{"id":"de70e062-04ed-44b8-80c7-f19e5c906099","arxiv_id":"2611.16354","paper_version":1,"verdict":"ACCEPT","confidence":"MODERATE","novelty_score":8.0,"correctness_risk":"low","formal_verification":"none","parameter_count":0,"one_line_summary":"The complete set of postcritically finite parameters in CM fields for the unicritical family is determined for every degree d.","lead":"This paper lists every constant c for which the polynomial z^d+c has a critical point with a finite orbit and where c lies in a CM field. It also gives the same list over the maximal abelian extension of the rationals, completing a classification begun for smaller fields.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek/deepseek-v4-flash","headline":"No significant objection identified","rationale":"The reader's verdict is ACCEPT with moderate confidence, correctness risk low. The reader's weakest_assumption is the reliance on Goksel's irreducibility theorem for G_{m,2}. This is indeed a load-bearing step in Lemma 5.3, but it is a published result from a peer-reviewed source. The paper does not rely on any unpublished or unverified claim. The proof is otherwise self-contained and rigorous. The cyclotomic average calculations are explicit and verifiable. The escape bound and CM integer/unit theorems are proved in the paper. The classification for d≥3 is direct. The quadratic case is handled with a clear reduction to four families, each excluded by computation. No internal inconsistency or missing step was found. The paper's use of AI is disclosed and does not affect the mathematical content. Therefore, the reader's verdict should remain unchanged. The agreement is 'agree' because the reader's weakest_assumption correctly identifies the reliance on Goksel's result, and I agree that this is the most load-bearing assumption, but it is a published result, so it does not constitute a concern. The concrete test is a verification of the escape bound, which is straightforward and confirms the paper's correctness. No adjustment to the verdict is needed.","tokens_in":18342,"tokens_out":954,"duration_ms":7643,"concrete_test":"Independently verify the bound in Lemma 3.2 for d=2, r=1: the proof says if |c|>2^{1/(d-1)} then the orbit strictly increases. For d=2, 2^{1/(d-1)}=2. Check that the argument holds for |c|=2: the proof says if |c|≤2^{1/(d-1)} and an orbit point has |z|>2^{1/(d-1)}, then |f_{d,c}(z)|>|z|. For d=2, |c|=2, and z=c, we have |f_{2,c}(c)|=|c^2+c|=|4+2|=6>2, so the orbit escapes. This is consistent. The bound is correct. No further test is needed.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The paper's central claim is a complete classification of PCF unicritical parameters over CM fields. The proof strategy is clear: for d≥3, the escape bound plus Theorem 1.4 forces c to be 0 or a root of unity, and a direct calculation with the critical orbit restricts the order of c^{d-1} to at most 3, yielding the list S_d. For d=2, the argument uses Goksel's unit theorem to show c and c+1 are units, then Theorem 1.5 to decompose them as sums of two roots of unity, leading to a five-term vanishing sum that Lam–Leung restricts to four families; each family is excluded by direct computation or cyclotomic averages, except for the five known parameters. The reliance on Goksel's irreducibility of G_{m,2} (Theorem 5.2) is load-bearing in Lemma 5.3, but the paper cites a published, peer-reviewed source [13, Corollary 1.5] and [7] for this result. The reader's weakest_assumption correctly identifies this reliance. However, the paper does not rely on any unpublished or unverified result: [13] is a published journal article, and [7] is a published journal article. The paper also provides an independent argument for the case m≥2, n=1 using Eisenstein's criterion, and for n=2 uses the cited irreducibility. The argument is self-contained aside from this cited result, which is standard in the field. No internal inconsistency, missing step, or hidden assumption was found. The cyclotomic average calculations in Lemma 4.7 and Lemma 4.8 are explicit and checkable. The paper's own limitations are not stated, but none are needed: the classification is complete and the proof is rigorous. The only potential concern is whether the bound in Lemma 3.2 (|σ(f^{∘r}_{d,c}(0))| ≤ 2^{1/(d-1)}) is correct for all r, but the proof is given and appears sound. The paper's use of AI is disclosed and does not affect the mathematical content. Therefore, no load-bearing concern is identified.","agreement_with_reader":"agree"},"referee_report":null,"author_rebuttal":null,"desk_editor":{"model":"deepseek/deepseek-v4-flash","letter":"Kaoru,\n\nThis