For a graph H, \langle H \rangle denotes the class of all subdivisions of H and tw(H) denotes the treewidth of H. In this paper, we prove the following. For k\geq 1, R\geq 1, let G,H be two graphs such that strong isometric path complexities (Chakraborty et al. [Disc. Math., 2026]) of both G and \langle H \rangle are at most k, and G admits an honest, “nicely rooted” R-bounded H-partition. Then, there is a graph F with tw(F)\leq tw(H) such that G admits a (1,33\cdot R\cdot k^2)-quasi-isometry to F. Using results of Albrechtsen, Distel, and Georgakopoulos (2025), we also obtain that K_{2,t}-asymptotic minor-free graphs admit quasi-isometries with additive distortion to K_{2,t}-minor-free graphs. This answers an open question raised by the above authors. As part of our proof, we combine the graph-partition based method and the layering partition based method (Chepoi et al. [Discrete Comput. Geom. 2012]) to obtain additive quasi-isometry when both the source and all subdivisions of the target graph have bounded strong isometric path complexity.
Let X be a graph, and let H be a finite graph. Then X has no K-fat H minor for some K \in \mathbb{N} if and only if X is f(K)-quasi-isometric to a graph with no H-minor, where f : \mathbb{N} \rightarrow \mathbb{N} depends on H only.
For which finite graphs H it is true that if a graph X has no K-fat H-minor for some K \in \mathbb{N}, then X admits a map of bounded additive distortion onto a graph with no H minor?
Is it true that if a graph X has no K-fat K_{2,t}-minor for some K \in \mathbb{N}, t\geq 3, then X admits a map of bounded additive distortion onto a graph with no K_{2,t}-minor?
Let H, G be graphs, and let \mathcal{H} be an honest, R-bounded H-partition of G for some R \in \mathbb{R}. Then G is (R + 1, R/(R + 1))-quasi-isometric to H.
For k\geq 1, R\geq 1, let G,H be two graphs such that sipco\left(G\right)\leq k, sipco\left(\langle H \rangle\right)\leq k, and G admits an honest, nicely rooted R-bounded H-partition. There is a graph F with tw(F) \leq tw(H) and G admits a (1,33\cdot R \cdot k^2)-quasi-isometry to F.
There is a function g:\mathbb{N}\times \mathbb{N}\rightarrow \mathbb{N} such that for any graph G without any K-fat K_{2,t}-minor for t\geq 3, there is a K_{2,t}-minor-free graph to which G admits a (1,g(K,t))-quasi-isometry.
A graph class \cal G contains H as an asymptotic minor if for every integer K\geq 1, there exists a graph G\in \cal G, such that G contains H as a K-fat minor.
An (M,A)-quasi-isometry between graphs G and H is a map f\colon V(G)\rightarrow V(H) such that the following holds for fixed constants M\geq 1, A\geq 0,
M^{-1}d(x,y)-A \leq d(f(x),f(y)) \leq Md(x,y)+A for every x,y\in V(G); and
for every z\in V(H) there is x\in V(G) such that d(z,f(x))\leq A.
The graphs G and H are quasi-isometric if there exists a quasi-isometric map f between V(G) and V(H).
A class \mathcal{G} of graphs admits a quasi-isometry with additive distortion to a class \mathcal{H} if there exists a constant A (depending only on \mathcal{G} and \mathcal{H}) such that any G\in \mathcal{G} admits a (1,A)-quasi-isometry to a graph in \mathcal{H}.
Let H, G be graphs, and let \mathcal{H} be an honest, R-bounded H-partition of G for some R\in \mathbb{R}. Then G is (R + 1, R/(R+1))-quasi-isometric to H.
For an integer R, let \mathcal{H}= (H, (V_h)_{h \in V(H)}) be an R-bounded honest graph partition of a graph G over a graph H which is also rooted at a node s. We say \mathcal{H} is “nicely rooted” if
for all j\in \mathbb{N}, the layer L^j is an independent set,
there exists a vertex r\in V_s such that V_s=B_G(r,R_s) with R_s\le R, and for every node h\neq s, V_h is level,
the heights of two nodes x,y\in V(H) lying in the same cluster w.r.t s are the same.
