{"id":"0af2243c-4883-4d3e-8e0f-ebfdd8488fee","arxiv_id":"2607.03854","paper_version":1,"verdict":"ACCEPT","confidence":"HIGH","novelty_score":6.5,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"Ehrhart polynomials of partial permutohedra P(m,n) are magic positive for every n≥m-1 except (m,n)=(2,1); for n=1,2,3 the positivity fails for all sufficiently large m, with only two exceptional positive cases when n=3.","lead":"The paper proves that Ehrhart polynomials of partial permutohedra are magic-positive for all n at least m-1 except the single case (2,1), and classifies the first three smaller-n families. This largely settles an open problem of Ferroni–Higashitani on which of these polytopes enjoy the strong positivity property.","discovery_kind":"extension","skeptic_critique":{"model":"grok-4.5","headline":"No significant objection identified","rationale":"The reader correctly isolates Behrend’s formula as the sole external pillar and notes that every subsequent reduction is internal and elementary. After re-examining the chain—magic transform \to Bi,r \to differential reduction \to tree-function representation \to coefficient monotonicity of R and C \to induction—I find no soft spot that threatens the non-negativity statement for n≥m-1. The exceptional case is computed explicitly and matches the claimed negative coefficient. The unstable-range classification (Theorem 1.6) is independent of the main argument and is supported by explicit product formulae whose first few coefficients are negative for all but two small pairs. Consequently the ACCEPT verdict stands; no adjustment is warranted.","tokens_in":15809,"tokens_out":611,"duration_ms":5270,"concrete_test":"Independently expand the generating function in (3) for the boundary pairs (i,r) with i+r≤6 (including the exceptional (1,1)) and verify that Bi,r(i+r-1) equals the coefficient [ui]R(u)rC(u) obtained from the closed forms (7) and (10); agreement to machine precision confirms the Lagrange-inversion reduction and the sign claims of Proposition 3.6.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The central claim (Theorem 1.3) rests on Behrend’s coefficient-extraction formula (1) for n≥m-1, which the authors convert into the magic coefficients μm,n,i = (m!/r!) Bi,r(n) (Lemma 2.1). Non-negativity of Bi,r(n) for n≥i+r-1 is then reduced, via the differential relation ∂/∂n Bi,r = Bi-1,r (Lemma 2.2), to boundary non-negativity of Bi,r(i+r-1). The boundary values are rewritten via the rooted-tree function T(u) and Lagrange inversion (Lemma 3.2) as [ui] R(u)r C(u). Coefficient estimates for R (Lemmas 3.3–3.4) and C (Lemma 3.5) establish the required sign pattern except for the single pair (i,r)=(1,1), which is handled by direct computation and yields the unique exception (m,n)=(2,1). The induction in Proposition 4.1 closes the argument cleanly. No hidden assumption, circular step, or range violation appears; the only external input is the cited formula of Behrend, whose domain matches exactly the stable range claimed. Residual arithmetic risk in the n=3 tables for 7≤m≤17 does not affect Theorem 1.3.","agreement_with_reader":"agree"},"referee_report":{"model":"grok-4.5","summary":"The paper studies magic positivity of the Ehrhart polynomials of partial permutohedra P(m,n). For n ≥ m-1 it proves that E_{m,n}(t) is magic positive except precisely for (m,n)=(2,1), by converting Behrend’s coefficient-extraction formula into magic coefficients μ_{m,n,i}=(m!/r!)B_{i,r}(n), reducing non-negativity via ∂/∂n B_{i,r}=B_{i-1,r} to boundary values B_{i,r}(i+r-1), and establishing those signs by expressing the boundary values through the rooted-tree function and Lagrange inversion as [u^i]R(u)^r C(u), with explicit coefficient estimates for R and C. As a corollary the parking-function polytope (integrally equivalent to P(m,m-1)) is magic positive for m≥3. For n<m-1 the paper classifies the cases n=1,2,3 completely: infinite families of counterexamples for n=1 (m≥3) and n=2 (m≥4), and for n=3 positivity only when m=5,6. This partially answers an open problem of Ferroni–Higashitani.","tokens_in":16121,"tokens_out":1114,"duration_ms":7797,"significance":"Magic positivity is a strong positivity property that implies both Ehrhart positivity and real-rootedness of the h*-polynomial (via Brändén). Partial permutohedra form a natural family of Y-generalized permutohedra for which the stable-range Ehrhart formula is known, so a complete magic-positivity classification in that range is a concrete advance. The parking-function corollary is of independent combinatorial interest. The argument is self-contained once Behrend’s formula is granted: every non-negativity claim is proved by explicit series comparison or induction, with no fitted parameters. The small-n analysis already exhibits both infinite counterexample families and exceptional positive cases, clarifying that the complementary range is mixed. The result therefore supplies a substantial partial resolution of the open problem posed in the Ferroni–Higashitani