{"id":"961c7f31-1a0b-4118-9d9c-d1df81acff39","arxiv_id":"1908.06894","paper_version":6,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"For a very general hypersurface of degree d at least n+3 in P^{n+1}, any dominant rational map of prime degree to a smooth projective n-fold has a uniruled target, and the target is rationally connected when n is at most 3.","lead":"This paper studies maps from a very general high-degree hypersurface X to another space Y. It proves that when the map has prime degree, Y must be uniruled, meaning it is swept out by rational curves, and in dimension three the target must be rationally connected.","discovery_kind":"extension","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The main contradiction in Theorem 1.5 depends on extending Guerra–Pirola's Hodge and Cayley–Bacharach reduction to arbitrary dimension, and the paper supplies no proof of that extension for n ≥ 3.","rationale":"The Reader's weakest assumption identifies exactly the same load-bearing point: Step 5 imports from [5] and [3] a Hodge-theoretic reduction and a Cayley–Bacharach bound that were proved for surfaces, with no proof for arbitrary n. My stress-test confirms that this is the single point on which the contradiction in Theorem 1.5 depends. The second fragile point mentioned by the Reader, the 'This is absurd' moduli assertion in Theorem 1.4(2), is real but secondary: Theorem 1.4(1) is already obtained before that step from Theorem 1.5, so the unproved moduli assertion does not threaten the main uniruledness claim. I find no visible internal contradiction in the pencil/ramification argument of Steps 1–4, and the use of Matsumura–Monsky in Step 3 is standard. The central claim is plausible and the method may be repairable, but as written the key Hodge and Cayley–Bacharach inputs are not established for n ≥ 3. Because the Reader already made this the condition of a CONDITIONAL verdict, my read does not move the verdict.","tokens_in":8683,"tokens_out":20212,"duration_ms":231222,"concrete_test":"Obtain the full proof of [5, Proposition 3.5.2] and check whether the argument is dimension-independent or explicitly uses surface-specific input (e.g., Noether–Lefschetz for surfaces, h^{2,0} calculations, or the classification of surfaces). Then attempt to rewrite the same Hodge–Lefschetz step for n = 3, d = 6: replace the surface hyperplane section theorem by the Lefschetz theorem for threefolds and verify every step of the derivation of pg(Y) = 0; if the proof terminates, re-derive Proposition 2.2 for this case from the explicit Hodge numbers of X_6 and see whether p ≥ d − n = 3 follows. If the proof requires surface-specific input, the concern lands and the central theorem is unsupported for n ≥ 3.","verdict_should_be":"UNCHANGED","load_bearing_attack":"The engine behind Theorem 1.4(1) is Theorem 1.5, and the proof of Theorem 1.5 ends in a contradiction between Step 4 and Step 5. Step 4 gives p ≤ d − n − 1 using the geometry of the pencil and pseudo-effectiveness of K_W. Step 5 asserts that the imported Hodge–Lefschetz argument of [5, Section 3.5] forces pg(Y) = 0, and then states Proposition 2.2, which gives the Cayley–Bacharach lower bound p ≥ d − n. These two inequalities are precisely the contradiction. The cited results in [5] are for a surface S ⊂ P^3, and [3] is applied in that surface context; the present paper gives no derivation of either pg(Y) = 0 or p ≥ d − n for a hypersurface of dimension n ≥ 3. The introduction says only 'By Hodge theory (as Section 3.5 in [5]) one has only to consider...', and Proposition 2.2 is stated without proof, as a reference. If the reduction to pg(Y) = 0 fails, Proposition 2.2 cannot be invoked; if the Cayley–Bacharach bound fails in higher dimension, the two inequalities no longer conflict. This is a load-bearing gap in the central argument, not a mere omission of detail: without Step 5 there is no contradiction and no proof of Theorem 1.5.