{"id":"bb032932-017e-4b7d-b570-d3135e45335b","arxiv_id":"2501.07698","paper_version":1,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":7.0,"correctness_risk":"medium","formal_verification":"none","parameter_count":0,"one_line_summary":"The automorphism group of the circle equals the automorphism group of the intersection graph of all its chords, and the rational-chord subgraph is strongly universal and locally-complementation-invariant.","lead":"This paper shows that the group of continuous symmetries of a circle is exactly the group of symmetries of the graph whose vertices are all chords of the circle, with edges for crossing chords. It also builds a countable subgraph of that graph that contains every countable circle graph and is invariant under local complementation, two rare properties.","discovery_kind":"unification","skeptic_critique":{"model":"deepseek-v4-flash","headline":"The pigeonhole argument in Lemma 3.1 rests on a false geometric inference: two chords in one interval do not force a third chord to hit the incident boundary chords.","rationale":"The reader's weakest_assumption correctly identifies the exact soft spot in Lemma 3.1. My independent check of the geometry confirms that the inference is not just under-explained but false: two chords lying in the same cap of an incident pair can be simultaneously crossed by a chord that stays inside that cap and avoids the boundary chords. Since Lemma 3.1 is the sole mechanism by which an arbitrary automorphism of the circle graph is shown to map boundary cliques to boundary cliques, and since Theorem 1.1's surjectivity proof depends on that lemma, the correctness of the headline result is not established by the manuscript as written. The rest of the paper—universality of C_Q, local complementation invariance, and the Rado graph observation—appears plausible and is not affected by this gap, but the central theorem requires a repaired argument. I therefore maintain the reader's CONDITIONAL verdict: the claim is credible but not fully supported until Lemma 3.1 is rigorously fixed.","tokens_in":6397,"tokens_out":8468,"duration_ms":81062,"concrete_test":"Directly verify the counterexample: on the unit circle, draw chords A from angle 0° to 90° and B from 0° to 180° (incident at 0°). Draw X from 10° to 80° and Y from 20° to 70° (both on the arc (0°,90°)). Draw Z from 50° to 85°. Check by the alternating-endpoint criterion that Z intersects both X and Y, while both endpoints of Z lie on the arc (0°,90°), so Z does not cross A or B. If this configuration is realizable as stated, the geometric inference used in Lemma 3.1 is false, confirming the proof gap.","verdict_should_be":"UNCHANGED","load_bearing_attack":"Theorem 1.1 depends entirely on Lemma 3.1, and its proof contains an invalid geometric inference. In the proof of (2), after h maps non-incident crossing chords C,D to incident chords h(C),h(D), the author picks four chords F_i, one in each of the four intervals of S^1\\setminus(C\\cup D), and a chord W_ij intersecting both F_i and F_j. Since F_i and F_j are in different intervals, W_ij must intersect C or D. The proof then argues that if two images h(F_i), h(F_j) lie in the same one of the three intervals determined by the incident pair h(C),h(D), any chord intersecting both h(F_i) and h(F_j) must also intersect h(C) or h(D). This is false. For example, let h(C) be the chord from angle 0° to 90° and h(D) the chord from 0° to 180°, sharing endpoint 0°. Place h(F_i) with endpoints 10° and 80°, and h(F_j) with endpoints 20° and 70°, both on the arc (0°,90°). The chord Z with endpoints 50° and 85° intersects both h(F_i) and h(F_j) (endpoints alternate in both cases), yet both endpoints of Z lie on the same arc (0°,90°) of h(C), so Z avoids both h(C) and h(D). Thus the claimed contradiction does not follow. The key assertion that h preserves pairs of incident chords is left unproved, and with it the boundary-clique preservation and the surjectivity of π collapse. The lemma may still be true, but the proof as written has a concrete gap that must be repaired before Theorem 1.1 is supported.","agreement_with_reader":"agree"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper studies the circle graph C whose vertices are all chords of S^1 and whose edges join intersecting chords. The main result (Theorem 1.1) asserts that the natural map π from the homeomorphism group Aut(S^1) to Aut(C) is an isomorphism, i.e. every graph automorphism of C is induced by a circle homeomorphism. The proof proceeds through Lemma 3.1, which claims that every automorphism of C sends each boundary clique (all chords sharing a