For an integer R, let \mathcal{H}= (H, (V_h)_{h \in V(H)}) be an R-bounded nicely rooted honest graph-partition of a graph G over a graph H which is also rooted at a node s. Let r\in V_s be a vertex such that V_s=B_G(r,R_s). Let g and h be two nodes of H that lie in the same cluster of H w.r.t s. For any w\in \delta^{\downarrow}_{h}, x\in \delta^{\downarrow}_{g}, d_{G}\left(r,w\right) = d_{G}\left(r,x\right). Moreover, for any y\in \delta^{\uparrow}_{h}, z\in \delta^{\uparrow}_{g}, d_{G}\left(r,y\right) = d_{G}\left(r,z\right).
We prove the lemma by induction on the layer that contains g and h. Assume g,h\in L^1. For two vertices w\in \delta^{\downarrow}_{g}, x\in \delta^{\downarrow}_{h}, there exist w',x'\in V_s such that \{ww',xx'\}\subseteq E(G). By Definition 11?, d_{G}\left(r,w'\right) = d_{G}\left(r,x'\right) =R_s. Hence, d_{G}\left(r,w\right) = d_{G}\left(r,w'\right) +1 = R_s+1 and d_{G}\left(r,x\right) = d_{G}\left(r,x'\right) +1 = R_s+1. Hence, d_{G}\left(r,w\right) = d_{G}\left(r,x\right). Suppose y\in \delta^{\uparrow}_{g},z\in \delta^{\uparrow}_{h}. As any (r,y)-isometric path contains a vertex of \delta^{\downarrow}_{g} and by the above arguments, all vertices of \delta^{\downarrow}_{g} are equidistant from r in G, it follows that d_{G}\left(r,y\right) = d_{G}\left(r,\delta^{\downarrow}_{g}\right) + R_g. Similarly, d_{G}\left(r,z\right) = d_{G}\left(r,\delta^{\downarrow}_{h}\right) + R_h. Since g and h lie in the same cluster w.r.t s, R_h=R_g (due to Definition 11?). By the above arguments d_{G}\left(r,\delta^{\downarrow}_{g}\right) = d_{G}\left(r,\delta^{\downarrow}_{h}\right) = R_s+1. Hence, d_{G}\left(r,y\right) = d_{G}\left(r,z\right) = R_s+1+R_h.
For an integer k\geq 1, assume the statement is true for all pairs of nodes that lie in the same cluster w.r.t s and lie in L^i with i\leq k. Now consider two nodes g and h that lie in the same cluster w.r.t s and g,h\in L^{k+1}. Let w\in \delta^{\downarrow}_{g},x\in \delta^{\downarrow}_{h} and let g',h'\in L^k be nodes such that there exist w'\in V_{g'}, x'\in V_{h'} with ww'\in E(G) and xx'\in E(G). Observe that, g' and h' lie in the same cluster w.r.t s and w'\in \delta^{\uparrow}_{g'}, x'\in \delta^{\uparrow}_{h'}. Hence, d_{G}\left(r,w'\right) =d_{G}\left(r,x'\right). Since d_{G}\left(r,w\right) = d_{G}\left(r,w'\right) +1 and d_{G}\left(r,x\right) = d_{G}\left(r,x'\right) +1, we have that d_{G}\left(r,w\right) =d_{G}\left(r,x\right).
For the second part, consider two vertices y\in \delta^{\uparrow}_{g}, z\in \delta^{\uparrow}_{h}. As any (r,y)-isometric path contains a vertex of \delta^{\downarrow}_{g} and by the above arguments, all vertices of \delta^{\downarrow}_{g} are equidistant from r in G, it follows that d_{G}\left(r,y\right) = d_{G}\left(r,\delta^{\downarrow}_{g}\right) + R_g. Similarly, d_{G}\left(r,z\right) = d_{G}\left(r,\delta^{\downarrow}_{h}\right) + R_h. Since g and h lie in the same cluster w.r.t s, R_h=R_g (due to Definition 11?) and d_{G}\left(r,\delta^{\downarrow}_{g}\right) = d_{G}\left(r,\delta^{\downarrow}_{h}\right).
For an integer R, let \mathcal{H}= (H, (V_h)_{h \in V(H)}) be an R-bounded nicely rooted honest graph-partition of a graph G over a graph H which is also rooted at a node s. Let r\in V_s be a vertex such that V_s=B_G(r,R_s). For gh\in E(H) with \ell_{g}=\ell_{h}-1 and u \in V_g, v\in V_h, d_{G}\left(r,u\right) < d_{G}\left(r,v\right).
