REVIEW 4 major objections 4 minor 47 references
Universal Triangle Covering Curve and Polygonal Chain: Escaping Forest and Fitting Worm
T0 review · 4 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash
Pith's one-line read A path escapes every translated, rotated triangle exactly when one weighted support-function inequality holds for every phase angle.
desk verdict The paper's support-function equivalence for arbitrary triangles is a genuine advance and probably correct; the existence and convergence proofs are the soft spot, not the central inequality. read the letter →
The pith
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
The reading
What carries the argument
The support function of the complete path, $h(\varphi)=\max_p r(p)\cdot(\cos\varphi,\sin\varphi)$, taken over the curve together with the segment from the origin to its start so that $h(\varphi)\ge 0$. The argument is carried by the weighted slack identity: with side lengths $L_1,L_2,L_3$ and distances $d_1,d_2,d_3$ from the origin to the three supporting lines of a translated triangle, the weighted sum $L_1d_1+L_2d_2+L_3d_3$ equals twice the triangle's area for every starting point. This identity, together with the phase-shifted normal directions, converts the requirement that the path reach at least one of the three lines for every translation and rotation into a single scalar inequality involving $h$ at the three phases. The intermediate value theorem supplies the transition from inequality to actual boundary crossing because the path is continuous and starts at the origin, which lies inside the translated triangle.
What would settle it
Choose a non-isosceles triangle, say base angles $20^\circ$ and $50^\circ$, and compute a path that satisfies the inequality at every phase. Then test the original problem directly: sample starting points densely inside the triangle and orientations $\theta\in[0,2\pi)$, and check whether the path intersects the translated and rotated triangle boundary in every case. A single sampled pair for which the path stays strictly inside the triangle while the inequality holds would refute Theorem 4, since the theorem asserts that no such pair can exist.
Extended reading notes
Core claim
The central claim is that escape from a triangle is not a family of geometric impossibilities spread over continuously many positions and orientations; it is one weighted inequality per phase. For the normalized triangle with unit base and base angles $\alpha,\beta$, the three sides have outward normals whose rotation phases are $t+\pi+\alpha$, $t+\pi-\beta$, and $t$, with side lengths $\sin\beta/\sin(\alpha+\beta)$, $\sin\alpha/\sin(\alpha+\beta)$, and $1$. The equilibrium identity $L_1 n_1+L_2 n_2+L_3 n_3=0$ forces the weighted sum of the three supporting-line offsets to equal the constant $\sin\alpha\sin\beta/\sin(\alpha+\beta)$, twice the triangle's area. Hence a path whose weighted support sum reaches that constant must, by continuity and the intermediate value theorem, cross at least one boundary line of every translated and rotated triangle; conversely, if the inequality fails, the proof constructs a translated triangle whose interior contains the whole path, so escape fails. This exact characterization is the load-bearing result, and the existence, polygonal convergence, and discretized convergence theorems all hang on it.
Load-bearing premise
Everything in the paper depends on the equivalence, proved in the author's earlier papers and invoked here, between Bellman's original escape problem and the reformulation in which the path is fixed at the origin while the triangle translates and rotates; if that equivalence has a flaw, the support inequality describes a different condition from the original problem.
Editorial extensions
If this is right
- For any triangle, the shortest escape path can be written as a minimum of arc length under a scalar support constraint, eliminating the permutation variables of the earlier TSPN formulation.
- The same formulation covers closed curves and closed polygonal chains by simply adding the closing segment to the objective.
- Solving the discretized problem yields escape paths for arbitrary non-isosceles triangles; the paper states these are the first such results, while matching known isosceles cases.
- The finite polygonal and angular-collocation problems converge to the continuous optimum as the segment count and grid resolution grow, with convergence proved in Theorems 6 and 8.
- Through the forest-worm duality, the same inequality yields an upper bound on the area of a triangle covering all unit arcs, namely $\tfrac{1}{2}L^2(\cot\alpha+\cot\beta)$ for optimal escape length $L$.
- The optimizer can be extended to arbitrary convex polygons by replacing the three side normals with a weighted combination of all polygon normals, yielding formulas for universal polygon covering curves and polygonal chains.