survey.","major_comments":[],"minor_comments":[{"comment":"In the abstract and introduction the phrase “integrally equivalent to P(m,m-1)” for the parking-function polytope is used without a one-line reference or definition of the equivalence; a short citation or parenthetical would help readers who know only the classical definition of P_m.","section":"Abstract / §1"},{"comment":"Lemma 3.5 computes the first few coefficients of C(u) by hand (c_2=c_3=0) and then gives a lower bound for k≥4; while correct, a brief remark that the same closed-form expression for [u^k]e^{-T} and the Lagrange formula for powers of T can be used to obtain an exact formula for all c_k would make the argument more uniform.","section":"§3.3, Lemma 3.5"},{"comment":"Table 1 lists the coefficient of y^4 for 7≤m≤17 as negative fractions; the fractions are correct but extremely large. Adding a short sentence that they were obtained by expanding the product formula of Proposition 5.7 (or by a computer-algebra script) would improve reproducibility.","section":"§5.3, Table 1"},{"comment":"A few typographical inconsistencies appear: “Brändén” is sometimes written without the umlaut, and the arXiv identifier of the concurrent Avila–Ferroni–Morales preprint is given as 2603.19194 (future-dated). Standardizing the orthography and confirming the identifier would be helpful.","section":"Throughout / References"},{"comment":"In the proof of Proposition 4.1 the three sub-cases for the induction step on B_{i-1,r} are exhaustive, but a one-line summary table of the exceptional pairs (i,r)=(1,0),(1,1),(2,0),(2,1) would make the case distinction easier to follow on a first reading.","section":"§4, Proposition 4.1"}],"recommendation":"accept","confidential_remarks":"The manuscript is clean, the logical skeleton is complete, and the only external input (Behrend’s formula) is correctly scoped. I see no load-bearing gap. The residual arithmetic verification for the n=3 table is routine and does not affect the main theorem. Suitable for acceptance after the minor presentation fixes listed above."},"author_rebuttal":null,"desk_editor":{"model":"grok-4.5","letter":"This paper settles the stable half of the open problem on magic positivity of partial permutohedra. Theorem 1.3 is the real payload: for every n≥m−1 the Ehrhart polynomial is magic positive except exactly (m,n)=(2,1). As a corollary the parking-function polytope is magic positive for m≥3, so its h*-polynomial is real-rooted. The n=1,2,3 analysis then supplies infinite counter-example families below the diagonal and two small positive exceptions (P(5,3), P(6,3)). That is a genuine partial answer to Ferroni–Higashitani Problem 4.23, going beyond the asymptotic results of Avila–Ferroni–Morales.\n\nWhat they do well is the reduction. Starting from Behrend’s coefficient formula (valid precisely when n≥m−1), they convert magic positivity into non-negativity of the series Bi,r(n). Differentiation in n lowers the first index, so everything reduces to the boundary n=i+r−1. They rewrite the boundary via the rooted-tree function and Lagrange inversion as [ui]R(u)rC(u), then prove the needed coefficient monotonicity for powers of R and the sign pattern C(u)=1−u+P(u). The induction that lifts the boundary to the whole half-plane is short and clean. The exceptional pair (i,r)=(1,1) is handled by direct computation and produces the single known counter-example. No free parameters, no circular definitions.\n\nThe only soft spots are minor and do not touch the main theorem. Everything rests on Behrend’s formula; if that identity ever needed re-checking the whole argument would need re-checking, but the domain matches exactly. For n=3 the tables of y4-coefficients for 7≤m≤17 are pure arithmetic; a transcription error could exist, yet the first-coefficient argument already kills m≥18 and the main theorem is unaffected. The paper does not claim a full classification for all n<m−1, and the authors say so.\n\nThis is for people who work on Ehrhart positivity, generalized permutohedra, or parking functions. The generating-function technique is reusable. I would send it to a serious referee without hesitation; the math is solid and the open-problem progress is real. Cite it if you need the parking-function statement or the stable-range result.","headline":"Clean partial solution of Ferroni–Higashitani 4.23: magic positivity for all n≥m−1 except (2,1), plus complete n=1,2,3 classification.","tokens_in":16764,"tokens_out":609,"would_cite":true,"duration_ms":5631,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05A15","52B20","52B05"],"pacs":[],"model":"grok-4.5","headline":"For n at least m-1 the Ehrhart polynomial of the partial permutohedron is magic positive except one case.","keywords":["lattice polytope","partial permutohedron","parking function polytope","Ehrhart polynomial","magic positivity","h*-polynomial","rooted-tree function"],"falsifier":"Compute the magic transform of the Ehrhart polynomial of P(m,n) for any pair with n greater than or equal to m-1 other than (2,1) and check