OctoInformBeta

OpenAI preprint, September 2026 · §4. Geodesic passages and the index form

Lemma 4.3 [Passage classification, finite-passage content]: Reflection completes a turning segment, while unit-energy returns have symmetric height and a controlled incidence speed

Finite neck passages meet the boundary transversely.

Hover over a blue or green term in the text, or a node in the graph, to find it in the other. Select it to read its declaration both in Lean and in words.

The Setting

Consider a geodesic passing through a neck, with geodesic time ss, neck height t(s)t(s), and height velocity v(s)v(s). A finite segment can reach a boundary or have an interior turn where its height velocity vanishes.

The transverse coordinate plane is ℝ2\mathbb R^2, and transverse matrices are real 2×22\times2 matrices. Write p=(p1,p2)𝖳p=(p_1,p_2)^{\mathsf T} for the conserved transverse momentum. For a transverse matrix BB and vectors u,w∈ℝ2u,w\in\mathbb R^2, the matrix bilinear pairing ℬ\mathcal B and the associated quadratic form qBq_B are defined by (1). ℬ(B,u,w)=u𝖳Bw,qB(p)=ℬ(B,p,p).\mathcal B(B,u,w)=u^{\mathsf T}Bw, \qquad q_B(p)=\mathcal B(B,p,p).(1)

The one-sided cutoff a:ℝ→ℝa:\mathbb R\to\mathbb R is the function in (2).✦ The integrated cutoff A:ℝ→ℝA:\mathbb R\to\mathbb R of aa is the function defined by the integral in (2). a(u)={e−1/(2/3−u),u<2/3,0,u≥2/3,A(u)=∫0ua(y)dy.a(u)= \begin{cases} e^{-1/(2/3-u)},&u<2/3,\\ 0,&u\ge2/3, \end{cases} \qquad A(u)=\int_0^u a(y)\,dy.(2)

Let ϑ\vartheta denote the standard smooth transition from zero to one. The shear transition χ:ℝ→ℝ\chi:\mathbb R\to\mathbb R and cometric shear σ\sigma at neck parameter δ\delta are the functions in (3). χ(t)=ϑ(6(t−5/12)),σ(δ,t)=δ2χ(t).\chi(t)=\vartheta\bigl(6(t-5/12)\bigr), \qquad \sigma(\delta,t)=\delta^2\chi(t).(3)

The diagonal coefficients d1d_1 and d2d_2 are the real-valued functions defined by (4). d1(δ,t)=1+δA(t),d2(δ,t)=1+δA(1−t).d_1(\delta,t)=1+\delta A(t), \qquad d_2(\delta,t)=1+\delta A(1-t).(4) The cometric P(δ,t)P(\delta,t) is the real matrix in (5), where the coefficients and shear are evaluated at (δ,t)(\delta,t). P(δ,t)=(d1+d2σ2d2σd2σd2).P(\delta,t)= \begin{pmatrix} d_1+d_2\sigma^2&d_2\sigma\\ d_2\sigma&d_2 \end{pmatrix}.(5)✦

Let τ\tau denote inward height from the terminal boundary. The endpoint frame matrix C2(δ)C_2(\delta) and terminal cometric Pend(δ,τ)P_{\mathrm{end}}(\delta,\tau) are the real matrices defined in (6). C2(δ)=(011−δ2),Pend(δ,τ)=C2(δ)P(δ,1−τ)C2(δ)𝖳.C_2(\delta)=\begin{pmatrix}0&1\\1&-\delta^2\end{pmatrix}, \qquad P_{\mathrm{end}}(\delta,\tau) =C_2(\delta)P(\delta,1-\tau)C_2(\delta)^{\mathsf T}.(6) For an endpoint choice r∈{left,right}r\in\{\mathrm{left},\mathrm{right}\}, the coordinate cometric Pr(δ,τ)P_r(\delta,\tau) is the matrix-valued function of inward height in (7). Pr(δ,τ)={P(δ,τ),r=left,Pend(δ,τ),r=right.P_r(\delta,\tau)= \begin{cases} P(\delta,\tau),&r=\mathrm{left},\\ P_{\mathrm{end}}(\delta,\tau),&r=\mathrm{right}. \end{cases}(7)✦