Let P be an (r,v)-isometric path in G. Let X=\displaystyle\bigcup\limits_{g'\in L^{\ell_{g}}} \delta^{\uparrow}_{g'} and by definition (of nicely rooted graph partitions), X is an (r,v)-separator in G. Hence, there exists a node g' in L^{\ell_{g}} such that P contains a vertex w_1\in \delta^{\uparrow}_{g'}. Observe that d_{G}\left(r,v\right) > d_{G}\left(r,w_1\right). Moreover, g,g' lie in the same cluster w.r.t s in H. By Lemma 12, for a vertex w_2 \in \delta^{\uparrow}_{g}, d_{G}\left(r,w_1\right) = d_{G}\left(r,w_2\right). Now observe that d_{G}\left(r,u\right) \leq d_{G}\left(r,w_2\right) = d_{G}\left(r,w_1\right) < d_{G}\left(r,v\right).
For an integer R, let \mathcal{H}= (H, (V_h)_{h \in V(H)}) be an R-bounded honest graph partition of a graph G over a graph H. For u,v\in V(G) and a (u,v)-isometric path P in G, trace_{H}\left(P\right) is a sequence \left(g_1,g_2,\ldots,g_k\right) of distinct elements such that for i\in [k], g_i\in V(H), u\in V_{g_1}, v\in V_{g_k}, and for i\in [k-1], there is an edge e_i=u_iv_i\in E(G) such that u_i\in V_{g_i}\cap V(P), v_i\in V_{g_{i+1}}\cap V(P).
For an integer R, let \mathcal{H}= (H, (V_h)_{h \in V(H)}) be an R-bounded nicely rooted honest graph-partition of a graph G over a graph H which is also rooted at a node s. Let r\in V_s be a vertex such that V_s=B_G(r,R_s).
For any u\in V(G), h\in V(H), u\in V_h and an (r,u)-isometric path P in G, trace_{H}\left(P\right) is an (s,h)-isometric path in H.
Let (x,y) be an r-rooted pair in G and g,h\in V(H) such that x\in V_g, y\in V_h. Then (g,h) is an s-rooted pair in H.
First we prove ?. Let P be an (r,u)-isometric path in G. We prove that for any node a\in V(H), V_a\cap V(P) induces a subpath of P.
To prove the above claim, suppose there exists a node a\in V(H) such that the graph F induced by V_a\cap V(P) has at least two components. For distinct connected components C,C' of F, let P(C,C') be the minimal subpath of P that has vertices adjacent to both C and C'. Let C_1 and C_2 be two components of F such that none of the vertices of P(C_1,C_2) lie in V_a. Without loss of generality, assume the vertices of C_1 are closer to r in G than those of C_2. In other words, for u\in V(C_1), v\in V(C_2), d_{G}\left(r,u\right) <d_{G}\left(r,v\right). Let Q=P(C_1,C_2). For i\in \{1,2\}, let f_i=u_iv_i\in E(G) such that u_i\in V(C_i) and v_i\in V(Q). Observe that d_{G}\left(r,u_1\right) < d_{G}\left(r,w\right) for all w\in V(Q).
If v_1 lies in a node b\in V(H) where \ell_{b}=\ell_{a}-1, then by Lemma 13, d_{G}\left(r,v_1\right) < d_{G}\left(r,u_1\right), which is a contradiction. Hence, v_1 lies in a node b\in V(H) where \ell_{b}=\ell_{a}+1. Observe that u_1\in \delta^{\uparrow}_{a}. Let w\in V(Q) be the vertex which is closest to u_1 in G and is adjacent to some vertex w_1 where w_1\in V_{c}\setminus \{u_1\}, c\in L^{\ell_{a}}. Since Q is a connected subgraph which has vertices adjacent to both C_1 and C_2, w always exists. Observe that w_1\in \delta^{\uparrow}_{c}. Moreover, a and c lie in the same cluster w.r.t s in H. Hence, by Lemma 13 and Lemma 12, d_{G}\left(r,u_1\right) =d_{G}\left(r,w_1\right), which is a contradiction.