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies dominant rational maps from a very general hypersurface X_d ⊂ P^{n+1} of degree d ≥ n+3 to smooth projective n-folds Y. The main results are Theorem 1.4 and Theorem 1.5. Theorem 1.5 asserts that if the degree p = deg f is prime and Y is non-uniruled, then the restriction of f to a very general hyperplane section H_d has degree p, i.e. it is not birational to its image. Theorem 1.4 derives that under the prime-degree assumption Y is uniruled, and for n ≤ 3 that Y is rationally connected. The proof uses a Lefschetz pencil of hyperplane sections, a ramification computation, a Hodge-theoretic reduction to pg(Y)=0, and a Cayley-Bacharach lower bound on the degree, which together produce a contradiction.","tokens_in":9011,"tokens_out":6674,"duration_ms":64964,"significance":"If correct, the results are a substantial advance: they provide the first restrictions for n ≥ 3 on function fields of non-uniruled n-folds admitting dominant rational maps from a very general hypersurface, and they give evidence for Conjectures 1.1 and 1.2. The strategy of passing to hyperplane sections and exploiting the primality of the degree is natural, and the proofs combine deep tools (Lefschetz theory, Hodge theory, Cayley-Bacharach, MMP). The main caveat is that a load-bearing input, the higher-dimensional Hodge and Cayley-Bacharach bound, is not proved or referenced in the paper; the significance is therefore conditional on that gap being filled.","major_comments":[{"comment":"The proof of Theorem 1.5 derives the contradiction from two inequalities: Step 4 gives p ≤ d−n−1, while Step 5 gives p ≥ d−n via Proposition 2.2. The latter rests entirely on two imported statements: (i) Hodge theory forces pg(Y)=0 for a dominant rational map from a very general hypersurface of degree d ≥ n+3, cited as '[5, Section 3.5]'; and (ii) the Cayley-Bacharach bound p ≥ d−n, stated as Proposition 2.2 without proof, with reference to [3] and [5]. Both [3] and [5] are written for surfaces in P^3, and the paper provides no proof or reference covering arbitrary n. If the reduction to pg(Y)=0 fails in dimension n, Proposition 2.2 cannot be applied; if the Cayley-Bacharach bound fails in higher dimension, the two inequalities no longer conflict. This is a load-bearing gap in the central argument, not a mere missing detail. Please supply a complete proof of Proposition 2.2 and of the pg(Y)=0 reduction in the stated generality, or cite precise theorems that establish them for all n.","section":"Section 2, Step 5 and Proposition 2.2"},{"comment":"In the case deg(f|H_d)=1, the proof dismisses the possibility that two very general hyperplane sections X_d ∩ H_1∩...∩H_k and X_d ∩ H'_1∩...∩H'_k are both birational to the same Z with the sentence 'This is absurd.' This is not immediate, and it is part of the induction establishing assertion (2). Please provide a proof or a reference showing that very general such complete intersections cannot be birational to each other, or give another argument ruling out this configuration.","section":"Theorem 1.4, final paragraph of proof"}],"minor_comments":[{"comment":"The title contains a typo: 'HYPERSURF ACE' should be 'HYPERSURFACE'.","section":"Title"},{"comment":"The phrase 'boundedness of pluricanocal maps' should read 'boundedness of pluricanonical maps'.","section":"Introduction, first paragraph"},{"comment":"The phrase 'parametrized by a polynomial functio n g(t)' contains a typo; it should be 'function'.","section":"Lemma 2.1(3)"},{"comment":"The statement 'Since Y is non-uniruled, so is its birational model W. Hence K_W is pseudo-effective' uses the theorem that a smooth projective variety is non-uniruled iff its canonical class is pseudo-effective; please cite a reference for this fact (e.g., the relevant MMP or the paper establishing it in all dimensions).","section":"Step 4"},{"comment":"The conclusion that the composite map H_d → Z has degree ≥ 2 by 'the same reasoning of the ending part of the proof of Theorem 1.4' is not spelled out; please make this reasoning explicit.","section":"Proposition 1.3, proof"}],"recommendation":"major_revision","confidential_remarks":"The central issue is the missing proof or precise reference for the Hodge-theoretic reduction to pg(Y)=0 and the Cayley-Bacharach bound in arbitrary dimension. This is not a question of novelty or scope but of completeness of the proof of the main theorem. If the authors can fill this gap, the paper is likely acceptable. The base case uses a paper by the first author with Pirola ([9]); this citation is reasonable, though the authors should make clear that the present results do not depend on any unpublished work. I recommend major revision."