Reading between the lines
- Because the underlying equivalence with Bellman's problem is inherited from earlier papers and not re-proved here, the strongest test of the paper's contribution is a direct numerical or formal check of Theorem 4 for a few non-isosceles triangles against the original grid formulation.
- The polygon formulas in Section 2.5 are stated with a proof described only as very similar; convergence for arbitrary polygons is therefore a plausible extension rather than an established result of this paper.
- The same phase-shifted support certificate could in principle certify global optimality for the worm problem at every angle, not just the isosceles and 30-60-90 cases singled out in the numerical band; the paper hints at this but does not compute the certificates.
- A natural next step would be to run the same support-function optimization at high precision for individual angle pairs, producing bounds that tighten the known universal-cover upper bound.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper proposes a support-function formulation for Bellman's lost-in-a-forest problem in a triangular forest and for the dual Moser worm problem of covering unit arcs by a triangle. The path is kept fixed at the origin while the triangle translates and rotates; the paper derives a scalar weighted support inequality (Theorem 4) claimed to be exactly equivalent to robust escape for all starting positions and orientations. It further claims existence of optimal continuous escape paths (Theorem 5), convergence of polygonal and collocation discretizations (Theorems 6 and 8), and an extension to arbitrary convex polygons (Section 2.5). Numerical results for triangles with various base angles are reported in Figures 2-4, together with a formula for a Moser-worm upper bound in Eq (22). A Lean formalization appendix is included.
Significance. If Theorem 4 is correct, it is a valuable reduction: the infinite family of escape constraints for all translations and rotations is collapsed into one scalar inequality per orientation, enabling a computational approach to arbitrary triangular forests and triangles as worm covers. The paper also reports what appear to be the first numerical escape paths for non-isosceles triangles, and it attempts to support the main theorems with machine-checked Lean proofs, which is commendable. However, the existence proof for the continuous optimum is incomplete, the numerical formulations do not exactly match the theorem's support function, and the worm upper-bound formula is unproved. These gaps currently prevent full confidence in the paper's central claims, although the algebraic core of Theorem 4 appears sound and repairable.
major comments (4)
- [§2.4, Theorem 5] The proof of Theorem 5 does not establish existence of a minimizer. Non-emptiness of the feasible set and boundedness of length via Finch's diameter bound do not imply that the infimum is attained; an explicit compactness argument is needed. The statement 'the proofs provide the compactness and liminf inequality' refers to an argument that is not supplied in the main text, and the appendix's abstract lemma compact_subsequence_is_optimal only lists hypotheses without verifying that the feasible set of escape paths is compact or that the length functional is lower semicontinuous on that set. Since Theorems 6 and 8 formulate convergence to an optimal continuous path, this gap is load-bearing and needs a concrete Arzelà-Ascoli step, including arclength reparameterization, closedness of the support constraints under uniform convergence, and lower semicontinuity of total variation.
- [§2.2, Eqs (5), (8), (9); §2.4, Eq (12)] The support function used in the numerical formulations is not the same as the support function in Theorem 4. Equation (5) defines h over r([0,2π]), while Theorem 4 defines h over eΓr, which includes the initial segment from the origin to r(0). The discretized problems in Eq (9) and Eq (12) take maxima over the discretized vertices only, do not explicitly include the origin as a point in the max, and Eq (12) restricts the max to 1≤i≤K even though Theorem 6 defines hK with q0=0. Unless r(0)=0 is intended, the solved optimization problems have a different feasible set and objective from the characterized problem, so the numerical lengths in Figures 2-4 are not certified to be escape-path lengths for the original problem. This inconsistency must be resolved, for example by defining the discrete support as max(0, max_i ...) and including the origin in the supporting set.
- [§3.1, Eq (22)] The worm upper-bound formula 1/2 L^2 (1/tan α + 1/tan β) is introduced without proof or derivation. It is not a standard quoted result in the references, and it is the entire basis for the numerical Moser-worm cover areas in Figure 3. The paper should either prove the formula from the cited forest-worm duality or clearly label the worm areas as heuristic estimates rather than established upper bounds.
