Pith. sign in

REVIEW 2 major objections 4 minor 16 references

Invariants of rational links represented by reduced alternating diagrams

T0 review · 2 major / 4 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper establishes that for every rational link, the braid index and the HOMFLY polynomial can be computed directly from the reduced alternating diagram (the nonalternating continued fraction form), by an algorithm that splits the…

desk verdict Useful new formulas for braid index and HOMFLY from minimal alternating diagrams, but the key sign-conversion proof leans too hard on 'left to the reader.' read the letter →

arxiv 1908.09458 v1 pith:272UGA7X submitted 2019-08-26 math.GN math.GT

Yuanan Diao , Claus Ernst , Gabor Hetyei This is my paper · ORCID
classification math.GNmath.GT MSC 57M2557M27
keywords rationallinkscontinuedfractionsbraidindexHOMFLYpolynomialalternatingdiagramsprimitiveblockdecompositionall-evenformFibonaccipolynomials
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

Rational links can be represented by many continued fractions for the same rational number. This paper establishes that the two standard invariants—braid index and HOMFLY polynomial—can be computed directly from the reduced alternating diagram, i.e. from the nonalternating continued fraction in which all partial denominators share one sign. The key is an algorithmic conversion, block by block, of that fraction into the all-even form on which the classical formulas were based. This matters because the all-even expansion is usually non-minimal and does not even exist when both integers defining the link are odd. The resulting formulas (Theorems 5.5 and 7.5) read the invariants off the minimal diagram, with mirror-image substitution covering the odd–odd case.

What carries the argument

The central object is the primitive block decomposition of a nonalternating continued fraction. A primitive block is either a single even partial denominator, or an odd-length stretch $a_m,a_{m+1},\dots,a_{m+2k}$ whose end entries are odd, whose interior even-position entries are even, and whose entries share one sign; except for a possible final exceptional block, the decomposition is unique when it exists. Alongside this sits a finite sign-tracking automaton (Figure 5) that records for each twistbox whether its crossing sign is positive or negative depending on the parity of the partial denominator and the position inside its block. The automaton is what proves that crossing signs are constant within each primitive block and opposite in adjacent blocks, and it is the machine from which both invariant formulas are read.

What would settle it

Compute the braid index of every rational link with denominator up to, say, $q=200$ using Theorem 5.5 and compare with the value from the all-even expansion of Definition 5.1; any mismatch would identify a counterexample to the sign rules, and the same comparison can be made for the HOMFLY matrix product against a direct skein computation for all rational links with at most 12 crossings.

Watch

Extended reading notes

Core claim

The paper's central claim is that the braid index and the HOMFLY polynomial of a rational link no longer need the special all-even continued fraction expansion. Starting from the reduced alternating diagram—equivalently, the nonalternating continued fraction $[0,a_1,\dots,a_n]$—the authors give an explicit primitive block decomposition and a sign-tracking rule (Theorems 4.3 and 4.4) that converts the fraction into all-even form block by block. This conversion yields a closed formula for the braid index (Theorem 5.5): $b(K)=1+\frac12$(sum of selected odd/even partial denominators chosen by the crossing signs of their twistboxes) plus a $0$ or $1/2$ correction term. It also yields a formula for the HOMFLY polynomial (Theorem 7.5) as the matrix product $H(a_n)\cdots H(a_1)$ applied to the standard column vector $(1,(a^2-1)/(az))^T$, where each $H(a_i)$ is a $2\times2$ matrix built from Fibonacci polynomials. When the numerator and denominator are both odd, so that no all-even form exists for the original diagram, the formulas apply to the mirror image and the substitution $a\mapsto a^{-1}$ recovers the invariant.

Load-bearing premise

Everything rests on the claim that, for every nonalternating fraction with a primitive block decomposition, crossing signs are constant within each block, opposite between adjacent blocks, and correctly encoded by the replacement rules of Theorem 4.4 with no hidden parity exception.

Editorial extensions

If this is right

  • The braid index of any rational link can be computed from its minimal alternating diagram alone, without first constructing a larger all-even diagram.
  • The HOMFLY polynomial becomes a finite product of $2\times2$ matrices with Fibonacci-polynomial entries, with the number of factors equal to the number of partial denominators in the reduced fraction rather than the inflated all-even expansion.
  • Rational links whose defining integers are both odd, which previously required a mirror-image workaround before applying the all-even formulas, are now handled by the same formulas with the substitution $a\mapsto a^{-1}$.
  • The two braid-index formulations for rational links—one based on the preferred standard form and one based on the alternative standard form—are shown to agree through elementary continued-fraction manipulations.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Because the sign automaton has only finitely many states, the parity rule in Theorem 4.4 could be checked exhaustively over all primitive blocks up to a fixed size; such a computer enumeration would give independent confirmation of the informal automaton analysis without requiring a formal case proof.
  • The same primitive-block scaffolding may extend to other alternating link families whose twistboxes share the same block-sign structure, potentially yielding HOMFLY formulas beyond rational links.
  • The Fibonacci-polynomial matrix entries suggest a path-counting interpretation: powers of the matrices $M(2)$ and $M(-2)$ count weighted lattice paths, so the HOMFLY polynomial of a rational link may be readable as a weighted path sum over the minimal diagram.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