whether every coefficient is non-negative; a single negative coefficient would refute the main theorem.","tokens_in":16695,"feed_emoji":"📐","tokens_out":939,"duration_ms":6868,"temperature":0.7,"pith_summary":"The paper settles most of an open classification question: which partial permutohedra have magic-positive Ehrhart polynomials. Magic positivity is a stronger form of Ehrhart positivity that also forces the associated h*-polynomial to be real-rooted. The authors prove that whenever the second parameter n is at least m-1, the Ehrhart polynomial is magic positive except for the single pair (m,n)=(2,1). In particular the parking-function polytope, integrally equivalent to P(m,m-1), is magic positive for every m greater than or equal to 3. Below the threshold n less than m-1 the picture is mixed: they exhibit infinite families of counterexamples for n=1 and n=2, while for n=3 only the two small polytopes P(5,3) and P(6,3) remain magic positive. The result therefore supplies a clean affirmative answer on the “stable” side of the parameter plane and already shows that the complementary side is not uniformly positive.","feed_headline":"Partial permutohedra are magic positive above the diagonal","feed_subtitle":"Except one case, their Ehrhart polynomials stay non-negative in the magic basis whenever n is at least m-1","key_machinery":"Behrend’s coefficient-extraction formula for the Ehrhart polynomial when n is at least m-1, rewritten as a generating function whose coefficients B_{i,r}(n) are shown non-negative by a differential reduction in n, evaluation of the boundary values via the rooted-tree function and Lagrange inversion, and coefficient-wise positivity estimates for two auxiliary series R(u) and C(u).","core_discovery":"For every pair of positive integers m,n with n greater than or equal to m-1 the Ehrhart polynomial of the partial permutohedron P(m,n) can be written with non-negative coefficients in the magic basis {t^i (t+1)^{m-i}}, with the sole exception of the two-dimensional polytope P(2,1). Equivalently, every magic coefficient is non-negative except that one case.","pith_inferences":["The same rooted-tree analysis may extend to other families of Y-generalized permutohedra once an analogous coefficient formula is available.","The two exceptional positive cases for n=3 suggest that a finite list of sporadic magic-positive polytopes may exist for each fixed n less than m-1.","Because magic positivity implies real-rootedness of h*, the result immediately yields new infinite families of real-rooted h*-polynomials coming from combinatorial polytopes."],"forward_implications":["The h*-polynomial of every partial permutohedron with n at least m-1 is real-rooted.","The parking-function polytope of length m is magic positive for all m greater than or equal to 3.","Partial permutohedra supply an infinite family of Y-generalized permutohedra that are magic positive for all admissible parameters except one explicit exception.","Below the line n = m-1 the property fails for infinitely many pairs, so any complete classification must treat the two regimes separately."],"fun_headline_variants":["Partial permutohedra are magic positive for n≥m-1 except P(2,1)","Ehrhart of P(m,n) magic positive when n≥m-1 save the case (2,1)","Magic positivity holds for partial permutohedra above the diagonal except one","All P(m,n) with n≥m-1 have magic-positive Ehrhart except P(2,1)","Parking-function polytopes have magic-positive Ehrhart for m≥3"],"cache_read_input_tokens":128,"weakest_assumption_plain":"The entire argument for the stable range rests on a single closed-form generating function for the Ehrhart polynomial that is known only when n is at least m-1.","fun_headline_variants_meta":{"raw":{"variants":["Partial permutohedra are magic positive for n≥m-1 except P(2,1)","Ehrhart of P(m,n) magic positive when n≥m-1 save the case (2,1)","Magic positivity holds for partial permutohedra above the diagonal except one","All P(m,n) with n≥m-1 have magic-positive Ehrhart except P(2,1)","Parking-function polytopes have magic-positive Ehrhart for m≥3"]},"model":"grok-4.5","effort":"low","cost_usd":0.006772,"raw_usage":{"total_tokens":1682,"prompt_tokens":782,"num_sources_used":0,"completion_tokens":125,"cost_in_usd_ticks":67720000,"prompt_tokens_details":{"text_tokens":782,"audio_tokens":0,"image_tokens":0,"cached_tokens":128},"completion_tokens_details":{"audio_tokens":0,"reasoning_tokens":775,"accepted_prediction_tokens":0,"rejected_prediction_tokens":0}},"tokens_in":782,"tokens_out":125,"duration_ms":5512,"temperature":1.0,"reasoning_tokens":775,"cache_read_input_tokens":128,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-07-11T23:29:11.633782+00:00","model_set":{"reader":"grok-4.5"},"falsifier":"Compute the magic transform of the Ehrhart polynomial of P(m,n) for any pair with n greater than or equal to m-1 other than (2,1) and check whether every coefficient is non-negative; a single negative coefficient would refute the main theorem.","supporting_citations":[],"review_version":1}