is not needed for the analytic identity itself. No internal inconsistency or hidden analytic gap appears in the load-bearing steps (absolute convergence of the zeta integrals, vanishing of residual orbits under regularity, reduction via Iwasawa to the augmented period, and the Langlands constant-term decomposition indexed by J_{1}(π)).","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper extends Bump–Friedberg type periods beyond the cuspidal spectrum. For the twisted period on GL_{2n} and an augmented variant on GL_1×GL_{2n}, under (η,Δ*)- or Δ_a*-regularity of the cuspidal datum, the periods defined on automorphic Schwartz functions extend continuously to functions of uniform moderate growth; the extensions are given by entire Whittaker-type zeta integrals obtained by unfolding, residual-orbit vanishing, and functional equations (Theorems 3.1.5.1, 4.1.3). A new period on GL_{2n+1} over SL_{n+1}×GL_n is then introduced. For Δ_1*-regular Eisenstein series E(φ), after normalization, P*(E(φ),Φ) equals a finite sum over i∈J_1(π) of products of partial L-values L^S(1,η_i^{-1}⊗Π_i)L^S(1,η_i⊗Π_i^∨)L^S(1/2,Π_i,η_i^{-1}⊗∧²) times normalized local zeta integrals (Theorem 5.4.1 / (1.3.1)). Under the hypothetical global Langlands correspondence, the summands match the fixed points and tangent-space L-factors of a candidate dual variety ˇX motivated by TWZ26, in line with the BZSV global numerical conjecture.","tokens_in":46485,"tokens_out":1032,"duration_ms":11641,"significance":"The work supplies a concrete, non-cuspidal instance of the multi-fixed-point phenomenon predicted by the BZSV numerical conjecture, with an explicit period formula whose L-factors match the expected tangent-space contributions. The analytic core—absolute convergence of the zeta integrals, vanishing of residual periods under the stated regularity, reduction via Iwasawa to the augmented period, and Langlands constant-term decomposition indexed by J_1(π)—is standard but carefully executed and of independent interest for relative Langlands beyond the cuspidal spectrum. The dual-variety comparison is correctly flagged as conditional; the period identity itself does not depend on it. Strengths include the systematic use of regularity to kill residual orbits and the transparent reduction chain from the new period to the classical twisted Bump–Friedberg integral.","major_comments":[],"minor_comments":[{"comment":"In §1.3 and Remark 1.3.2 the normalization conventions for W^M_φ and W_{E(φ)} are stated carefully, but a short explicit sentence early in §5.4 recalling which global L-factor is absorbed by which normalization would help the reader track the appearance/disappearance of L(1,π,ˆn_P^-).","section":null},{"comment":"The candidate dual variety ˇX is introduced only in §5.5 with a citation to TWZ26 Table 14 Line 19. A one-sentence pointer already in the introduction (or in the statement of Theorem 1.3.1) would make the compatibility claim easier to locate.","section":null},{"comment":"Several lemmas (e.g. 3.3.3, 4.2.4, 5.3.3) are declared “identical” or “similar” to earlier ones and left to the reader. A brief indication of the single changed numerical relation (the +1 or +2 in the sum of r^+ versus r^-) would improve readability without lengthening the text much.","section":null},{"comment":"Typographical: “ad´ eles”, “Fr´ echet”, “math´ ematiques” appear with broken accents in the source; these should be cleaned for the published version. Also “ene can check” (p. 37) → “one can check”.","section":null},{"comment":"In Definition 3.1.4.1(2) the holomorphy of L(x,π_i,∧²⊗η^{-1}|·|^s) at x=1 is required for even n_i; a parenthetical remark that this is automatic for odd n_i (by the functional equation or known non-vanishing) would clarify why the condition is stated only for even blocks.","section":null}],"recommendation":"accept","confidential_remarks":"The manuscript is a solid, technically careful contribution that fits well in a representation-theory / automorphic-forms journal. The dual-variety discussion is appropriately hedged; I see no reason to demand a proof of duality that is outside the paper’s scope. The self-citation density (LX25, LXZ25) is high but the lemmas used are standard technical tools rather than circular inputs. I recommend acceptance with only light copy-editing."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper does the analytic work needed to push twisted Bump–Friedberg periods off the cuspidal spectrum and then evaluates a new SL_{n+1}\times GL_n period on GL_{2n+1} Eisenstein series as an explicit finite sum of L-values. That is the real content.\n\nWhat is new is clean. Under (η,Δ*)-regularity the classical period on Schwartz functions extends continuously to the uniform-moderate-growth space T_χ and is recovered as the value at (0,0) of an entire Whittaker-type zeta integral (Thm 3.1.5.1). The same strategy produces an “augmented” period on GL_1\times GL_{2n} and, after Iwasawa reduction of the constant term along P_{(1,2n)}, a new period on GL_{2n+1}. For Δ_1*-regular Eisenstein series the latter evaluates to a sum over the G_1-blocks of products L^S(1,η_i^{-1}⊗Π_i)L^S(1,η_i⊗Π_i^∨)L^S(1/2,Π_i,η_i^{-1}⊗∧^{2}) times normalized local factors (Thm 5.4.1). The absolute-convergence lemmas, residual-orbit vanishing under the regularity hypotheses, functional equations via Poisson, and Langlands constant-term reduction are all standard but carefully written; the Euler product step simply quotes the known unramified computation.\n\nThe only soft spot is the dual-variety comparison in §5.5. It assumes both the global Langlands correspondence and that the proposed ˇX is actually dual to X; both are flagged as conjectural and are not used to derive the period identity itself. If either fails the numerical-conjecture match collapses, but the analytic formula remains. That is a minor, clearly labelled limitation, not a load-bearing flaw.