{"id":"8dd68236-cc8d-4d4f-8891-72ffac7633bf","arxiv_id":"2509.07266","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For unicritical algebraic correspondences z^{p/q}+c, this paper proves asymptotic self-similarity of the Julia set at Misiurewicz points and asymptotic similarity between the Multibrot and Julia sets under a transversality condition, proven algebraically for the (4,2) family.","lead":"This paper proves that the Julia sets and Multibrot sets of a family of multi-valued maps called algebraic correspondences are asymptotically self-similar at small scales near Misiurewicz parameters, extending a classical result of Tan Lei for the quadratic family. The main theorem is conditional on a transversality hypothesis that the author proves, using 2-adic valuations, for the semigroup generated by z^2+c and -z^2+c.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Theorem 5.2 hinges on the unproven equality K_a=J_a / density of repelling cycles in K_a, cited only to a literal 'Theorem??' and the unpublished [19, Theorem D]; without it, Lemma 5.6's dense set X'(a) cannot be constructed.","rationale":"The reader's weakest_assumption identifies the same gap: Lemma 5.6 relies on [19, Theorem D] for the density of repelling periodic points in K_a. This is indeed the single most load-bearing concern. The transversality condition is explicitly stated as a hypothesis in Theorem B and proved for (4,2); even if the algebraic proof in §6 had gaps, it would affect only the unconditional corollary. The K_a=J_a/density issue, by contrast, is needed to construct the dense holomorphic-motion set X'(a) in every application of Tan Lei's theorem, and it is not proved in the manuscript. The paper contains a literal 'Theorem??' in Lemma 5.6, which is a clear sign of missing support. The internal Remark 5.1 further concedes the difficulty. Thus the central claim is conditionally established at best: it is sound if [19, Theorem D] is correct and the uniformity hypotheses in Theorem 5.1 can be verified, but neither is demonstrated here. I therefore agree with the reader's CONDITIONAL verdict and recommend no change.","tokens_in":22132,"tokens_out":12895,"duration_ms":146805,"concrete_test":"Obtain the proof of Theorem D in the author's preprint [19]. Verify (i) that it actually proves K_a=J_a for all Misiurewicz parameters of (1.1) (not just for special cases), and (ii) that it does not invoke the conclusion of Theorem 5.2 or Lemma 5.6 (no circularity). If the proof is missing or circular, attempt an independent proof of K_a=J_a for the semigroup family ⟨z^2+c,-z^2+c⟩ by showing the filled Julia set has empty interior and every point of K_a is a limit of repelling cycles. If no such proof can be given, Lemma 5.6 is unsubstantiated; Theorem 5.2 and Corollary 5.1 should be reformulated with K_a=J_a as an explicit hypothesis or additional evidence.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The proof of Lemma 5.6 (Section 5) is the load-bearing step for Theorem 5.2. It defines X'(a)=φ_a∘g_a(R_a∩Ω_a^∘)∪∂D_r and asserts that its closure is X(a). This requires repelling periodic points R_a to be dense in K_a. By Definition 2.1, J_a is the closure of R_a, so this density is exactly the assertion K_a=J_a. The manuscript cites a literal 'Theorem??' and the unpublished preprint [19, Theorem D] for this equality, and the Remark after Theorem C concedes that for correspondences 'the Julia and Fatou sets are no longer completely invariant, and many classical tools fail to apply.' If K_a contains any open Fatou component, then R_a is not dense in K_a, the set X'(a) is not dense in X(a), and hypothesis (ii) of Tan Lei's Theorem 5.1 fails, so the conclusion of Theorem 5.2 does not follow. Remark 5.1 itself warns that without further analysis of the structure of K_a, the density of R_a∩Ω_a^∘ in J_a∩Ω_a cannot be shown. Thus Theorem 5.2 — and the unconditional Corollary 5.1 for (4,2) — rest on an external, non-self-contained result; this is the weakest point of the argument.