2 major / 4 minor

Summary. The paper presents an algorithmic conversion of a rational link's continued fraction from the nonalternating (reduced alternating diagram) form to the all-even form used in Murasugi's braid-index formula and Lickorish-Millett's HOMFLY formula. The conversion is organized around primitive blocks (Definition 3.3) and a replacement rule (Theorem 4.4) that tracks crossing signs via the automaton of Figure 5. From this conversion the paper derives a braid-index formula (Theorem 5.5), compares it with the authors' earlier formula from [4] (Section 6), and derives a matrix-product HOMFLY formula stated directly in terms of the original partial denominators and crossing signs (Theorem 7.5). Worked examples include the knots and links 1402/1813, 3244/4195, and the two-component link 5075/17426.

Significance. Should the formulas hold in full generality, they give a practical way to compute both invariants from a minimal alternating diagram without first constructing the highly non-minimal all-even diagram, and they cover the case pq odd via mirror images. The paper provides several internal consistency checks: Example 5.4 verifies the Cromwell-Murasugi index of 1402/1813 both with Theorem 5.3 and with Definition 5.1; Example 6.5 computes the braid index of a two-component link in several ways; Example 7.4 illustrates the HOMFLY block product. These checks lend credence to the main formulas. The main weakness is proof completeness: the key sign-tracking steps underpinning Theorem 4.4 and Theorem 7.5 are delegated to the reader rather than proved or machine-checked.

major comments (2)
  1. [Section 4, Theorem 4.4] The replacement rules in Theorem 4.4 are the pivot on which both Theorem 5.5 and Theorem 7.5 rest, yet their proof is not a formal case enumeration. The proof states that the rules "follow from Proposition 3.2, after verifying that (-1)^tau(i) sign(a1) is the correct sign," and the rest is an informal description of the automaton states. In particular, rule (4) generates alternating strings of length |a_i|-1, so an off-by-one parity error at a block boundary would silently change every subsequent sign and hence both invariant formulas. Please supply a complete proof, or a machine-checked enumeration, covering all primitive-block types and all four replacement rules, and specify exactly how the automaton determines the parity count entering tau(i).
  2. [Section 7, Theorem 7.5] Theorem 7.5 is the paper's central HOMFLY result, but its proof ends with "The details of the verification are left to the reader." This is not a minor omission: the passage from the block products (7.10)-(7.11) to the per-entry matrices H(a_i) requires matching the all-even form produced by Theorem 4.4 with the order and conjugation conventions in Proposition 7.1, case by case. Please provide the detailed verification, and ideally add an independent check of a nontrivial example against Proposition 7.1 or the Duzhin-Shkolnikov formula.
minor comments (4)
  1. [Example 5.4] The displayed computation "1 + 1 + 2/2 + 3/2 + 5/2 + 3/2 = 8" is arithmetically 8.5; the intended value 8 is obtained as 1 + (1 + 2 + 3 + 5 + 3)/2. Please correct the displayed formula.
  2. [Example 7.4] The notation M(-2/2), M(-4/2), M(6/2), M(4/2) is ambiguous: read literally, M(-2/2) would be M(-1), which is not of the form M(2r) used in the paper. Please clarify whether these denote M(-2), M(-4), M(6), M(4) or the corresponding r-values, and correct the example.
  3. [Theorem 4.4 and Figure 5] The automaton in Figure 5 is introduced under the assumption ai > 0 and epsilon(B1) = +1, but Theorem 4.4 is stated for a general nonalternating continued fraction with s = sign(a1). Please state explicitly how the automaton and the parity rules are adapted when all ai have the opposite sign.
  4. [Section 6, Theorem 6.4] The proof of Theorem 6.4 verifies cases (i) and (ii) in detail and leaves cases (iii) and (iv) to the reader. Since this is a consistency check rather than a load-bearing step, a brief completion of the remaining cases would remove the asymmetry.