Reduced Motion and Reflection

For fixed rr, δ\delta, and pp, the momentum potential VpV_p and reduced force FpF_p are the scalar functions of inward height defined by (8); primes on PrP_r denote differentiation in that height. Vp(τ)=ℬ(Pr(δ,τ),p,p),Fp(τ)=ℬ(Pr′(δ,τ),p,p).V_p(\tau)=\mathcal B(P_r(\delta,\tau),p,p), \qquad F_p(\tau)=\mathcal B(P_r'(\delta,\tau),p,p).(8) Reduced neck motion on an interval KK consists of real functions t,v:ℝ→ℝt,v:\mathbb R\to\mathbb R satisfying (9) at every s∈Ks\in K; here t(s)t(s) is measured inward from the chosen endpoint. t′(s)=v(s),v′(s)=−12Fp(t(s)).t'(s)=v(s),\qquad v'(s)=-\tfrac12 F_p(t(s)).(9) Reduced neck motion t,vt,v on an interval KK has unit energy when (10) holds on KK. Vp(t(s))+v(s)2=1(s∈K).V_p(t(s))+v(s)^2=1\qquad(s\in K).(10)

The coordinate cometric PrP_r satisfies the cometric concavity condition when the quadratic-form inequality (11) holds. q−Pr″(δ,u)(q)≥0(u∈[0,1],q∈ℝ2).q_{-P_r''(\delta,u)}(q)\ge0 \qquad\bigl(u\in[0,1],\ q\in\mathbb R^2\bigr).(11) The turning constant a0a_0 is the real number in (12). a0=e−3/2.a_0=e^{-3/2}.(12)✦

For a real function hh and a time cc, its even reflection Rc+hR_c^+h and odd reflection Rc−hR_c^-h are the functions defined by (13). (Rc+h)(s)={h(s),s≤c,h(2c−s),s>c,(Rc−h)(s)={h(s),s≤c,−h(2c−s),s>c.(R_c^+h)(s)= \begin{cases}h(s),&s\le c,\\h(2c-s),&s>c,\end{cases} \qquad (R_c^-h)(s)= \begin{cases}h(s),&s\le c,\\-h(2c-s),&s>c.\end{cases}(13)

The Claim

Theorem 1 (Finite neck passages).

  1. Let δ≥0\delta\ge0 and let t,vt,v be reduced neck motion on [α,β][\alpha,\beta], where α<β\alpha<\beta. If t(α)=0t(\alpha)=0 and t(s)>0t(s)>0 for α<s<β\alpha<s<\beta, lower entry satisfies (14). v(α)>0.v(\alpha)>0.(14)✦ If t(α)=1t(\alpha)=1 and t(s)<1t(s)<1 for α<s<β\alpha<s<\beta, upper entry satisfies (15). v(α)<0.v(\alpha)<0.(15)✦ If 0<t(s)<10<t(s)<1 for α<s<β\alpha<s<\beta and t(β)∈{0,1}t(\beta)\in\{0,1\}, the exit satisfies (16). (t(β)=0andv(β)<0)or(t(β)=1andv(β)>0).\bigl(t(\beta)=0\ \text{and}\ v(\beta)<0\bigr) \quad\text{or}\quad \bigl(t(\beta)=1\ \text{and}\ v(\beta)>0\bigr).(16)✦