Let trace_{H}\left(P\right)=(a_1,a_2,\ldots,a_k). Since for i\in [k], V_{a_i}\cap V(P) induces a subpath of P, observe that, for j\in [2,k], a_j is the unique node of H that contains the endpoint of the path induced by V(P)\setminus \{V_{a_1}\cup \ldots \cup V_{a_{j-1}}\} which is also adjacent to a vertex of V_{a_{j-1}}. Next, we prove that for j\in [k-1], d_{H}\left(s,a_{j+1}\right) = d_{H}\left(s,a_j\right) +1.
To prove the above claim, suppose for some j\in [k-1], d_{H}\left(s,a_{j+1}\right) \leq d_{H}\left(s,a_j\right). Let w_1w_2\in E(P) be the unique edge such that w_1\in V_{a_j}\cap V(P) and w_2\in V_{a_{j+1}}\cap V(P). Clearly, d_{H}\left(s,a_{j+1}\right) =d_{H}\left(s,a_{j}\right) -1 and d_{G}\left(r,w_1\right) < d_{G}\left(r,w_2\right). But this contradicts Lemma 13. Hence, trace_{H}\left(P\right) is an (s,h)-isometric path.
To prove ?, let P be an (r,y)-isometric path in G that contains x. By ?, Q=trace_{H}\left(P\right) is an (s,h)-isometric path in H. Since x\in V(P), g\in V(Q). Hence (g,h) is an s-rooted pair in H.
The “strong isometric path complexity” of a connected graph G, denoted by sipco\left(G\right), is the minimum integer k such that for every vertex r\in V(G), ipco\left(r,G\right)\leq k.
The strong isometric path complexity of graphs that do not contain K_{2,t} as a K-fat minor is at most (48K+5)t-1.
Let X and Y be two graphs, f\colon V(X)\rightarrow V(Y) be a function, and r\in V(X) be a vertex. Let t\geq 1, k\geq 1 be two integers such that the following hold.
For any edge ab\in E(X), d_{Y}\left(f(a),f(b)\right) \leq k+1;
For any r-rooted pair (a,b) in X, d_{Y}\left(f(a),f(b)\right) \leq d_{X}\left(a,b\right) + t.
Then, for \{u,v\}\subseteq V(X), d_{Y}\left(f(u),f(v)\right) \leq d_{X}\left(u,v\right) + t\cdot sipco\left(X\right) + k(sipco\left(X\right)-1).
Let P be any (u,v)-isometric path in X and \mathcal{Q} be a minimum cardinality set of r-rooted isometric paths in X that covers P and for any Q\in \mathcal{Q}, |V(P)\cap V(Q)| is maximized. Let z\coloneq |\mathcal{Q}| and by definition, z\leq sipco\left(X\right). Observe that for any Q\in \mathcal{Q}, the path induced by V(P)\cap V(Q) is an isometric path in X. Moreover, there exists exactly one path Q_1\in \mathcal{Q} that contains u. Otherwise, if there exists a path Q'\in \mathcal{Q}\setminus \{Q_1\} containing u, then either V(P)\cap V(Q') contains V(P)\cap V(Q_1), or vice versa. But this contradicts the minimality of \mathcal{Q}.
Let Q_1\in \mathcal{Q} be the path containing u, and define P_1\coloneq P[V(P)\cap V(Q_1)]. Let a_1\coloneq u, and b_1 be the end-vertex of P_1 distinct from a_1, unless P_1 is reduced to a single vertex in which case b_1=a_1.
For i\in\{2,\ldots, z\}, let Q_i\in \mathcal{Q} be a path that contains the end-vertex of the path induced by V(P)\setminus \displaystyle\bigcup\limits_{j<i} V(Q_j) which is distinct from v. For i\in\{2,\ldots, z\}, let P_i= P[V(P)\cap V(Q_i)], and a_i be the end-vertex of P_i which is adjacent to b_{i-1} in P, and let b_i be the end-vertex of P_i distinct from a_i. Note that b_z=v. Hence,
This completes the proof.
Let H be a graph such that sipco\left(\langle H\rangle\right)\leq k. Then, for any path P of H and a component C of H-V(P), |N_H(C)\cap V(P)|\leq k.