- [Appendix] The Lean code in the appendix is not readable as supplied: identifiers, operators, and binders are replaced by the placeholder glyph '' throughout, including in the statements of traceEscapes3_iff_support and robustEscape3_iff_weightedSupport. Consequently the claimed machine-checked proofs cannot be verified from the manuscript. Please provide a clean, compilable version of the Lean code, especially because the appendix is invoked to fill the compactness gap in Theorem 5.
minor comments (4)
- [§2.2, p.6] There are several typos and infelicities: 'apths' should be 'paths', 'efficiently' should be 'efficiently', and the phrase 'the topology of r(p) is strictly defined as open curve' is unclear.
- [§2.5, Eqs (20)-(21)] The claimed extension to arbitrary convex polygons is stated without proof. The phrase 'proof ... is very similar' is not sufficient, because the weighted-simplex surjectivity and the existence of the weights λj require assumptions that are not stated for general m-gons.
- [§3.1] The numerical results are reported with no data tables, no certificates of global optimality, and no verification that the reported solutions satisfy the continuum support constraints. The paper itself acknowledges that 'numerical global optimality requires a certified global solver', but Figures 3 and 4 are nevertheless presented as quantitative results; this should be made conditional.
- [References] Reference [16] has a duplicated year '(2026). (2026).', and some arXiv identifiers in the references are inconsistently formatted.
Circularity Check
No significant circularity: the weighted support inequality is derived from triangle geometry and proved self-containedly; the self-citations are scaffolding, and the compactness omission is a gap, not a circular step.
full rationale
The central inequality Eq (7) is obtained from the triangle geometry through the edge-length equilibrium identity L1n1 + L2n2 + L3n3 = 0 and the weighted slack identity L1d1(s) + L2d2(s) + L3d3(s) = sin(alpha) sin(beta) / sin(alpha + beta), neither of which is fitted or presupposes the conclusion. Theorem 4 proves both directions: if the weighted support inequality holds, having Hj < dj for all three coordinates contradicts the weighted slack identity; if it fails, the proof constructs explicit distances dj = Hj + delta lying on the distance simplex, realized by a starting point s, so the path cannot escape. This algebra is self-contained and is additionally formalized in the Lean appendix, including the distance-simplex surjectivity lemma. No parameter is fitted and then renamed as a prediction; the numerical results are solutions of the stated optimization problems rather than forecasts from fitted constants. The citations to the author's prior papers [14] and [15] justify the path-fixing and TSPN reformulation and reference compactness and lower-semicontinuity arguments, but the support-function characterization itself is re-derived in this paper, so these citations are not circular inputs. The proof of Theorem 5 does not actually supply the Arzelà-Ascoli compactness step needed for existence, and Theorem 6 refers to the same argument rather than demonstrating it; this is a correctness gap, not circularity. Overall, no load-bearing step reduces to its own inputs, and the central derivation chain is self-contained against the robust escape condition it claims to characterize.
Assumptions & free parameters
assumptions (5)
- domain assumption TSPN transformation equivalence, Theorem 1 of [14] and [15]: the escape path problem can be reformulated by fixing the path and translating/rotating the triangle.
- domain assumption Duality between Bellman's forest problem and Moser's worm problem, Theorem 3 of [1].
- domain assumption Finch's theorem that the diameter of a closed convex shape is an escape path.
- standard math Arzela-Ascoli compactness and lower semicontinuity of path length under uniform convergence.
- ad hoc to paper A weighted support certificate analogous to Eq (7) holds for any convex polygon with m sides, as stated in Eq (20).
Cite this review
Pith. "Pith review of Universal Triangle Covering Curve and Polygonal Chain: Escaping Forest and Fitting Worm." pith.science (2026). https://pith.science/paper/45VS3JNP
@misc{pith2026260801393,
author = {Pith},
title = {Pith review of: Universal Triangle Covering Curve and Polygonal Chain: Escaping Forest and Fitting Worm},
year = {2026},
howpublished = {\url{https://pith.science/paper/45VS3JNP}},
note = {Machine review of arXiv:2608.01393}
}
read the original abstract
In this paper, we present a general formulation to address the problems of covering curves and polygonal chains with triangle, and fitting these curves into triangle. These problems can be formulated as special cases of Bellman's lost-in-a-forest problem (escaping triangular forest) and Moser's worm problem (covered by triangle). We model and reformulate the problem by keeping the curve stationary while allowing the triangle to translate and rotate. Subsequently, we derive the functional minimization formulation with support function constraints to solve. We also prove the equivalence and convergence of the formulas. Finally, we employ numerical methods and present results for covering curves with arbitrary triangles of various angles. We also present some corollaries and variant results, including closed curves and closed polygonal chains.