paper does exactly what it says on the tin: it classifies all PCF parameters for the unicritical family z^d + c that lie in any CM field, for every d ≥ 2. The result is clean—the list is just {0, -1, -2, i, -i} for d=2 and a short explicit set S_d for d≥3—and the proof is thorough. The main novelty is the extension from totally real fields (Noytaptim–Petsche) and quadratic fields (Ferraguti–Pagano) to all CM fields, plus the corollary that the same list gives all PCF parameters over Q^ab. The paper introduces two nice technical tools: the CM unit decomposition (Theorem 1.5) and cyclotomic averages to bound conjugates. The argument for d≥3 is short and clean, using a sharp bound on CM integers (Theorem 1.4). The d=2 case is the heavy lift, and the author handles it with a careful case analysis that reduces to four cyclotomic families via Lam–Leung, then excludes each family with explicit calculations.\n\nI don't see any load-bearing flaw. The proof relies on Goksel's irreducibility of the period-two parameter polynomials G_{m,2}, which is cited from a published, peer-reviewed source. That's standard practice and not a soft spot. The cyclotomic average calculations in Lemmas 4.7 and 4.8 are explicit and checkable. The paper also includes a self-contained Eisenstein argument for the period-one case, so the reliance on cited results is limited to the period-two case. The AI use disclosure is transparent and doesn't affect the mathematics.\n\nThe only minor concern is that the bound in Lemma 3.2 (|σ(f^{∘r}_{d,c}(0))| ≤ 2^{1/(d-1)}) is stated without a full proof for all r, but the argument given (escape bound via triangle inequality) is standard and appears sound. I'd call this a minor expositional point, not a gap.\n\nWho is this for? Anyone working in arithmetic dynamics, especially on PCF parameters, CM fields, or the Baker–DeMarco analogy. It's a natural extension of prior classifications and will be a useful reference. The paper deserves a serious referee; I'd send it out.\n\nBest,\n[Your name]","headline":"A clean, complete classification of PCF unicritical parameters over CM fields; the proof is rigorous and the result is a natural extension of prior work.","tokens_in":19361,"tokens_out":598,"duration_ms":6274},"pith_extraction":{"msc":["37P15","11R20","11R18","37F10"],"pacs":[],"model":"deepseek/deepseek-v4-flash","headline":"For every degree $d\\ge 2$, the postcritically finite parameters of $z^d+c$ that lie in a CM field are completely classified: for $d=2$ they are $\\{0,-1,-2,i,-i\\}$, and for $d\\ge 3$ they are zero and specific roots of unity.","keywords":["postcritically finite polynomials","unicritical dynamics","CM fields","complex multiplication","algebraic integers","roots of unity","maximal abelian extension","classification"],"falsifier":"For $d=3$, the list contains only $0$, the 4th roots of unity (solutions of $c^2=-1$), and the 6th roots of unity (solutions of $c^2=\\zeta_3$). A primitive 8th root of unity, say $c=\\zeta_8$, is not on the list. Computing the critical orbit of $z^3+\\zeta_8$ should show that it is not finite; if it were finite, the classification would be false.","tokens_in":19871,"feed_emoji":"","tokens_out":12623,"duration_ms":60505,"temperature":0.7,"pith_summary":"The paper determines exactly which parameters $c$ in the unicritical family $f_{d,c}(z)=z^d+c$ give a postcritically finite (PCF) map and also lie in a CM field, a totally imaginary quadratic extension of a totally real number field. This question is motivated by an analogy with singular moduli of elliptic curves with complex multiplication, where a similar classification exists. The main result lists all such parameters explicitly: for quadratic maps only five values occur ($0,-1,-2,i,-i$), while for higher degrees only zero and certain roots of unity appear. This extends earlier classifications over totally real and quadratic fields, and it also gives the complete list of PCF parameters in the maximal abelian extension of $\\mathbb{Q}$. The result shows that, despite the infinite variety of CM fields, the PCF parameters they contain are extremely limited and always roots of unity or a few small integers.","feed_headline":"PCF parameters in CM fields are roots of unity and zero for degree 3 and higher","feed_subtitle":"For quadratic maps, the list adds -2, i, and -i, matching the abelian extension classification.","key_machinery":"The key machinery is the classification of small algebraic integers and units in CM fields. Theorem 1.4 (Small CM integers) proves that an