  2. Let f:ℝ→ℝf:\mathbb R\to\mathbb R be smooth, and suppose t′=vt'=v and v′=−f(t)/2v'=-f(t)/2 on [α,β][\alpha,\beta]. Assume c∈(α,β)c\in(\alpha,\beta), v(c)=0v(c)=0, t(α)=0t(\alpha)=0, and 0<t(s)<10<t(s)<1 for α<s<β\alpha<s<\beta. The reflected functions satisfy (17). β≤2c−α,Rc+t=t,Rc−v=von [α,β],(Rc+t)(α)=(Rc+t)(2c−α)=0,0<(Rc+t)(s)<1(α<s<2c−α).\begin{gathered} \beta\le2c-\alpha,\qquad R_c^+t=t,\quad R_c^-v=v\quad\text{on }[\alpha,\beta],\\ (R_c^+t)(\alpha)=(R_c^+t)(2c-\alpha)=0,\\ 0<(R_c^+t)(s)<1\qquad(\alpha<s<2c-\alpha). \end{gathered}(17)✦

  3. Let L>0L>0 and suppose the cometric concavity condition (11) holds. Let t,vt,v be reduced neck motion of unit energy on [0,L][0,L], with t(0)=t(L)=0t(0)=t(L)=0 and 0<t(s)<10<t(s)<1 for 0<s<L0<s<L. Set T=t(L/2)T=t(L/2). The return satisfies (18). v(0)>0,v(L/2)=0,0<T<1,t([0,L])⊆[0,T],t(L−s)=t(s),v(L−s)=−v(s)(0≤s≤L),Fp(u)>0(0≤u≤T),v is convex and t is concave on [0,L/2],∫0Lv(s)2ds≤Lv(0)23.\begin{gathered} v(0)>0,\qquad v(L/2)=0,\qquad 0<T<1,\qquad t([0,L])\subseteq[0,T],\\ t(L-s)=t(s),\qquad v(L-s)=-v(s)\qquad(0\le s\le L),\\ F_p(u)>0\qquad(0\le u\le T),\\ v\text{ is convex and }t\text{ is concave on }[0,L/2],\\ \int_0^L v(s)^2\,ds\le\frac{L\,v(0)^2}{3}. \end{gathered}(18)✦

  4. Let δ≥0\delta\ge0 and suppose (11) holds. Let t,v:ℝ→ℝt,v:\mathbb R\to\mathbb R satisfy t(0)=0t(0)=0, v(L/2)=0v(L/2)=0, and the unit energy identity (10) at s=0s=0 and s=L/2s=L/2. Assume T=t(L/2)∈(0,1)T=t(L/2)\in(0,1) and Fp(u)>0F_p(u)>0 for u∈[0,T]u\in[0,T]. The incidence estimate is (19). v(0)2≤δa0p12.v(0)^2\le\delta a_0p_1^2.(19)✦

In the paper, the result reads: with κ\kappa denoting the paper’s force bound, every finite endpoint of a neck passage JJ is transverse; exactly one alternative holds, either J=ℝJ=\mathbb R and tt is constant, or vv never vanishes and JJ is a finite crossing between different boundaries or a half-infinite approach to an interior equilibrium, or, in inward coordinates from one boundary and after translating time, J¯=[0,L]\overline J=[0,L], t(0)=t(L)=0t(0)=t(L)=0, v(0)=−v(L)=v0>0v(0)=-v(L)=v_0>0, and there is a unique turn at time L/2L/2 of height t*=t(L/2)∈(0,1)t_*=t(L/2)\in(0,1); for a return, t(L−s)=t(s)t(L-s)=t(s), v(L−s)=−v(s)v(L-s)=-v(s), 0<Vp′(u)≤κp120<V_p'(u)\le\kappa p_1^2 on [0,t*][0,t_*], p1≠0p_1\ne0, and 0<v02=Vp(t*)−Vp(0)≤κp120<v_0^2=V_p(t_*)-V_p(0)\le\kappa p_1^2; an infinite-end equilibrium has Vp=1V_p=1 and Vp′=0V_p'=0; a boundary-tangent geodesic does not enter the open neck at that time.

What Is Checked

The statements concern the real scalar neck equations for either endpoint coordinate cometric, unit-energy returns under cometric concavity, reflection for an arbitrary smooth scalar force law, and the boundary and midpoint energy identities with positive force on the intervening height interval for the incidence estimate.