Assume for contradiction that there is a path P in H and a component C of H-V(P) such that |N_H(C)\cap V(P)|\geq k+1. For each u\in N_H(C)\cap V(P), let r_u be a vertex of C adjacent to u. Let T be a spanning tree of C and r be any vertex of C. Set t_1=|E(P)|, t_2=\max_{u\in V(C)} d_{T}\left(r,u\right), t_3=t_1+t_2+2 and t_4=2t_3+t_1+2. Define a function \phi\colon E(H)\rightarrow \mathbb{Z}^+ as follows.
For each e\in E(T)\cup E(P), define \phi(e)=1.
For each u\in N_H(C)\cap V(P), define \phi(ur_u)=t_3-d_{T}\left(r,r_u\right).
For any edge e that does not satisfy the above two conditions, define \phi(e)=t_4.
Construct a graph H^* by replacing each edge e of H with a path of length \phi(e). Clearly, H^*\in \langle H \rangle. We prove the following claim.
The path P is isometric in H^*.
By definition, P is a path in H^*. Let x,y be the endpoints of P and Q be any (x,y)-path distinct from P. By definition of H^*, Q must contain a subpath of length at least t_1+2 and hence |E(Q)|>|E(P)|.
Any r-rooted isometric path in H^* contains at most one vertex of N_H(C)\cap V(P).
By construction of H^*, for any u\in N_H(C)\cap V(P), d_{H^*}\left(r,u\right) = t_3. (Indeed, for any u\in N_H(C)\cap V(P), there is an (r,u)-path in H^* through r_u that has length t_3 and any other (r,u)-path will have length at least t_4>t_3.) Hence, any path of H^* that has r as an endpoint and contains two vertices of N_H(C)\cap V(P) will have length at least t_3+1, and thus not isometric in H^*.
By Claims ? and ?, ipco\left(r,H^*\right)\geq k+1. This implies sipco\left(H^*\right)\geq k+1, a contradiction.
Treewidth of F is at most that of H and sipco\left(F\right)\leq sipco\left(\langle H\rangle\right)+2.
For every node g and vertex v\in V_g\setminus \{r_g\}, delete from F, all vertices of Q_v\setminus \{r_{p(g)}\}. The resulting graph, say F^*, is a subdivision of H. Since subdivision of edges preserves treewidth, we have that tw(F)\leq tw(H). For the second part, let k=sipco\left(\langle H\rangle \right) and P be any isometric path of F. Let E_1 be the set of edges of P that lie in some pendant path of F. Clearly E_1 induces at most two components C_1,C_2 and E(P)\setminus E_1 induces an isometric path Q of F^*. Let v be any vertex of F.
Suppose v\in V(F^*). Since F^*\in \langle H \rangle, sipco\left(F^*\right)\leq k i.e. vertices of Q can be covered with at most k many v-rooted isometric paths in F^*, and thus in F. Since C_1,C_2 are subpaths of pendant paths, each of them can be covered with one v-rooted isometric path in F. Therefore, P can be covered with k+2 many v-rooted isometric paths.
Suppose v\not\in V(F^*). If v is incident to some edge in C_1 or C_2, then P can be covered with two v-rooted paths. Otherwise, there exists a vertex v'\in F^* such that any isometric path in F containing v as an endpoint also contains v'. Since F^*\in \langle H \rangle, sipco\left(F^*\right)\leq k, similar arguments as above imply that P can be covered with k+2 many v-rooted paths.
The graph F satisfies the following.
For an edge e=gh\in E(H) with \ell_{g}=\ell_{h}-1, d_{G}\left(r_g,r_h\right) \leq 2R+1, the length of P_e is at most 2R+1, and d_{F}\left(r_g,r_h\right) \leq 2R+1.
For a node h\in V(H), and a vertex v\in V_h\setminus \{r_h\}, the length of Q_v is at most 3R+1 and, d_{F}\left(v,r_{p(h)}\right) \leq3R+1.
For any v\in V(G), h\in V(H) with v\in V_{h}, d_{F}\left(v,r_h\right) \leq 5R+2.
For an edge uv\in E(G), d_{F}\left(u,v\right) \leq 12R+5.
To prove ?, assume without loss of generality \ell_{g}=\ell_{h}-1. By definition, r_h is a vertex of \delta^{\downarrow}_{h} and there exists x\in V_g, x'\in \delta^{\downarrow}_{h} such that xx'\in E(G). Since d_{G}\left(x,r_g\right) \leq R and d_{G}\left(x',r_h\right) \leq R, we have d_{G}\left(r,r_h\right) - d_{G}\left(r,r_g\right) \leq d_{G}\left(r_h,r_g\right) \leq 2R+1. Now the statement follows from the definition of P_e and the definition of F.