Figures
Figures from the paper (8 more)
Reference graph
Works this paper leans on
-
[1]
Finch, S. R., & Wetzel, J. E. (2004). Lost in a forest. The American Mathematical Monthly, 111(8), 645-654
work page 2004
-
[2]
Norwood, R., Poole, G., & Laidacker, M. (1992). The worm problem of Leo Moser. Discrete & Computational Geometry, 7(2), 153-162
work page 1992
-
[3]
Brass, P., Moser, W. O., & Pach, J. (2005). Research problems in discrete geom- etry (Vol. 18). New York: Springer
work page 2005
-
[4]
Croft, H. T., Falconer, K., & Guy, R. K. (2012). Unsolved problems in geometry: unsolved problems in intuitive mathematics. Springer Science & Business Media
work page 2012
-
[5]
Gross, O. A. (1955). A search problem due to Bellman
1955
-
[6]
Gerriets, J., & Poole, G. (1974). Convex regions which cover arcs of constant length. The American Mathematical Monthly, 81(1), 36-41
1974
-
[7]
Isbell, J. R. (1957). An optimal search pattern. Naval Research Logistics Quar- terly, 4(4), 357-359
1957
-
[8]
Zalgaller, V. A. (2005). A question of Bellman. Journal of Mathematical Sciences, 131(1), 5286-5306
2005
Show all 47 references
-
[9]
Besicovitch, A. S. (1965). On arcs that cannot be covered by an open equilateral triangle of side 1. The Mathematical Gazette, 49(369), 286-288
1965
-
[10]
Coulton, P., Movshovich, Y. (2006). Besicovitch triangles cover unit arcs. Ge- ometriae Dedicata, 123(1), 79-88. 23 10◦ − 10◦ 10◦ − 10◦ 10◦ − 10◦ 10◦ − 20◦ 10◦ − 20◦ 10◦ − 20◦ 10◦ − 30◦ 10◦ − 30◦ 10◦ − 30◦ 10◦ − 40◦ 10◦ − 40◦ 10◦ − 40◦ 10◦ − 50◦ 10◦ − 50◦ 10◦ − 50◦ 10◦ − 60◦ 1...
2006
-
[11]
Temerev, A., & Doria, A. (2026). The exact solution of Bellman’s lost-in-a-forest problem for the golden gnomon. arXiv preprint arXiv:2607.24483
2026 arXiv
-
[12]
Gibbs, P. E. (2016). Lost in an isosceles triangle. Working paper
2016
-
[13]
Gibbs, P. (2016). Bellman’s Escape Problem for Convex Polygons
2016
-
[14]
Deng, Z. (2024). A General Solution to Bellman’s Lost-in-a-forest Problem. arXiv preprint arXiv:2412.10686
2024 arXiv
-
[15]
Deng, Z. (2026). Proof and More Variations of Bellman’s Lost-in-a-forest Prob- lem. arXiv preprint arXiv:2606.13987
2026 arXiv
-
[16]
Deng, Z. (2026). (2026). Revisit escape path for infinite unit strip forest and unit broadworm. arXiv preprint arXiv:2607.18563
2026 arXiv
-
[17]
Wetzel, J. E. (2003). Fits and covers. Mathematics magazine, 76(5), 349-363
2003
-
[18]
O., & Pach, J
Brass, P., Moser, W. O., & Pach, J. (2005). Research problems in discrete geometry (Vol. 18). New York: Springer
2005
-
[19]
Poole, G., Gerriets, J. (1973). Minimum covers for arcs of constant length. Bulletin of the American Mathematical Society, 79(2), 462-463
1973
-
[20]
Adhikari, A., & Pitman, J. (1989). The shortest planar arc of width 1. The American Mathematical Monthly, 96(4), 309-327
1989
-
[21]
Norwood, R., Poole, G., Laidacker, M. (1992). The worm problem of Leo Moser. Discrete & Computational Geometry, 7, 153-162
1992
-
[22]
Norwood, Poole. (2003). An improved upper bound for Leo Moser’s worm prob- lem. Discrete & Computational Geometry, 29, 409-417
2003
-
[23]
A., Poole, G