Circularity Check

0 steps flagged · score 1.0 of 10

No substantive circularity: the invariant formulas are derived from external Murasugi and Lickorish-Millett results via independent continued-fraction transformations; the self-citation to [4] is a non-load-bearing consistency check.

full rationale

The derivation chain is self-contained relative to established external results. Section 3 proves arithmetic transformations (Lemma 3.1, Proposition 3.2) purely at the level of continued fractions, without using link invariants. Theorem 4.3 derives the crossing-sign pattern from the diagram automaton of Figure 5, and Theorem 4.4 converts primitive blocks to all-even form; these are internal, noncircular steps. Theorem 5.5 is obtained by applying Murasugi's Cromwell-Murasugi index (external, Proposition 5.2) to the converted all-even form; Theorem 5.3 computes that index directly from the primitive-block structure. Theorem 7.5 is likewise obtained by substituting the all-even conversion into Lickorish-Millett's matrix product (Proposition 7.1) and compressing alternating runs via Fibonacci polynomials. The only self-citation is to the authors' preprint [4] (Theorem 6.1), and it is used solely as a comparison/consistency check in Section 6, not as a premise for the new formulas. No parameter is fitted, no invariant is defined in terms of the claimed output, and no uniqueness theorem is imported from the authors' prior work. The skeptics' concern about Theorem 4.4 and Theorem 7.5 leaving verification 'to the reader' is a rigor/correctness issue, not circularity.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The paper introduces no free parameters and no invented entities. It relies on established classification theorems and invariant formulas from the literature, which are cited; the main internal burden is the correctness of the primitive-block sign tracking, which is proven informally.

assumptions (5)
  • domain assumption Schubert classification of rational links (Theorem 4.1)
    Used to identify oriented and unoriented rational links from their fractions and to justify the transformation between non-preferred and preferred standard forms in Section 4.
  • domain assumption Murasugi's braid index formula for preferred standard form (Proposition 5.2)
    External result from Cromwell's book [3] used as the starting point for deriving the primitive-block braid index formula in Section 5.
  • domain assumption Lickorish-Millett HOMFLY formula for all-even continued fractions (Proposition 7.1)
    External result from [11] used as the starting point for the new HOMFLY matrix formula in Section 7.
  • standard math Standard arithmetic of continued fractions and matrix identities (Propositions 2.1, 2.2, Lemma 2.4)
    Underpins the Lagrange identity and the conversion lemma 3.1, and is used throughout Sections 2 and 3.
  • standard math Mirror image relation changes HOMFLY variable a to a^{-1}
    Invoked in Remark 7.6 and in the discussion around Figure 4 to extend formulas to non-preferred diagrams.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Invariants of rational links represented by reduced alternating diagrams." pith.science (2026). https://pith.science/paper/272UGA7X

@misc{pith2026190809458,
  author       = {Pith},
  title        = {Pith review of: Invariants of rational links represented by reduced alternating diagrams},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/272UGA7X}},
  note         = {Machine review of arXiv:1908.09458}
}
read the original abstract

A rational link may be represented by any of the (infinitely) many link diagrams corresponding to various continued fraction expansions of the same rational number. The continued fraction expansion of the rational number in which all signs are the same is called a {\em nonalternating form} and the diagram corresponding to it is a reduced alternating link diagram, which is minimum in terms of the number of crossings in the diagram. Famous formulas exist in the literature for the braid index of a rational link by Murasugi and for its HOMFLY polynomial by Lickorish and Millet, but these rely on a special continued fraction expansion of the rational number in which all partial denominators are even (called {\em all-even form}). In this paper we present an algorithmic way to transform a continued fraction given in nonalternating form into the all-even form. Using this method we derive formulas for the braid index and the HOMFLY polynomial of a rational link in terms of its reduced alternating form, or equivalently the nonalternating form of the corresponding rational number.

Figures

Figures reproduced from arXiv: 1908.09458 by the authors.

Figure 1
Figure 1. Without the orientation: the sign convention used to define the standard form of non-oriented rational links; With orientation: preferred standard form of ori￾ented rational links. presented in this form by the symbol b(q, p) or by the vector (a1, a2, . . . , an), and call such a diagram a standard diagram of the unoriented rational link. Notice that the left end of the diagram is fixed, the closing on the right end… view at source ↗
Figure 2
Figure 2. The crossing sign convention at a crossing in an oriented link diagram. sign given in [PITH_FULL_IMAGE:figures/full_fig_p008_2.png] view at source ↗
Figure 3
Figure 3. The crossing sign convention for crossings in an oriented link diagram: the signs in front of the aj ’s indicate which strand is on top as indicated in [PITH_FULL_IMAGE:figures/full_fig_p009_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: The procedure of changing a two component rational link in a non-preferred standard form to a preferred standard form. Define the sign of a twistbox Bi as the crossing sign of all crossings in it and denote it (by abuse of notation) by ε(Bi). Under the assumption that …
Figure 5
Figure 5. Figure 5: Automaton, parsing the signs of crossings in an alternating link. o and e stand for an odd or even number of crossings in a twistbox respectively. Theorem 4.3. Suppose ˜b(q, p) is represented by a nonalternating continued fraction p/q = [0, a1, . . . , an] that has a p…
Figure 6
Figure 6. Figure 6: The two bridge link ¯b(17426, 5075) = (0, 3, 2, 3, 3, 1, 2, 3, 4, 4) in an alter￾native standard form with signed vector (3, 2, 3, 3, −1, −2, −3, 4, −4). Notice that it is not in a preferred standard form. Example 6.5. Consider the rational link given in [PITH_FULL_IM…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 16 canonical work pages