To prove ?, observe that d_{G}\left(v,r_{p(h)}\right) \leq d_{G}\left(v,r_{h}\right) + d_{G}\left(r_{h},r_{p(h)}\right). By definition, d_{G}\left(v,r_{h}\right) \leq R. Now ? implies d_{G}\left(r,v\right) - d_{G}\left(r,r_{p(h)}\right) \leq d_{G}\left(v,r_{p(h)}\right) \leq 3R+1. Now the statement ? follows from the definition of Q_v, and the definition of F.
To prove ?, observe that d_{F}\left(v,r_h\right) \leq d_{F}\left(v,r_{p(h)}\right) + d_{F}\left(r_h,r_{p(h)}\right). Now ? and ? imply d_{F}\left(v,r_h\right) \leq 5R+2.
To prove ?, consider an edge xy\in E(G). First assume there exists a vertex h\in V(H) such that x,y\in V_h. Observe that d_{F}\left(x,y\right) \leq d_{F}\left(x,r_h\right) + d_{F}\left(r_h,y\right). By ?, d_{F}\left(x,y\right) \leq 10R+4. Now consider the case when there is an edge e=gh\in E(H) with \ell_{g}=\ell_{h}-1 and x\in V_h, y\in V_g. Observe that d_{F}\left(x,y\right) \leq d_{F}\left(x,r_h\right) + d_{F}\left(r_h,r_g\right) + d_{F}\left(r_g,y\right). Now ? and ? imply, d_{F}\left(x,y\right) \leq 12R+5.
For any s-rooted pair (g,h) in H, d_{F}\left(r_g,r_h\right) \leq d_{G}\left(r_g,r_h\right).
If g=h, the lemma is trivially true. So, assume g\neq h and let (g,h) be an s-rooted pair in H and P be any (s,h)-isometric path in H that contains g. Let e_1,e_2,\ldots,e_k be the edges of the (g,h)-subpath of P where e_1 is incident to g, for i\in [k-1], e_i,e_{i+1} share an endpoint and e_k is incident to h. For i\in [k], let e_i=g_ih_i such that d_{H}\left(s,g_i\right) =d_{H}\left(s,h_i\right) -1. Observe that, g=g_1, h=h_k, and i\in [k-1], h_i=g_{i+1}. Thus,
This completes the proof.
There is a function \beta\colon V(G)\rightarrow V(F), such that all of the following hold.
For each x\in V(F), there is a vertex y\in V(G) such that d_{F}\left(x,\beta(y)\right) \leq 3R+1.
For each edge xy\in E(G), d_{F}\left(\beta(x),\beta(y)\right) \leq 12R+5.
For any r-rooted pair (x,y) in G, d_{F}\left(\beta(x),\beta(y)\right) \leq d_{G}\left(x,y\right) + 12R+4.
By definition, any vertex of G is also a vertex of F. For x\in V(G), define \beta(x)=x. Hence for x,y\in V(G), d_{F}\left(\beta(x),\beta(y)\right) =d_{F}\left(x,y\right).
To prove ?, any vertex of F is contained in some path in \mathcal{S}, and that the endpoints of the paths in F are identified with vertices of V(G). Now ? follows from Lemma 23? and ?.
Statement ? follows from Lemma 23?.
To prove ?, consider two vertices x,y\in V(G) such that there is an (r,y)-isometric path in G that contains x. Let g,h\in V(H) be the vertices of H such that x\in V_g and y\in V_h. If g=h, then d_{F}\left(x,y\right) \leq d_{F}\left(x,r_g\right) + d_{F}\left(r_g,y\right). Lemma 23? implies d_{F}\left(x,y\right) \leq 10R+4. Otherwise, g\neq h, and by Lemma 15?, there is a (s,h)-isometric path in H that contains g i.e. (g,h) is an s-rooted pair in H. Therefore,
This completes the proof.
Let (g,h) be an s-rooted pair in H such that (r_g,r_h) is an r-rooted pair in G. Then, d_{G}\left(r_g,r_h\right) = d_{F}\left(r_g,r_h\right).