common endpoint) to a boundary clique. The paper also proves that the rational subgraph C_Q is strongly universal for countable circle graphs (Theorem 1.3) and is invariant under local complementation (Theorem 1.2), and that the Rado graph has this latter property (Observation 5.1).","tokens_in":6778,"tokens_out":4446,"duration_ms":40377,"significance":"If Theorem 1.1 is correct, it provides a natural and elegant analogue of Ivanov's theorem for the circle, identifying the automorphism group of a very concrete intersection graph with a classical topological group. The universality and local-complementation invariance of C_Q are also interesting, since such invariance is rare and connects to Bouchet's vertex-minor characterization. However, the proof of Theorem 1.1 rests entirely on Lemma 3.1, and the proof of that lemma contains a concrete geometric gap. Until that gap is repaired, the central claim is not established as written.","major_comments":[{"comment":"The pigeonhole step in the proof of (2) is invalid. After h(C) and h(D) are incident, the proof places four chords h(F_i) in the three intervals of S^1 \\ (h(C) ∪ h(D)) and concludes that if h(F_i) and h(F_j) lie in the same interval, then any chord meeting both must meet h(C) or h(D). This is false: a chord whose endpoints both lie on the arc of that interval can cross two nested chords inside the cap without touching either boundary chord. For example, let h(C) have endpoints at angles 0° and 90° and h(D) have endpoints at 0° and 180°. Let h(F_1) have endpoints at 10° and 80°, and h(F_2) at 20° and 70°. The chord Z with endpoints at 50° and 85° intersects both h(F_1) and h(F_2) (its endpoints alternate with each pair in cyclic order), yet both endpoints of Z lie on the arc (0°,90°), so Z avoids both h(C) and h(D). Thus the claimed contradiction does not follow. Since claim (2) is the basis for preserving boundary cliques, and Lemma 3.1 is the basis for the surjectivity of π in Theorem 1.1, this is a load-bearing gap. The lemma may still be true, but the proof as written needs a substantially more careful argument to rule out this configuration.","section":"§3, Lemma 3.1, proof of claim (2)"}],"minor_comments":[{"comment":"The phrase 'at least two of the h(Fi), ∈ [4]' contains a typo; it should read 'h(F_i), i ∈ [4]'.","section":"§3, proof of Lemma 3.1"},{"comment":"'Charactering' should be 'Characterizing'.","section":"§1, Introduction"},{"comment":"The reference to a Wikipedia page is unusual for a published article; a standard textbook or survey reference on distance-hereditary graphs would be more appropriate.","section":"§6, reference [18]"},{"comment":"The recursive placement of the new endpoints p_j, q_j into Q1 is plausible but terse; an explicit statement of how to extend a given finite cyclic order of rational points by two new rational points using the density property (1) would help the reader.","section":"§4, proof of Theorem 1.3"}],"recommendation":"major_revision","confidential_remarks":"The main theorem is attractive and the supplementary results on C_Q are of independent interest. The gap in Lemma 3.1 is concrete and specific, but I cannot tell from the manuscript whether it is repairable within the current framework or requires a new idea. The author should be given the opportunity to fix it; if the lemma turns out false, Theorem 1.1 would need substantial reformulation."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"Here's my take: the headline theorem is plausible and would be a nice result, but the proof as written has a genuine gap in Lemma 3.1, and the paper should not be accepted in this form.\n\nWhat's new: the claim that Aut(S^1) is the automorphism group of the chord intersection graph is natural and, as far as I know, new. The rational circle graph CQ being strongly universal for countable circle graphs is a clean, well-proved result; Theorem 1.3's inductive use of density of Q is straightforward and correct. The observation that CQ is invariant under local complementation is also interesting; the blow-up construction is clever. The Rado graph remark is correct.\n\nThe soft spot is the proof of Lemma 3.1, specifically step (2). The pigeonhole argument is invalid: two chords in the same interval determined by a pair of incident chords do not force a common transversal to hit those boundary chords. The counterexample is easy: for incident chords 0-90 and 0-180, place h(F_i) as chord 10-80 and h(F_j) as chord 20-70; then chord 50-85 crosses both but stays entirely within the cap 0-90, avoiding both boundary chords. So the claimed contradiction does not follow. This is not a minor typo; the proof of Theorem 1.1 rests on this lemma. I don't see an immediate repair in the text.