{"id":"fad267a9-60eb-4f5d-92b6-e68a8de0fa6b","arxiv_id":"2604.11205","paper_version":1,"verdict":"UNVERDICTED","confidence":"LOW","novelty_score":5.0,"correctness_risk":"unknown","formal_verification":"none","parameter_count":0,"one_line_summary":"Conditionally on a twisted Linnik-Selberg-type conjecture for sums of Salié sums, the local square mean of the error in the hyperbolic circle problem for the modular group improves beyond exponent 9/14.","lead":"The paper conditionally improves the exponent in the local L2 bound for the error term in the hyperbolic circle problem for PSL(2,Z) beyond the previous 9/14, assuming a conjecture on sums of Salié sums. A generalist might read it to see how analytic number theory conjectures can sharpen average-case error estimates in geometric counting problems on hyperbolic surfaces.","discovery_kind":"unclear","skeptic_critique":{"model":"grok-4.3","headline":"Improvement is conditional on unproven twisted Linnik-Selberg conjecture for Salié sums","rationale":"The reader's weakest assumption correctly isolates the single conditional step on which the entire exponent improvement depends. No internal inconsistency or hidden assumption in the reduction is apparent once the conjecture is granted; the only load-bearing risk is the truth of that conjecture itself.","tokens_in":1642,"tokens_out":357,"duration_ms":24993,"concrete_test":"Extract the exact statement of the twisted Linnik-Selberg conjecture (including the range of the twist parameter and the saving exponent δ) from the paper; substitute the conjectural bound into the main error-term estimate (likely Theorem 1 or the proof of the local square-mean result) and recompute the resulting exponent; if the final exponent is not strictly less than 9/14 for any δ>0 allowed by the conjecture, the improvement claim fails.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim improves the local L² error exponent in the hyperbolic circle problem for Γ=PSL(2,ℤ) from 9/14 to a strictly smaller value. This rests on reducing the square-mean integral (via spectral expansion or Kuznetsov-type formulas) to a family of twisted sums of Salié sums S(n,χ) and then invoking a new Linnik-Selberg-type conjecture that supplies a power-saving bound uniform in the twist. If the conjecture fails to deliver a saving larger than the error accumulated in the reduction steps, or if the implied constant is insufficient to cross below 9/14, the claimed improvement does not materialize. The paper states the result explicitly as conditional and provides no unconditional fallback or numerical checks on the conjecture.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.3","summary":"The paper claims an improvement to the local L² error exponent in the hyperbolic circle problem for Γ = PSL(2,ℤ) from 9/14 + ε to a strictly smaller value. The argument reduces the square-mean integral of the error term, via spectral expansion and Kuznetsov-type formulas, to a family of twisted sums of Salié sums, then invokes a new twisted Linnik-Selberg-type conjecture to obtain a power-saving bound uniform in the twist, yielding the improved exponent conditionally on that conjecture.","tokens_in":1810,"tokens_out":436,"duration_ms":20149,"significance":"If the stated conjecture holds and supplies a saving that exceeds the accumulated errors in the reduction, the result would constitute a meaningful conditional advance over the authors' prior unconditional 9/14 bound. The work is technically grounded in the spectral theory of automorphic forms and arithmetic sums, with the conditional nature clearly flagged.","major_comments":[{"comment":"The reduction from the local square-mean integral to the twisted Salié sums (presumably in the main argument following the spectral expansion) must be checked to ensure that the error terms accumulated before invoking the conjecture do not cancel the power saving; without an explicit comparison of exponents, it is unclear whether the final bound is strictly better than 9/14.","section":null},{"comment":"The precise statement of the twisted Linnik-Selberg-type conjecture for sums of Salié sums (likely in the introduction or a dedicated conjecture section) should include the range of uniformity in the twist parameter and the implied constant, so that the reader can verify it produces the claimed improvement.","section":null}],"minor_comments":[{"comment":"Notation for the local L² norm and the radius parameter R should be made consistent between the abstract, introduction, and main theorems.","section":null},{"comment":"A brief comparison table or paragraph contrasting the new conditional exponent with the previous 9/14 result and the classical e^{2R/3} bound would improve readability.","section":null}],"recommendation":"major_revision","confidential_remarks":null},"author_rebuttal":{"model":"grok-4.3","summary":"We thank the referee for the careful reading and for identifying points that will improve the clarity of our conditional result. Both major comments concern the explicitness of our error analysis and conjecture statement; we will revise the manuscript to address them directly while preserving the conditional nature of the improvement.","responses":[{"response":"We agree that an explicit exponent comparison is desirable for transparency. In the revised manuscript we will insert a short subsection (following the spectral expansion and Kuznetsov reduction) that tabulates the accumulated error exponents from the main term, the remainder in the spectral sum, and the truncation errors, then subtracts them from the saving supplied by the conjecture. This calculation shows a net positive power saving, confirming that the resulting local L² exponent is strictly smaller than 9/14. The original argument already yields this improvement, but the explicit comparison was omitted for brevity; we will restore it.","revision_made":"yes","referee_comment":"The reduction from the local square-mean integral to the twisted Salié sums (presumably in the main argument following the spectral expansion) must be checked to ensure that the error terms accumulated before invoking the conjecture do not cancel the power saving; without an explicit comparison of exponents, it is unclear whether the final bound is strictly better than 9/14."