Johnson, J. A., Poole, G. D., Wetzel, J. E. (2004). A small cover for convex unit arcs. Discrete & Computational Geometry, 32, 141-147
2004
-
[24]
Wang, Wei (2006), An improved upper bound for the worm problem, Acta Mathematica Sinica, 49 (4): 835–846
2006
-
[25]
Wetzel, J. E. (2013). Bounds for covers of unit arcs. Geombinatorics, 22(3), 116-122. 28
2013
-
[26]
Khandhawit, T., Pagonakis, D., & Sriswasdi, S. (2013). Lower bound for convex hull area and universal cover problems. International Journal of Computational Geometry & Applications, 23(03), 197-212
2013
-
[27]
Movshovich, Y., Wetzel, J. E. (2017). Drapeable unit arcs fit in the unit 30° sector. Advances in Geometry, 17(4), 497-506
2017
-
[28]
E., Wichiramala, W
Wetzel, J. E., Wichiramala, W. (2019). Sectorial covers for unit arcs. Mathe- matics Magazine, 92(1), 42-46
2019
-
[29]
Movshovich, Y. (2025). Recent advances in the worm problem. European Jour- nal of Mathematics, 11(4), 71
2025
-
[30]
Wichiramala, W., & Panraksa, C. (2026). Wetzel’s 30-60-90 Triangle Covers Unit Arcs. arXiv preprint arXiv:2606.14625
2026
-
[31]
Ball, S., & Lavrauw, M. (2019). Arcs in finite projective spaces. EMS Surv. Math. Sci, 6(1-2), 133-172
2019
-
[32]
E., & Wichiramala, W
Sroysang, B., Wetzel, J. E., & Wichiramala, W. (2008). Covers for angleworms. The American Mathematical Monthly, 115(1), 61-65
2008
-
[33]
Füredi, Z., & Wetzel, J. (2011). Covers for closed curves of length two. Periodica Mathematica Hungarica, 63(1), 1-17
2011
-
[34]
Panraksa, C., & Wichiramala, W. (2021). Wetzel’s sector covers unit arcs. Pe- riodica Mathematica Hungarica, 82(2), 213-222. 4 Appendix-Formalized proofs in Lean The appendix provides formalized proofs of Theorems 4-8 in Lean 4 code. The following is the Lean 4 code for Theore...
2021
-
[35]
a trace/support lemma
-
[36]
a weighted-simplex separation lemma
-
[37]
the exact slack-coordinate description of the normalized triangle
-
[38]
-/ section TraceSupport variable {X : Type*} /-- A number `H` is an attained support value of `p` on Γ``
the all-starting-points and all-angles support theorem. -/ section TraceSupport variable {X : Type*} /-- A number `H` is an attained support value of `p` on Γ``. -/ def IsAttainedSupport Γ( : Set X) (p : X → R) (H : R) : Prop := ( x Γ, p x H) x Γ, p x = H /-- Strict...
-
[39]
the scaled equilibrium identity for the three triangle normals
-
[40]
the weighted side-slack identity
-
[41]
the weighted certificate for all nonnegative slack triples
-
[42]
the intermediate-value boundary-crossing lemma
-
[43]
stability of an attained support maximum
-
[44]
the weighted angular Lipschitz estimate
-
[45]
the collocation-error estimate
-
[46]
repair of an approximate support constraint by dilation
-
[47]
fixed-complexity and joint convergence by quantitative squeezing. -/ noncomputable section open Filter Set namespace TriangleCovering abbrev Vec2 := R × R def dot (u v : Vec2) : R := u.1 * v.1 + u.2 * v.2 def n ( : R) : Vec2 := (-Real.sin , Real.cos ) def n ( : R) : Vec2...
Reviewed August 7, 2026 · model on record in the stance chip above.
Discussion (0). Continue with ORCID to comment.