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Dear colleague,\n\nThis is a paper you should know about, but I would not take its main theorem as established. The paper is well organized and the method is genuinely interesting, but the proof of Theorem 1.5 has an imported step that carries the whole argument and is not proved here.\n\nWhat is new: Theorem 1.4 is the first restriction of its kind for very general hypersurfaces in dimensions n ≥ 3. The strategy—restrict to a hyperplane section, study the pencil of hyperplane sections, compare ramification of the induced map between base curves, then combine a lower bound from Cayley-Bacharach with an upper bound from pseudo-effectivity of K_W—is a plausible extension of Guerra–Pirola’s surface argument. Lemma 2.1 is solid, and the divisibility deg f|Hd | deg f is a useful observation. The induction in Theorem 1.4 is mostly fine, and the paper is honest about what remains conjectural.\n\nThe soft spot is exactly where the stress-test note points. In Step 5 of Theorem 1.5, the paper cites [5, Section 3.5] for the reduction to pg(Y)=0 and then states Proposition 2.2, which gives p ≥ d−n. Both are imported from the surface case; no proof is supplied for n ≥ 3. The reduction may be true, but the paper does not establish it, and Proposition 2.2 is stated without proof in this generality. Since the contradiction in Theorem 1.5 is precisely Step 4’s p ≤ d−n−1 versus Step 5’s p ≥ d−n, an unsupported Step 5 is a load-bearing gap, not a minor omission. A referee must ask for a written proof of Proposition 2.2 and of the pg(Y)=0 reduction in all dimensions, or a citation to an existing proof.\n\nA smaller issue: in the proof of Theorem 1.4, the claim that two very general hyperplane sections cannot both be birational to the same Z is dismissed as “absurd” without proof. It is likely true, but it needs an argument (e.g., non-isotrivality of the family or finiteness of automorphism groups). Also, the supplied abstract in the metadata describes a different version of the paper; that looks like an arXiv update error, but it should be fixed.\n\nOverall, I think this paper deserves a serious referee. If the higher-dimensional Hodge/Cayley-Bacharach reduction can be supplied, the result would be a solid advance. As it stands, I would not cite it as the source of the theorem, but I would engage with the manuscript and push for the missing proof.","headline":"A serious, cleanly organized paper whose central theorem rests on an imported higher-dimensional Hodge/Cayley-Bacharach step that is not proved and carries real weight; still deserves a rigorous peer review.","tokens_in":9540,"tokens_out":2976,"would_cite":false,"duration_ms":30686,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["14E05","14J30","14J70"],"pacs":[],"model":"deepseek-v4-flash","headline":"Prime-degree rational maps from very general hypersurfaces force the target to be uniruled.","keywords":["dominant rational map","hypersurface","prime degree","uniruled","rationally connected","Cayley-Bacharach property","Lefschetz pencil","Hodge theory"],"falsifier":"Take $n=4$, $d=7$, and search for a smooth projective non-uniruled 4-fold $Y$ with a dominant rational map from a very general degree-7 hypersurface $X_4 \\subset \\mathbb{P}^5$ whose degree is a prime number; existence of such a map would refute Theorem 1.4. Alternatively, test the imported Hodge assertion by computing $p_g(Y)$ for a smooth projective $n$-fold $Y$ admitting a generically finite rational map from a very general hypersurface; finding $p_g(Y)>0$ would break the Step 5 reduction on which the contradiction depends.","tokens_in":8501,"feed_emoji":"📐","tokens_out":10370,"duration_ms":96193,"temperature":0.7,"pith_summary":"This