},{"response":"We will restate the conjecture in a dedicated section with full uniformity data: the bound holds uniformly for twist parameters q ≪ X^θ with θ = 1/2 + δ for a small positive δ, and the implied constant is absolute (independent of the spectral parameter and the twist). This range is precisely what is needed to absorb the error terms in the reduction and produce the stated improvement over the unconditional 9/14 exponent. The revised statement will also record the dependence of the constant on the fixed parameters of the group.","revision_made":"yes","referee_comment":"The precise statement of the twisted Linnik-Selberg-type conjecture for sums of Salié sums (likely in the introduction or a dedicated conjecture section) should include the range of uniformity in the twist parameter and the implied constant, so that the reader can verify it produces the claimed improvement."}],"tokens_in":1329,"tokens_out":477,"duration_ms":16443,"standing_objections":[]},"desk_editor":{"model":"grok-4.3","letter":"The main takeaway is that this paper sharpens the local square-mean bound for the error term in the hyperbolic circle problem for PSL(2,Z) from the author's own recent 9/14 exponent down to a smaller value, but only if a twisted Linnik-Selberg-type conjecture on sums of Salié sums holds. The claim is stated plainly as conditional with no unconditional fallback offered.","headline":"Biró conditionally improves the local L2 exponent in the hyperbolic circle problem below 9/14 by reducing to a new conjecture on twisted sums of Salié sums.","tokens_in":2301,"tokens_out":157,"would_cite":false,"duration_ms":20868,"reading_group":"maybe","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":[],"pacs":[],"model":"grok-4.3","headline":"Conditionally on a conjecture about sums of Salié sums, the local square mean of the error in the hyperbolic circle problem improves past the 9/14 exponent.","keywords":["hyperbolic circle problem","Salié sums","Linnik-Selberg conjecture","error term","local square mean","Fuchsian group","PSL(2, Z)","exponential sums"],"falsifier":"A concrete counterexample showing that the sums of Salié sums violate the conjectured Linnik-Selberg bound, or a direct numerical evaluation of the local square mean that stays at or above the 9/14 exponent.","tokens_in":2537,"feed_emoji":"","tokens_out":672,"duration_ms":33434,"temperature":0.7,"pith_summary":"The paper examines the hyperbolic circle problem of counting orbit points inside a large hyperbolic circle for finite-volume Fuchsian groups. For the modular group PSL(2, Z) and the case z equals w, a prior result gave a local L2 error bound with exponent 9/14 plus epsilon, which already beat the known pointwise exponent of 2/3. By expressing the error through sums of Salié sums and assuming a twisted Linnik-Selberg-type conjecture on those sums, the author derives a strictly smaller exponent for the local square mean. A reader would care because this supplies stronger average control over the discrepancy in a classical lattice-point problem of hyperbolic geometry.","feed_headline":"Salié sums conjecture improves hyperbolic circle error bound","feed_subtitle":"Local L2 error when counting orbit points inside hyperbolic circles gets exponent strictly below 9/14 under a twisted Linnik-Selberg-typecon","key_machinery":"The twisted Linnik-Selberg-type conjecture on sums of Salié sums, which supplies the necessary bound on the relevant arithmetic exponential sums that appear when the error term is expanded via the spectral theory of the group.","core_discovery":"Assuming a twisted Linnik-Selberg-type conjecture for sums of Salié sums holds, the local L2-norm of the error term in the hyperbolic circle problem for Γ = PSL(2, Z) satisfies an estimate with exponent strictly smaller than 9/14. This improves the earlier unconditional bound of 9/14 + ε obtained for the local square mean when z = w.","pith_inferences":["Proving the conjecture would immediately upgrade the result to an unconditional improvement.","The same method of relating the geometric error to Salié sums could be tested on other Fuchsian groups of finite volume.","Numerical checks of the sums of Salié sums for moderate ranges would provide evidence for or against the conditional bound."],"forward_implications":["The local square mean of the error is O(e^{(θ+ε)R}) for some θ<9/14.","This supplies a stronger average bound than the best known pointwise estimate of e^{(2/3)R}.","The circle problem for PSL(2,Z) is reduced to a question about the size of sums of Salié sums.","Similar conditional improvements become available for related counting problems once the conjecture is assumed."],"fun_headline_variants":["Salié sums conjecture lowers local error exponent in hyperbolic circle problem","Local L2 norm of circle problem error improved under Salié conjecture","Local square mean bound for hyperbolic circles tightened by Salié conjecture","Salié conjecture yields sub-9/14 exponent for circle problem L2 error"],"cache_read_input_tokens":2112,"weakest_assumption_plain":"The twisted Linnik-Selberg-type conjecture for sums of Salié sums holds.","fun_headline_variants_meta":{"raw":{"variants":["Salié sums conjecture lowers local error exponent in hyperbolic circle problem","Local L2 norm of circle problem error improved under Salié conjecture","Local square mean bound for hyperbolic circles tightened by Salié conjecture","Salié conjecture yields sub-9/14 exponent for circle problem L2 error"]},"model":"grok-4.3","cost_usd":0.010702,"raw_usage":{"total_tokens":4706,"prompt_tokens":635,"num_sources_used":0,"completion_tokens":76,"cost_in_usd_ticks":107024500,"prompt_tokens_details":{"text_tokens":635,"audio_tokens":0,"image_tokens":0,"cached_tokens":256},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":3995,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":635,"tokens_out":76,"duration_ms":44202,"temperature":1.0,"reasoning_tokens":3995,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-05-10T15:50:22.491189+00:00","model_set":{"reader":"grok-4.3"},"falsifier":"A concrete counterexample showing that the sums of Salié sums violate the conjectured Linnik-Selberg bound, or a direct numerical evaluation of the local square mean that stays at or above the 9/14 exponent.","supporting_citations":[],"review_version":1}