The case g=h is trivial, and hence assume g\neq h. Let P be an (r_g,r_h)-isometric path in F, and let r_{g_1},r_{g_2},\ldots,r_{g_k} be the ordering of the representative vertices encountered on P when traversed from g_1=g to g_k=h. By the construction of F,
Therefore,
where the last equality holds because (r_g,r_h) is an r-rooted pair in G. Together with Lemma 24, we have the proof.
Let g,h\in V(H) be two nodes such that (r_g,r_h) is an r-rooted pair in F. Then, d_{G}\left(r_g,r_h\right) \leq d_{F}\left(r_g,r_h\right) + 2R\cdot sipco\left(\langle H \rangle\right).
Let k=sipco\left(\langle H\rangle \right). If g=h, then the lemma is trivially true. If g=s, then the lemma is true by Lemma 26. Therefore, we assume s\neq g\neq h. Let P be a (r,r_h)-isometric path in F that contains r_g. Let Q=\left(g_1,g_2,\ldots,g_t\right) be the sequence such that for i\in [t], g_i\in V(H), s=g_1, h= g_t, and for i\in [t-1], P contains a (r_{g_i},r_{g_{i+1}})-subpath that does not contain any vertex of G apart from r_{g_i} and r_{g_{i+1}}. Clearly, Q is an (s,h)-isometric path in H that contains g. Therefore, (g,h) is an s-rooted pair in H. Let Q_1 be the (g,h)-subpath of Q. Applying Lemma 19 on Q_1 and the component C of H-V(Q_1) that contains s, we have that |N_H(C)\cap V(Q_1)|\leq k. Let A=N_H(C)\cap V(Q_1).
For each node h'\in V(Q_1)\setminus \{g\}, there exists a node g'\in A\setminus \{h'\} such that (g',h') is an s-rooted pair in H and d_{G}\left(r_{g'},r_{h'}\right) \leq d_{G}\left(r,r_{h'}\right) - d_{G}\left(r,r_{g'}\right) + 2R.
Fix a node h'\in V(Q_1)\setminus \{g\} and let h_1 be the node of Q_1 where d_{Q_1}\left(g,h_1\right) =d_{Q_1}\left(g,h'\right) -1. Since H is an honest partition of G, there exists an edge uv\in E(G) such that v\in V_{h'} and u\in V_{h_1}. Observe that v\in \delta^{\downarrow}_{h'} and by Lemma 12,
By Lemma 13, d_{G}\left(r,v\right) =d_{G}\left(r,u\right) +1, and therefore there exists an (r,v)-isometric path T in G that contains u. Let Q_2=trace_{H}\left(T\right) and h_2\in V(Q_2)\cap V(Q_1) be the node closest to s. Observe that h_2\neq h', and h_2\in A. Let w be the vertex of V(T)\cap V_{h_2} that is closest to r in G. Observe that w\in \delta^{\downarrow}_{h_2} and by Lemma 12,
By (?), (?) we have that,
Setting g'=h_2 proves the claim.
Let (g'_1, g'_2, \ldots g'_{t'}) be the maximum subsequence of Q_1 such that g'_{t'}=h, and for each i\in [t'-1,1], g'_{i} is the node closest to s that satisfies Claim ? when applied to g'_{i+1}. Observe that, t'-1\leq k. Observe that, g'_1=g. Hence,
This completes the proof.
There is a function \alpha\colon V(F)\rightarrow V(G) satisfying the following.
For each edge xy\in E(F), d_{G}\left(\alpha(x),\alpha(y)\right) \leq 3R+1.
For any r-rooted pair (x,y) in F, d_{G}\left(\alpha(x),\alpha(y)\right) \leq d_{F}\left(x,y\right) + 4R+2+2R\cdot sipco\left(\langle H\rangle\right).
The function \alpha\colon V(F)\rightarrow V(G) is defined as follows. Let x\in V(F) be a vertex. If x is also a vertex of G, then define \alpha(x)=x. Otherwise, if there exist y\in V(G), h\in V(H) such that y\in V_h\setminus \{r_h\}, x\in V(Q_y) where Q_y\in \mathcal{S}, then define \alpha(x)=r_{p(h)}. Otherwise there exists e=gh\in E(H) with \ell_{g}=\ell_{h}-1 and x\in V(P_e) where P_e\in \mathcal{S}. In this case define \alpha(x)=r_g. This concludes the definition of \alpha.