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the unicritical family of algebraic correspondences f_c(z)= q√(z^p+c). For a Misiurewicz parameter a (unique bounded critical orbit, strictly pre-periodic), it proves (Theorem 4.1) that the filled Julia set K_c is asymptotically self-similar about the points of the associated pre-periodic orbit. The central result (Theorem 5.2) asserts that, under a transversality condition w'(a)≠0, the Multibrot set M_{p,q} and the Julia set K_a are asymptotically similar about a, with common scale λ(a) and limit models coinciding up to a constant. For (p,q)=(4,2), Theorem 3.1 proves transversality by a 2-adic valuation argument, giving an unconditional Corollary 5.1.","tokens_in":22459,"tokens_out":15598,"duration_ms":166151,"significance":"If correct, this is a substantial extension of Tan Lei's theorem to algebraic correspondences. The reduction of similarity to transversality and the algebraic proof of transversality for the (4,2)-semigroup are original and potentially useful. The paper is honest about its dependence on the unpublished preprint [19] and includes Remark 5.1 flagging the exact difficulty. However, the main theorem is only as strong as the unproved density of repelling periodic points in K_a, so the significance is contingent.","major_comments":[{"comment":"The proof invokes 'By Theorem??, this set is dense in J_a=K_a'. This theorem is not identified; the only external source offered is the unpublished preprint [19, Theorem D]. The density of repelling periodic points in K_a is exactly what makes X'(a) dense in X(a) and is therefore load-bearing for hypothesis (ii) of Theorem 5.1. Without it, Theorem 5.2 and Corollary 5.1 do not follow. Remark 5.1 explicitly concedes that this density cannot currently be shown. This must be repaired by a proof or a precise, verifiable reference.","section":"§5, Lemma 5.6"},{"comment":"In case (b) (and similarly (d)) the proof asserts 'v(ζ−η) equals min{v(ζ),v(η)}'. This is not a valuation identity: equality holds only when the valuations differ; otherwise cancellation can make the left side larger. The congruence (6.3), and hence the proof of Theorem 3.1, depends on this step. The conclusion may be true, but the argument as written is incomplete and needs a correct valuation estimate.","section":"§6, Lemma 6.1"},{"comment":"The displayed computation of F_c(a,z_ℓ) is invalid: adding the term (F(a,z_ℓ)−F(c,z_ℓ(c)))/(c−a) changes the limit by a generally nonzero quantity, so the chain of equalities does not hold. The intended formula F_c(a,z_ℓ)=−z'_ℓ(a) follows instead by differentiating the identity F(c,z_ℓ(c))=0 with respect to c. As printed, the proof of u'(a)=w'(a)/(λ(a)−1) is not correct. This is a local error, but it occurs in the central theorem.","section":"§5, Eq. (5.8)"}],"minor_comments":[{"comment":"The placeholder 'Theorem??' should be replaced by an actual citation; please also clarify the status of [19, Theorem D] for the equality K_a=J_a.","section":"§5, Lemma 5.6"},{"comment":"The equation contains a duplicated expression '=φ_c(V_{c,r}∩K_c)∩D_{λ(c)r}'; likely a typo.","section":"§4, Eq. (4.4)"},{"comment":"In the second case, 'F'_{ℓ+n-1}(a)−F'_ℓ(a)' should probably be 'F'_{ℓ+n}(a)−F'_ℓ(a)' to match the first case and the definition of w(c).","section":"Proof of Theorem 3.1"},{"comment":"The arrow in 'ˇz_1 = a fc → ˇz_2' should be 'f_a' for clarity.","section":"§6.2"},{"comment":"The text mentions magnifications of 10^3 and 10^5, but the caption is not fully explicit about the scale factor for each panel; please make it consistent.","section":"Figure 1 caption"}],"recommendation":"major_revision","confidential_remarks":"The manuscript's main theorem depends on a load-bearing fact that is only cited to an unpublished preprint and a literal 'Theorem??' placeholder. Even if the author's preprint [19] is available, a journal referee should verify that the density statement is actually proved there and is not just taken for granted. The valuation gap in §6 and the flawed computation in Eq. (5.8) are repairable, but together they make the current version unsuitable for acceptance as is."