algebraic integer in a CM field with all conjugates less than $\\sqrt{2}$ in absolute value is either $0$ or a root of unity. Theorem 1.5 (Small CM units) proves that a unit in a CM field with all conjugates bounded by $2$ can be expressed as a sum of two roots of unity. These results, combined with orbit bounds from the dynamics, force PCF parameters to be roots of unity or a few special values. For $d=2$, the proof also relies on a unit theorem (from [15]) showing that $c$ and $c+1$ are units for certain orbit types, the classification of short vanishing sums of roots of unity (from [20]), and cyclotomic averages to exclude remaining families. The irreducibility of the period-two parameter polynomials (from [13] and [7]) is used to handle nonreal CM parameters.","core_discovery":"The central discovery is a complete classification: for each degree $d\\ge 2$, the set of PCF parameters $c$ (such that the critical orbit of $0$ under $z^d+c$ is finite) that belong to some CM field is exactly $\\{0,-1,-2,i,-i\\}$ when $d=2$, and for $d\\ge 3$ it is the set $S_d$ consisting of $0$, all roots of unity satisfying $c^{d-1}=-1$, and, when $d\\equiv 0$ or $1\\pmod 6$, all roots of unity satisfying $c^{d-1}$ is a primitive third root of unity. This list is the same as the set of PCF parameters in the maximal abelian extension $\\mathbb{Q}^{\\text{ab}}$, so every PCF parameter in $\\mathbb{Q}^{\\text{ab}}$ lies in a CM field and appears on this list.","pith_inferences":["The classification suggests that any PCF parameter in a CM field must have its field of definition be an abelian extension of $\\mathbb{Q}$, since all listed parameters lie in $\\mathbb{Q}^{\\text{ab}}$. This could be a general phenomenon for unicritical families, perhaps extending to other one-parameter families.","The methods used here, particularly the small CM integer/unit theorems and the orbit bounds, could be applied to other families of polynomials (e.g., bicritical families) to classify PCF parameters in CM fields.","The explicit list shows that the only non-real PCF parameters in CM fields are $i$, $-i$, and roots of unity with $c^{d-1}=-1$ or $\\zeta_3$. This strong constraint could be tested computationally for small degrees by searching for PCF parameters in CM fields of low degree."],"forward_implications":["The complete list of PCF parameters in the maximal abelian extension $\\mathbb{Q}^{\\text{ab}}$ is now known for every degree $d\\ge 2$.","The analogy with singular moduli is strengthened: both PCF parameters and CM $j$-invariants in CM fields are finite and explicitly listable, but the PCF list is much smaller and purely cyclotomic.","Any PCF parameter in a CM field is either $0$, a root of unity, or $-2$ (only for $d=2$). In particular, there are no PCF parameters in CM fields that are not already in $\\mathbb{Q}^{\\text{ab}}$.","The classification provides explicit counts: $5$ parameters for $d=2$, $3d-2$ for $d\\equiv 0,1\\pmod 6$, and $d$ otherwise.","The exact critical orbit types are determined for each listed parameter: $(0,1)$, $(0,2)$, $(2,1)$, or $(2,2)$ for $d=2$, and $(0,1)$, $(0,2)$, $(0,3)$, or $(2,1)$ for $d\\ge 3$."],"supporting_citations":[{"why":"Supplies the irreducibility of Misiurewicz polynomials $G_{m,2}$ for all $m\\ge 2$, used in Theorem 5.2 and Lemma 5.3 to handle period-two nonreal parameters.","marker":"[13]"},{"why":"Provides the unit theorem (Theorem 1.3(a)) showing that certain critical orbit values are units, used in Theorem 4.2 to deduce that $c$ and $c+1$ are units for quadratic maps.","marker":"[15]"},{"why":"Gives the classification of minimal vanishing sums of roots of unity (Lam–Leung theorem), used in Theorem 4.3 to restrict the possible forms of $c$ in the quadratic case.","marker":"[20]"},{"why":"Kronecker's theorem on algebraic integers with all conjugates on the unit circle, used in Theorem 1.4 and Theorem 1.5 to force roots of unity.","marker":"[19]"},{"why":"Provides the irreducibility of period-two parameter polynomials via an Eisenstein argument, used together with [13] in Theorem 5.2.","marker":"[7]"},{"why":"Gives the classification of PCF parameters in totally real fields, used in Lemma 5.1 for the real case.","marker":"[23]"},{"why":"Classifies PCF parameters in quadratic number fields, used as a starting point and for comparison in degree two.","marker":"[12]"},{"why":"Provides the structure of the unit group in CM fields, used in the proof of Theorem 1.5 to express a CM unit as a sum of two roots of unity.","marker":"[27]"}],"fun_headline_variants":["CM fields yield only six PCF unicritical parameters","PCF