\n\nAlso, Observation 4.1 is only sketched, and the 'elementary compactness argument' deserves more detail. The local-complementation-invariance proof (Theorem 1.2) is also more of a sketch than a full argument, particularly the step where the flipped realization 'coincides with C'. These are secondary but should be expanded.\n\nWho this is for: people working on infinite graphs, automorphism groups of natural objects, and circle graphs. The universality result alone is worth citing. But the main theorem needs a rigorous fix before I'd rely on it. The paper is short, readable, and the mistakes look repairable. It deserves a serious referee, not a desk rejection. I'd send it to review and ask for a revised proof of Lemma 3.1, plus details for the two sketches.","headline":"The universality and local-complementation results are solid and citable, but the main theorem's proof has a concrete gap in Lemma 3.1 that is not a minor typo.","tokens_in":7266,"tokens_out":3752,"would_cite":true,"duration_ms":34460,"reading_group":"yes","serious_thinker":"yes","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["05C63","05C25","05C10","05C62","05E18"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper proves that the automorphism group of the circle graph—the intersection graph of all chords of the circle—is exactly the homeomorphism group of the circle.","keywords":["circle graph","automorphism group","homeomorphism group of the circle","rational circle graph","strongly universal graph","local complementation","Rado graph","chord intersection graph"],"falsifier":"A concrete way to test Theorem 1.1 is to search for an automorphism of $C$ that maps three chords sharing a common endpoint on $S^1$ to three chords with no common endpoint; such a map would violate Lemma 3.1 and could be looked for by computing the intersection graph of a finite set of chords and checking whether any graph automorphism moves a boundary clique to a non-boundary clique.","tokens_in":6207,"feed_emoji":"⭕","tokens_out":11931,"duration_ms":104053,"temperature":0.7,"pith_summary":"This paper proves that the automorphism group of the circle graph is exactly the homeomorphism group of the circle, with the natural map $\\pi$ sending each homeomorphism to its induced action on chords being an isomorphism. In other words, every symmetry of the combinatorial object formed by all chord intersections comes from a continuous symmetry of the circle itself. The same argument shows that the countable graph of rational chords has automorphism group equal to the homeomorphisms fixing the rational points. The paper further proves that this rational circle graph is a strongly universal countable circle graph (it contains every countable circle graph as an induced subgraph) and is invariant under local complementation, so it behaves like a canonical infinite representative of the whole circle-graph class.","feed_headline":"Circle graph's symmetries are exactly the circle's homeomorphisms","feed_subtitle":"Chords as vertices: the intersection graph pins down every continuous symmetry of the circle.","key_machinery":"The workhorse is the boundary clique $K_x$, the set of all chords of $S^1$ incident with a given point $x$; these are maximal cliques of $C$, and distinct points give distinct boundary cliques. Lemma 3.1 proves that every automorphism of $C$ permutes these cliques, which converts a symmetry of an infinite graph into a symmetry of the underlying circle. Proposition 2.1 then supplies the conversion from a bijection preserving nestedness of pairs of points to a homeomorphism. For the rational subgraph, the density of the rationals lets arbitrary countable circle graphs be embedded by placing endpoints in cyclic order at rational points, and a blow-up construction eliminates shared endpoints so that local complementation can be realized by flipping one interval of the circle.","core_discovery":"The central discovery is that the intersection pattern of all chords of the circle is a complete combinatorial shadow of the circle's homeomorphism group. Theorem 1.1 states that the canonical map $\\pi$ from Aut($S^1$) onto Aut($C$) is an isomorphism; every automorphism of the chord graph is induced by a unique homeomorphism of $S^1$. The proof reconstructs the homeomorphism by first showing that any graph automorphism sends the maximal clique of chords through a point $x$ to the maximal clique of chords through some point $y$, then reading off a bijection of the circle and upgrading it to a homeomorphism via preservation of