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Plainly: this is a real extension of Tan Lei's theorem to unicritical algebraic correspondences, with one serious gap in the part that connects the Multibrot set to the Julia set. The self-similarity of the Julia set (Theorem A / Theorem 4.1) is new and looks sound; the algebraic proof of transversality for (p,q)=(4,2) is detailed and self-contained, and that alone is a worthwhile result.\n\nThe paper does a good job of setting up filled Julia sets and the Multibrot set for the family z^{p/q}+c, and of reducing the Multibrot–Julia similarity to a transversality hypothesis, exactly in the spirit of Tan Lei. Theorem 4.1 uses holomorphic motion of the repelling cycle and the Kœnigs linearization to get asymptotic self-similarity of K_c about each point of the pre-periodic orbit; that part does not depend on the questionable density lemma. The valuation argument in Section 6 for the semigroup <z^2+c, -z^2+c> appears to prove transversality, and I did not see a circular step there.\n\nThe soft spot is Lemma 5.6, and it is load-bearing for Theorem 5.2. To apply Tan Lei's Theorem 5.1 you need a dense set of points in X(a) that move holomorphically. The paper builds X'(a) from repelling periodic points R_a, which requires R_a dense in K_a and K_a=J_a. That fact is cited to a literal 'Theorem??' and to the author's unpublished preprint [19, Theorem D]. Remark 5.1 openly concedes that without more structure on K_a, the density of R_a∩Ω_a^∘ in J_a∩Ω_a cannot be shown. If that density fails, the dense set X'(a) cannot be constructed, and Theorem 5.2 does not follow. So the headline claim about the Multibrot set is conditional on an external result that is not established here. The paper also simplifies Tan Lei's Proposition 4.1 and asserts uniformity of the self-similarity radius; that looks fixable but is not fully written out.\n\nThis is still a serious paper. Theorem A and the transversality proof deserve referee time, and the conditional Theorem B is a clean reduction. The right outcome is major revision, not rejection: the author should either prove the density fact, or state Theorem B as conditional on it, and fix the missing reference. Specialists in holomorphic correspondences will want to read this; the valuation proof in Section 6 is the most solid part.","headline":"A genuine extension of Tan Lei's similarity theorem to algebraic correspondences, with a solid transversality proof, but the Multibrot–Julia similarity hinges on an unproven density claim that the paper itself flags.","tokens_in":22960,"tokens_out":3281,"would_cite":true,"duration_ms":35876,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["37F05","37F10","37F32"],"pacs":[],"model":"deepseek-v4-flash","headline":"At Misiurewicz points of the correspondence family z ↦ (z^p+c)^(1/q), the Multibrot set and the filled Julia set are asymptotically similar, sharing a common scaling factor and limit model up to a nonzero constant; for (p,q)=(4,2) the resul","keywords":["Misiurewicz points","Multibrot set","Julia set","algebraic correspondences","asymptotic self-similarity","transversality condition","holomorphic motions","unicritical family"],"falsifier":"Take any Misiurewicz parameter a for the family (1.1) with (p,q) not (4,2), and compute w'(a) = d/dc [h_c(g_c(c))−g_c(c)] at c=a by following the branches explicitly. A vanishing derivative for one such parameter would disprove the transversality conjecture and remove the hypothesis of Theorem B for that point. Alternatively, numerically magnify M_{p,q} and K_a about a by powers of λ(a) and measure the Hausdorff distance between the scaled sets; if that distance does not tend to zero, the asymptotic similarity is false.","tokens_in":22007,"feed_emoji":"🌀","tokens_out":11459,"duration_ms":120210,"temperature":0.7,"pith_summary":"This paper studies the parameter space of a multi-valued iterated map family, f_c(z)= the q-th root of (z^p + c), with p > q > 0. The key claim is that at a Misiurewicz point—a parameter where the critical point 0 has exactly one bounded orbit, eventually periodic—the Multibrot set (parameters for which 0 has a bounded orbit) and the filled Julia set (points with at least one bounded orbit) are asymptotically similar: under magnification by powers