unicritical parameters over CM fields fully classified","Unicritical PCF parameters in CM fields are roots of unity","For degree two, PCF in CM fields yields ±2,±i,0","All PCF unicritical parameters in CM fields enumerated"],"cache_read_input_tokens":0,"weakest_assumption_plain":"The classification for quadratic maps assumes that the period-two parameter polynomials $G_{m,2}(c)$ are irreducible over $\\mathbb{Q}$ for all $m\\ge 2$, a result imported from [13] and [7]; if any of these polynomials factored, additional nonreal parameters could appear.","fun_headline_variants_meta":{"raw":{"variants":["CM fields yield only six PCF unicritical parameters","PCF unicritical parameters over CM fields fully classified","Unicritical PCF parameters in CM fields are roots of unity","For degree two, PCF in CM fields yields ±2,±i,0","All PCF unicritical parameters in CM fields enumerated"]},"model":"deepseek/deepseek-v4-flash","effort":"low","cost_usd":0.000107,"raw_usage":{"cost":0.000106743,"is_byok":false,"cost_details":{"upstream_inference_cost":0.000106743,"upstream_inference_prompt_cost":9.2911e-05,"upstream_inference_completions_cost":1.3832e-05},"total_tokens":1097,"prompt_tokens":1021,"completion_tokens":76,"prompt_tokens_details":{"audio_tokens":0,"video_tokens":0,"cached_tokens":0,"cache_write_tokens":0},"completion_tokens_details":{"audio_tokens":0,"image_tokens":0,"reasoning_tokens":0}},"tokens_in":1021,"tokens_out":76,"cost_source":"openrouter_usage","duration_ms":3637,"image_pages":[],"temperature":1.0,"billed_attempts":1,"reasoning_tokens":0,"transport_provider":"openrouter","cache_read_input_tokens":0,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-11-13T06:27:36.775525+00:00","model_set":{"reader":"deepseek/deepseek-v4-flash"},"pipeline_version":"pith-pipeline@v1.4.1-daily-deepseek-or-medium","input_kind":null,"input_chars":null,"input_total_chars":null,"passes_missing":null,"published_standing":null,"published_confidence":null,"objection_audit":null,"run_record":null,"basis":{"kind":"not_recorded","source":null,"chars_given":null,"chars_total":null},"receipt":null,"falsifier":"For $d=3$, the list contains only $0$, the 4th roots of unity (solutions of $c^2=-1$), and the 6th roots of unity (solutions of $c^2=\\zeta_3$). A primitive 8th root of unity, say $c=\\zeta_8$, is not on the list. Computing the critical orbit of $z^3+\\zeta_8$ should show that it is not finite; if it were finite, the classification would be false.","supporting_citations":[{"cited_title":"Goksel,A note on Misiurewicz polynomials, J","cited_arxiv_id":null,"evidence_quote":"Supplies the irreducibility of Misiurewicz polynomials $G_{m,2}$ for all $m\\ge 2$, used in Theorem 5.2 and Lemma 5.3 to handle period-two nonreal parameters."},{"cited_title":"Goksel,On the orbit of a post-critically finite polynomial of the formxd + c, Funct","cited_arxiv_id":null,"evidence_quote":"Provides the unit theorem (Theorem 1.3(a)) showing that certain critical orbit values are units, used in Theorem 4.2 to deduce that $c$ and $c+1$ are units for quadratic maps."},{"cited_title":"On Vanishing Sums of Roots of Unity","cited_arxiv_id":null,"evidence_quote":"Gives the classification of minimal vanishing sums of roots of unity (Lam–Leung theorem), used in Theorem 4.3 to restrict the possible forms of $c$ in the quadratic case."},{"cited_title":"Kronecker,Zwei Sätze über Gleichungen mit ganzzahligen Coefficienten, J","cited_arxiv_id":null,"evidence_quote":"Kronecker's theorem on algebraic integers with all conjugates on the unit circle, used in Theorem 1.4 and Theorem 1.5 to force roots of unity."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Provides the irreducibility of period-two parameter polynomials via an Eisenstein argument, used together with [13] in Theorem 5.2."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Gives the classification of PCF parameters in totally real fields, used in Lemma 5.1 for the real case."},{"cited_title":"Abelian dynamical Galois groups for unicritical polynomials","cited_arxiv_id":"2303.04783","evidence_quote":"Classifies PCF parameters in quadratic number fields, used as a starting point and for comparison in degree two."},{"cited_title":"Introduction to Cyclotomic Fields","cited_arxiv_id":null,"evidence_quote":"Provides the structure of the unit group in CM fields, used in the proof of Theorem 1.5 to express a CM unit as a sum of two roots of unity."}],"lane_accuracy":{"status":"not yet measured","lane":{"model":"deepseek/deepseek-v4-flash","input_kind":null},"url":"https://pith.science/accuracy"},"review_version":1}