  1. [4]

    Y. Diao, C. Ernst, G. Hetyei and P. Liu, A diagrammatic approach for determining the braid index of alternating links, preprint 2018

    work page 2018

  2. [1]

    Burde, H

    G. Burde, H. Zieschang and M. Heusener Knots, De Gruyter Studies in Mathematics 5, 2013

    work page 2013

  3. [2]

    Conway, An enumeration of knots and links, and some of their algebraic properties, 1970 Computational Problems in Abstract Algebra (Proc

    J.H. Conway, An enumeration of knots and links, and some of their algebraic properties, 1970 Computational Problems in Abstract Algebra (Proc. Conf., Oxford, 1967) pp. 329–358 Pergamon, Oxford

    work page 1970

  4. [3]

    Cromwell, Knots and links, Cambridge University Press, 2004

    P. Cromwell, Knots and links, Cambridge University Press, 2004

    work page 2004

  5. [5]

    Duzhin and M

    S. Duzhin and M. Shkolnikov, A formula for the HOMFLY polynomial of rational links, Arnold Math. J. 1 (2015), 345–359

    work page 2015

  6. [6]

    Euler, Elements of algebra, Translated from the German by John Hewlett, Reprint of the 1840 edition, with an introduction by C

    L. Euler, Elements of algebra, Translated from the German by John Hewlett, Reprint of the 1840 edition, with an introduction by C. Truesdell, Springer-Verlag, New York, 1984, lx+593 pp

    work page 1984

  7. [7]

    Franks and R

    J. Franks and R. Williams Braids and The Jones Polynomial , Trans. Amer. Math. Soc., 303 (1987), 97–108

    work page 1987

  8. [8]

    Freyd, D

    P. Freyd, D. Yetter, J. Hoste, W. Lickorish, K. Millett and A. Ocneanu A New Polynomial Invariant of Knots and Links, Bull. Amer. Math. Soc. (N.S.) 12 (1985), 239–246

    work page 1985

Show all 16 references
  1. [9]

    Goldman, and L.H

    J.R. Goldman, and L.H. Kauffman, Rational tangles, Adv. in Appl. Math. 18 (1997), 300–332

    1997

  2. [10]

    Kauffman, and S

    L.H. Kauffman, and S. Lambropoulou, On the classification of rational tangles, Adv. in Appl. Math. 33 (2004), 199–237

    2004

  3. [11]

    Lickorish and Kenneth C

    W.B.R. Lickorish and Kenneth C. Millett, A polynomial invariant of oriented links. Topology 26 (1987), 107–141

    1987

  4. [12]

    Morton Seifert Circles and Knot Polynomials , Math

    H. Morton Seifert Circles and Knot Polynomials , Math. Proc. Cambridge Philos. Soc. 99 (1986), 107–109

    1986

  5. [13]

    Murasugi On The Braid Index of Alternating Links, Trans

    K. Murasugi On The Braid Index of Alternating Links, Trans. Amer. Math. Soc. 326 (1991), 237–260

    1991

  6. [14]

    Przytycki and P

    J. Przytycki and P. Traczyk Conway Algebras and Skein Equivalence of Links , Proc. Amer. Math. Soc. 100 (1987), 744–748. INV ARIANTS OF RATIONAL LINKS REPRESENTED BY REDUCED ALTERNATING DIAGRAMS 23

    1987

  7. [15]

    Schubert, Knoten mit zwei Br¨ ucken, Mathematische Zeitschrift,65 (1956), 133–170

    K. Schubert, Knoten mit zwei Br¨ ucken, Mathematische Zeitschrift,65 (1956), 133–170

    1956

  8. [16]

    Yamada The Minimal Number of Seifert Circles Equals The Braid Index of A Link , Invent

    S. Yamada The Minimal Number of Seifert Circles Equals The Braid Index of A Link , Invent. Math. 89 (1987), 347–356. † Department of Mathematics and Statistics, University of North Carolina Charlotte, Charlotte, NC 28223 ∗ Department of Mathematics, Western Kentucky University...

    1987

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.