of a single complex scaling factor, both sets converge to the same limit shape up to a nonzero constant. This extends the classical 1990 result for quadratic polynomials by replacing polynomials with algebraic correspondences, which are multi-valued maps. The proof works whenever a natural transversality condition holds; the paper proves that condition algebraically for the family (p,q)=(4,2), and gives experimental evidence for other exponents.","feed_headline":"Multibrot and Julia sets zoom in the same way at Misiurewicz points","feed_subtitle":"Extends the classical Mandelbrot–Julia similarity to multi-valued maps; transversality proven for the (4,2) family.","key_machinery":"The key objects are the (p:q) algebraic correspondences f_c(z)=(z^p+c)^(1/q), which send a point to q images and admit p preimages, and their Misiurewicz parameters: parameter values where the critical point 0 has exactly one bounded orbit and that orbit is strictly preperiodic, landing on a repelling cycle. The argument's engine is a holomorphic motion of that cycle: Lemma 3.1 shows the cycle, and the univalent branches along it, persist holomorphically as c varies, so the preperiodic orbit continues as a unique bounded orbit ξ(c). The transversality condition—the requirement that w(c)=h_c(g_c(c))−g_c(c), with g_c the branch composition to the cycle and h_c the cycle composition, satisfy w'","core_discovery":"The paper's central claim is Theorem B (Theorem 5.2 in the text): let a be a Misiurewicz parameter for the correspondence family (1.1), and suppose the transversality condition holds at a. Then both the Multibrot set M_{p,q} and the filled Julia set K_a are asymptotically self-similar about a with the same scaling factor λ(a), and their limit models coincide up to multiplication by a nonzero complex constant. Here λ(a) is the multiplier of the repelling cycle on which the critical orbit lands. When (p,q)=(4,2), the transversality condition is proved in full (Theorem 3.1), so the similarity becomes an unconditional theorem (Corollary 5.1). The paper also proves an independent statement (Theor","pith_inferences":["If the density claim cited from the preprint weakens, the similarity theorem may still hold for the filled Julia set but would need a different dense set; Remark 5.1 already shows the natural replacement Y(c) does not work, so this is a concrete open problem.","The 2-adic valuation proof of transversality for (p,q)=(4,2) is algebraic and may transfer to other even-exponent families where the sign-cancellation structure is analogous; testing p=6,q=2 at a Misiurewicz point would give a cheap check.","A quantitative computational experiment—recording Hausdorff distances between λ(a)^n-scaled M_{p,q} and K_a at, say, p/q=3/2—would estimate whether the conjectured transversality holds and with what convergence rate; the paper's figures are suggestive but not numerical.","The theorem's 'up to a nonzero complex constant' allows the parameter-space and dynamical-plane models to differ by a rotation-dilation; identifying that constant µ_a explicitly for small periods may reveal a scaling-universality statement across the parameter and dynamical planes."],"forward_implications":["At every Misiurewicz point satisfying transversality, M_{p,q} and K_a have the same asymptotic scaling: magnifying either about a by powers of λ(a) produces the same limit set up to a nonzero complex constant.","For (p,q)=(4,2), the transversality check is algebraic, so the similarity theorem is unconditional for that family: K_a and M_{4,2} are asymptotically similar at every Misiurewicz point.","For the Julia-side result (Theorem 4.1), no transversality is needed: the filled Julia set K_c is asymptotically λ(c)-self-similar about every point of the preperiodic orbit and of the repelling cycle.","The paper reduces the full similarity theorem to a single condition: prove transversality at a Misiurewicz point and the whole similarity result follows by the same argument.","If the transversality conjecture holds for all integer exponents, the similarity theorem would hold for every rational exponent p/q > 1 in the