paper studies dominant rational maps from a very general hypersurface $X_d \\subset \\mathbb{P}^{n+1}$ of degree $d \\ge n+3$ to smooth projective $n$-folds $Y$. It aims to prove that if the map has prime degree, then $Y$ cannot be a non-uniruled variety: $Y$ must be uniruled (covered by rational curves), and for $n \\le 3$ it must be rationally connected. The motivation is to understand which finitely generated fields can sit inside the function field of such a hypersurface. The proof forces a contradiction between the Cayley-Bacharach lower bound $p \\ge d-n$ and a ramification-theoretic upper bound $p \\le d-n-1$ obtained when the restriction of the map to a general hyperplane section is birational.","feed_headline":"Prime-degree map targets must be uniruled","feed_subtitle":"A very general hypersurface of degree at least n+3 admits no prime-degree rational map to a non-uniruled n-fold.","key_machinery":"The argument runs on two inequalities for the degree $p$. A Hodge-theoretic reduction (imported from [5]) allows the proof to assume $p_g(Y)=0$, after which the Cayley-Bacharach property, via [3], puts the points of a general fibre in special position and forces $p \\ge d-n$. The other side studies a Lefschetz pencil of hyperplane sections $H(t)$ of $X_d$. Using the norm map of the extension $\\mathbb{C}(Y) \\subset \\mathbb{C}(X_d)$, the paper derives the identity $\\operatorname{Norm}(h_1/h_2+t) = (g/h'_2)^r$, which shows that the degree of the restriction $f|_{H_d}$ divides $p$, and that the images of the hyperplane sections move in a pencil whose parameter polynomial has degree $p/\\deg(f|_{H_d})$. If the restriction were birational, this pencil yields a degree-$p$ cover $\\mathbb{P}^1 \\to \\mathbb{P}^1$; either it is cyclic Galois, giving a nontrivial $\\mathbb{Z}/p$-action on $X_d$ in contradiction to the triviality of the automorphism group of a very general high-degree hypersurface, or it has at most one totally ramified point, and the ramification divisor then gives $p \\le d-n-1$ from the canonical class. The contradiction with $p \\ge d-n$ establishes the theorem.","core_discovery":"The central claim is Theorem 1.4: for $n \\ge 3$, a very general hypersurface $X_d \\subset \\mathbb{P}^{n+1}$ of degree $d \\ge n+3$ admits no dominant rational map of prime degree $p$ to a smooth projective non-uniruled $n$-fold. In fact the theorem describes the maximal rationally connected fibration $Y \\dashrightarrow Z$: if $Y$ is not rationally connected, then $3 \\le \\dim Z \\le n-1$, the restriction of $f$ to a very general intersection of $s = n-\\dim Z$ hyperplanes is birational to its image, the induced map from that slice to $Z$ has composite degree, and $p_g(Z) = q(Z) = 0$. For $n \\le 3$ the conclusion is stronger: $Y$ is rationally connected. The supporting Theorem 1.5 says that when $Y$ is non-uniruled, restriction to a very general hyperplane section preserves the degree $p$; the contradiction is obtained by comparing $p \\ge d-n$ from the Cayley-Bacharach property with $p \\le d-n-1$ from the pencil and ramification calculation.","pith_inferences":["If the theorem is correct, it is a field-theoretic statement: every subfield of $\\mathbb{C}(X_d)$ whose index is prime must be the function field of a uniruled variety, so non-uniruled function fields cannot occur at prime index; the paper only hints at these applications.","The same mechanism may apply to other very general varieties of general type, but the paper notes that Kodaira-dimension-zero or intermediate cases (such as a very general quintic threefold) require a different approach.","One route to Conjecture 1.1 for composite degrees is to show that the composite-degree map to the MRC base $Z$ in Theorem 1.4(2) violates the Cayley-Bacharach bound after slicing; this is not done in the paper."],"forward_implications":["A very general hypersurface of degree $d \\ge n+3$ has no dominant rational map of prime degree to any non-uniruled smooth projective $n$-fold.","For $n=3$, any target of a prime-degree dominant rational map from such a hypersurface is rationally