\n\nThe paper is for people working on relative Langlands, automorphic periods, or the BZSV numerical conjecture. The math is self-contained enough that a serious referee can check it line-by-line. I would send it to peer review and would cite the period formula when I next need a non-cuspidal Bump–Friedberg-type identity.","headline":"Solid analytic extension of Bump–Friedberg periods past cuspidals, with an explicit multi-term L-value formula for a new SL_{n+1}\times GL_n period on Eisenstein series that matches the BZSV fixed-point prediction under Langlands.","tokens_in":47089,"tokens_out":604,"would_cite":true,"duration_ms":7759,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["11F70","22E55","11F66"],"pacs":[],"model":"grok-4.5","headline":"A Bump–Friedberg period on GL_{2n+1} extends continuously and evaluates on regular Eisenstein series as a finite sum of L-values matching fixed-point predictions.","keywords":["Bump–Friedberg periods","automorphic forms of uniform moderate growth","Whittaker zeta integrals","Eisenstein series","relative Langlands duality","global numerical conjecture","L-functions"],"falsifier":"Compute the period of a low-rank Δ₁*-regular Eisenstein series (e.g., n=1) both by the explicit sum of L-values and by direct integration or numerical approximation of the automorphic form; any mismatch of poles, residues or numerical values would refute the formula.","tokens_in":47075,"feed_emoji":"∑","tokens_out":840,"duration_ms":8669,"temperature":0.7,"pith_summary":"Classical Bump–Friedberg periods are defined only for cuspidal forms. This paper shows how to push three related periods past the cuspidal spectrum: a twisted period on GL_{2n}, an augmented period on GL_1 × GL_{2n}, and a new period on GL_{2n+1} integrating over SL_{n+1} × GL_n. Under explicit regularity conditions on the cuspidal datum, each period extends continuously from Schwartz functions to functions of uniform moderate growth, and the extension is realized by an entire Whittaker-type zeta integral. For the new period, the value on a regular Eisenstein series is computed explicitly: after normalization it becomes a finite sum, indexed by the GL_1 blocks of the inducing data, of products of three special L-values times local zeta integrals. Under the global Langlands correspondence those summands line up exactly with the fixed points of the L-parameter on a proposed dual variety and with the tangent-space L-factors predicted by the global numerical conjecture of relative Langlands duality. The result therefore supplies a concrete instance in which a non-cuspidal period produces a multi-term spectral expression rather than a single L-value.","feed_headline":"Non-cuspidal Bump–Friedberg periods sum L-values","feed_subtitle":"On GL_{2n+1} the extended period of a regular Eisenstein series equals a fixed-point sum of three L-factors.","key_machinery":"The entire Whittaker-type zeta integral Z(f, Φ, λ, s₂) (and its twisted/augmented analogues) obtained by successive Fourier expansions under regularity conditions that kill all non-principal orbits; its value at (0,0) supplies the continuous extension of the period, and for Eisenstein series the Langlands constant-term formula reduces it to a sum of twisted Bump–Friedberg integrals.","core_discovery":"Under a Δ₁*-regularity condition, the SL_{n+1} × GL_n period on GL_{2n+1} extends continuously to functions of uniform moderate growth via an entire zeta integral, and its value on a corresponding Eisenstein series equals (after normalization) a finite sum over the GL_1 blocks of products L^S(1,η_i^{-1} ⊗ Π_i) L^S(1,η_i ⊗ Π_i^∨) L^S(1/2, Π_i, η_i^{-1} ⊗ ∧²) times normalized local zeta integrals; under global Langlands these terms match the fixed-point contributions of a candidate dual variety.","pith_inferences":[],"forward_implications":[],"fun_headline_variants":["Non-cuspidal Bump–Friedberg periods equal fixed-point L-sums","GL_{2n+1} SL_{n+1}×GL_n periods yield triple L-factor products","Eisenstein Bump–Friedberg values match dual fixed-point L-factors","Regular Eisenstein periods sum L^S products over GL_1 blocks","Beyond-cuspidal periods extend via entire Whittaker zeta integrals"],"cache_read_input_tokens":32896,"weakest_assumption_plain":"The match with fixed points and tangent-space L-factors of the dual variety rests on the unproved global Langlands correspondence and on the unproved claim that the proposed dual variety is correct.","fun_headline_variants_meta":{"raw":{"variants":["Non-cuspidal Bump–Friedberg periods equal fixed-point L-sums","GL_{2n+1} SL_{n+1}×GL_n periods yield triple L-factor products","Eisenstein Bump–Friedberg values match dual fixed-point L-factors","Regular Eisenstein periods sum L^S products over GL_1 blocks","Beyond-cuspidal periods extend via entire Whittaker zeta integrals"]},"model":"grok-4.5","effort":"low","cost_usd":0.004992,"raw_usage":{"total_tokens":1460,"prompt_tokens":846,"num_sources_used":0,"completion_tokens":113,"cost_in_usd_ticks":49920000,"prompt_tokens_details":{"text_tokens":846,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":501,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":846,"tokens_out":113,"duration_ms":4139,"temperature":1.0,"reasoning_tokens":501,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-14T12:52:29.433853+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Compute the period of a low-rank Δ₁*-regular Eisenstein series (e.g., n=1) both by the explicit sum of L-values and by direct integration or numerical approximation of the automorphic form; any mismatch of poles, residues or numerical values would refute the formula.","supporting_citations":[],"review_version":1}