family."],"supporting_citations":[{"why":"The classical similarity theorem that Theorem 5.1 adapts; its hypotheses (self-similar slices plus dense holomorphic motions) structure the proof of Theorem B.","marker":"[13]"},{"why":"Unpublished preprint cited for Theorem D, which supplies K_a=J_a and the density of repelling periodic points needed in Lemma 5.6.","marker":"[19]"},{"why":"Provides the escaping radius, forward invariance, and the intersection formula K_c = ∩ f_c^{-n}(B_R) used to define the filled Julia set and to control perturbations.","marker":"[17]"},{"why":"Used for compactness of K_c and for holomorphic-motion tools for this family of correspondences, supporting Lemma 3.1.","marker":"[20]"},{"why":"Origin of the transversality condition for polynomial families, whose analogue is the key hypothesis of Theorem B and is proved algebraically for (4,2).","marker":"[8,9]"},{"why":"Supplies the linearization theorem for repelling fixed points and the holomorphic dependence of the linearizing maps on parameters, used to define φ_c and the limit models.","marker":"[16]"}],"fun_headline_variants":["Multibrot and Julia sets share scaling factor at Misiurewicz","Similarity proven for Multibrot and Julia at Misiurewicz points","Mandelbrot–Julia similarity extends to multivalued correspondences","Transversality gives unconditional Multibrot-Julia similarity"],"cache_read_input_tokens":2688,"weakest_assumption_plain":"The argument rests on a density property of the Julia set—that periodic points which repel nearby orbits are dense—which the paper cites from an unpublished preprint; if that property fails, the dense holomorphic motions required by the similarity theorem cannot be built.","fun_headline_variants_meta":{"raw":{"variants":["Multibrot and Julia sets share scaling factor at Misiurewicz","Similarity proven for Multibrot and Julia at Misiurewicz points","Mandelbrot–Julia similarity extends to multivalued correspondences","Transversality gives unconditional Multibrot-Julia similarity"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000206,"raw_usage":{"total_tokens":1240,"prompt_tokens":758,"completion_tokens":482,"prompt_tokens_details":{"cached_tokens":256},"prompt_cache_hit_tokens":256,"prompt_cache_miss_tokens":502,"completion_tokens_details":{"reasoning_tokens":403}},"tokens_in":502,"tokens_out":482,"duration_ms":5501,"temperature":1.0,"reasoning_tokens":403,"cache_read_input_tokens":256,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-04T22:34:46.511669+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take any Misiurewicz parameter a for the family (1.1) with (p,q) not (4,2), and compute w'(a) = d/dc [h_c(g_c(c))−g_c(c)] at c=a by following the branches explicitly. A vanishing derivative for one such parameter would disprove the transversality conjecture and remove the hypothesis of Theorem B for that point. Alternatively, numerically magnify M_{p,q} and K_a about a by powers of λ(a) and measure the Hausdorff distance between the scaled sets; if that distance does not tend to zero, the asymptotic similarity is false.","supporting_citations":[{"cited_title":"3, 587–617","cited_arxiv_id":null,"evidence_quote":"The classical similarity theorem that Theorem 5.1 adapts; its hypotheses (self-similar slices plus dense holomorphic motions) structure the proof of Theorem B."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Unpublished preprint cited for Theorem D, which supplies K_a=J_a and the density of repelling periodic points needed in Lemma 5.6."},{"cited_title":"8, 2661–2692","cited_arxiv_id":null,"evidence_quote":"Provides the escaping radius, forward invariance, and the intersection formula K_c = ∩ f_c^{-n}(B_R) used to define the filled Julia set and to control perturbations."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Used for compactness of K_c and for holomorphic-motion tools for this family of correspondences, supporting Lemma 3.1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the linearization theorem for repelling fixed points and the holomorphic dependence of the linearizing maps on parameters, used to define φ_c and the limit models."}],"review_version":1}