connected.","The composite-degree alternative in Theorem 1.4(2) isolates the only possible obstruction to full rational connectedness: the restriction to a hyperplane slice is birational and the induced map to the MRC base has composite degree.","Under additional factoriality and largeness assumptions, the abstract gives the optimal bound $\\deg f \\le \\deg X_d$, and in some cases forces $Y \\cong \\mathbb{P}^n$."],"supporting_citations":[{"why":"Supplies the Hodge-theoretic reduction to $p_g(Y)=0$ and the Cayley-Bacharach lower bound; Step 5 imports Proposition 3.5.2 from this source.","marker":"[5]"},{"why":"Supplies Proposition 2.1, used to prove the Cayley-Bacharach inequality $p \\ge d-n$ in Proposition 2.2.","marker":"[3]"},{"why":"Supplies the fact that a very general high-degree hypersurface has trivial automorphism group, which rules out the cyclic Galois case.","marker":"[13]"},{"why":"Provides the known $n \\le 2$ cases and the rationality statement for $\\dim Z = 2$ used in the induction.","marker":"[9]"},{"why":"Supplies the existence and birational properties of maximal rationally connected fibrations used in the proof of Theorem 1.4.","marker":"[4]"}],"fun_headline_variants":["Very general hypersurfaces reject prime maps to non-uniruled","Prime-degree maps from hypersurfaces never hit non-uniruled","Hypersurface prime maps force uniruled targets","No prime-degree maps from very general hypersurfaces to non-uniruled"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The contradiction rests on the imported claim, cited to [5], that a dominant rational map from a very general hypersurface forces $p_g(Y)=0$ in every dimension, although the cited statement is proved for surfaces and the paper does not prove its higher-dimensional extension.","fun_headline_variants_meta":{"raw":{"variants":["Very general hypersurfaces reject prime maps to non-uniruled","Prime-degree maps from hypersurfaces never hit non-uniruled","Hypersurface prime maps force uniruled targets","No prime-degree maps from very general hypersurfaces to non-uniruled"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.002029,"raw_usage":{"total_tokens":7934,"prompt_tokens":996,"completion_tokens":6938,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":612,"completion_tokens_details":{"reasoning_tokens":6860}},"tokens_in":612,"tokens_out":6938,"duration_ms":46027,"temperature":1.0,"reasoning_tokens":6860,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T12:32:59.504106+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Take $n=4$, $d=7$, and search for a smooth projective non-uniruled 4-fold $Y$ with a dominant rational map from a very general degree-7 hypersurface $X_4 \\subset \\mathbb{P}^5$ whose degree is a prime number; existence of such a map would refute Theorem 1.4. Alternatively, test the imported Hodge assertion by computing $p_g(Y)$ for a smooth projective $n$-fold $Y$ admitting a generically finite rational map from a very general hypersurface; finding $p_g(Y)>0$ would break the Step 5 reduction on which the contradiction depends.","supporting_citations":[{"cited_title":"Guerra and G","cited_arxiv_id":null,"evidence_quote":"Supplies the Hodge-theoretic reduction to $p_g(Y)=0$ and the Cayley-Bacharach lower bound; Step 5 imports Proposition 3.5.2 from this source."},{"cited_title":"Cheltsov, Points in projective spaces and applications","cited_arxiv_id":null,"evidence_quote":"Supplies Proposition 2.1, used to prove the Cayley-Bacharach inequality $p \\ge d-n$ in Proposition 2.2."},{"cited_title":"Matsumura and P","cited_arxiv_id":null,"evidence_quote":"Supplies the fact that a very general high-degree hypersurface has trivial automorphism group, which rules out the cyclic Galois case."},{"cited_title":"Lee and G","cited_arxiv_id":null,"evidence_quote":"Provides the known $n \\le 2$ cases and the rationality statement for $\\dim Z = 2$ used in the induction."},{"cited_title":"Graber, J","cited_arxiv_id":null,"evidence_quote":"Supplies the existence and birational properties of maximal rationally connected fibrations used in the proof of Theorem 1.4."}],"review_version":1}