nestedness of pairs.","pith_inferences":["Because the proof recovers the homeomorphism from maximal cliques of incident chords, the same strategy may transfer to $d$-dimensional sphere graphs, where the analogue would be maximal cliques of hyperplane slices through a common boundary point; this is the open problem the paper raises for $d>1$.","The combination of strong universality and local-complementation invariance makes the rational circle graph a natural infinite litmus test for conjectural vertex-minor structure theories: any dichotomy conjectured for finite circle graphs can be checked against this one countable representative.","One could test how robust the theorem is by replacing the rationals with other dense proper subsets of the circle; the rational case gives the result for one such subset, and other choices would either extend the conclusion or reveal where the boundary-clique argument breaks."],"forward_implications":["The circle graph is a faithful combinatorial model of the full homeomorphism group of the circle.","The rational circle graph contains every countable circle graph as an induced subgraph, so it is a single countable host for all finite and countable chord-intersection graphs.","Local complementation never leaves the rational circle graph, so the graph is closed under the operation that generates vertex minors; consequently the finite forbidden-vertex-minor characterization extends to the countable setting.","Every automorphism of the rational circle graph is induced by a circle homeomorphism fixing the rational points, tying the graph's symmetry group to the arithmetic of the circle."],"supporting_citations":[{"why":"Supplies the classical fact that a bijection of $S^1$ preserving nestedness of pairs of points is a homeomorphism, used to finish the proof of Theorem 1.1.","marker":"[3]"},{"why":"Supplies the finite forbidden-vertex-minor characterization of circle graphs that Observation 4.1 extends to countable graphs.","marker":"[2]"},{"why":"Supplies the flip construction that realizes local complementation of a circle graph as a circle graph, used in the proof of Theorem 1.2.","marker":"[15]"}],"fun_headline_variants":["Chord intersections completely determine circle symmetries","Circle's homeomorphism group equals its chord graph automorphisms","Every circle symmetry appears as a chord graph symmetry","Chord graph: the circle's symmetry group in graph form"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is the geometric pigeonhole step in Lemma 3.1: after an automorphism maps two non-incident crossing chords to incident chords, two auxiliary chords lying in the same interval cut out by the image chords are claimed to force their connecting chord to avoid both boundary chords; if this fails for nested chords inside one cap, the proof that automorphisms preserve incident pairs collapses.","fun_headline_variants_meta":{"raw":{"variants":["Chord intersections completely determine circle symmetries","Circle's homeomorphism group equals its chord graph automorphisms","Every circle symmetry appears as a chord graph symmetry","Chord graph: the circle's symmetry group in graph form"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000813,"raw_usage":{"total_tokens":3470,"prompt_tokens":755,"completion_tokens":2715,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":371,"completion_tokens_details":{"reasoning_tokens":2653}},"tokens_in":371,"tokens_out":2715,"duration_ms":20385,"temperature":1.0,"reasoning_tokens":2653,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-10T20:40:16.982413+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"A concrete way to test Theorem 1.1 is to search for an automorphism of $C$ that maps three chords sharing a common endpoint on $S^1$ to three chords with no common endpoint; such a map would violate Lemma 3.1 and could be looked for by computing the intersection graph of a finite set of chords and checking whether any graph automorphism moves a boundary clique to a non-boundary clique.","supporting_citations":[{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the classical fact that a bijection of $S^1$ preserving nestedness of pairs of points is a homeomorphism, used to finish the proof of Theorem 1.1."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the finite forbidden-vertex-minor characterization of circle graphs that Observation 4.1 extends to countable graphs."},{"cited_title":null,"cited_arxiv_id":null,"evidence_quote":"Supplies the flip construction that realizes local complementation of a circle graph as a circle graph, used in the proof of Theorem 1.2."}],"review_version":1}