Using \alpha, we define a function \alpha'\colon \mathcal{S}\rightarrow 2^{V(G)} as follows. Let Q\in \mathcal{S} be a path. Assume for some e'=gh\in E(H) with \ell_{g}=\ell_{h}-1, Q=P_{e'}. Define \alpha'(Q)=\{r_g,r_h\}. Now assume for some z\in V(G) and h'\in V(H), Q=Q_z and z\in V_{h'}. In this case, define \alpha'(Q)=\{z,r_{p(h')}\}. By Lemma 23? and ?, for any Q\in \mathcal{S} and \{a,b\}=\alpha'(Q), d_{G}\left(a,b\right) \leq 3R+1.
To prove (a), let e=xy be any edge of F. Observe that there exists a path T_e\in \mathcal{S} such that e\in E(T_e). By definition, \{\alpha(x),\alpha(y)\}\subseteq \alpha'(T_e). The above arguments imply, d_{G}\left(\alpha(x),\alpha(y)\right) \leq 3R+1.
To prove (b), let x,y\in V(F) such that there is an (r,y)-isometric path Q in F that contains x. If x=y, then the lemma is trivially true. For x\ne y, let e_x be the edge immediately after x toward y along Q, and e_y be the edge immediately before y. Let T_x,T_y\in \mathcal{S} be two paths containing e_x and e_y, respectively. If T_x=T_y, then \{\alpha(x),\alpha(y)\}\subseteq \alpha'(T_x) and thus d_{G}\left(\alpha(x),\alpha(y)\right) \leq 3R+1. Assume T_x\neq T_y. Then there exists e_1=g_1h_1\in E(H) with \ell_{g_1}=\ell_{h_1}-1 such that P_{e_1}=T_x. By definition, \alpha(x)\in \{r_{g_1}, r_{h_1}\}.
Case 1. Suppose there exists e_2=g_2h_2\in E(H) with \ell_{g_2}=\ell_{h_2}-1 such that P_{e_2}=T_y. Then \alpha(y)\in \{r_{g_2},r_{h_2}\}. Since T_x\neq T_y, Q contains r_{g_1}, x, r_{h_1}, r_{g_2},y. This implies Q contains \alpha(x),x,\alpha(y),y and d_{F}\left(r,\alpha(x)\right) \leq d_{F}\left(r,x\right) \leq d_{F}\left(r,\alpha(y)\right) \leq d_{F}\left(r,y\right). Thus,
Case 2. Suppose there exists h_2\in V(H), and z\in V_{h_2}\setminus \{r_{h_2}\} such that Q_{z}=T_y. Then \alpha(y)\in \{y,r_{p(h_2)}\}. Since T_x\neq T_y, and Q contains r_{g_1}, x, r_{h_1}, r_{p(h_2)},y and d_{F}\left(r,r_{g_1}\right) \leq d_{F}\left(r,x\right) \leq d_{F}\left(r,r_{h_1}\right) \leq d_{F}\left(r,r_{p(h_2)}\right) \leq d_{F}\left(r,y\right). This implies Q contains \alpha(x),x,\alpha(y),y and d_{F}\left(r,\alpha(x)\right) \leq d_{F}\left(r,x\right) \leq d_{F}\left(r,\alpha(y)\right) \leq d_{F}\left(r,y\right). If \alpha(y)=r_{p(h_2)} then,
Otherwise, \alpha(y)=y and thus,
This completes the proof.
For k\geq 1, R\geq 1, let G,H be two graphs such that sipco\left(G\right)\leq k, sipco\left(\langle H \rangle\right)\leq k, and G admits an honest, nicely rooted R-bounded H-partition. There is a graph F that has H as a core and G admits a (1,33\cdot R \cdot k^2)-quasi-isometry to F.
There is a function f:\mathbb{N}\times \mathbb{N}\rightarrow \mathbb{N} such that for any graph G without any K-fat K_{2,t}-minor, there is an honest, nicely rooted f(K,t)-bounded H-partition of G where H is \Theta_t minor-free.
Apply Lemma 3.4 of [albrechtsen2025excluding] with 3K. Since G contains no K-fat K_{2,t}-minor, Lemma 2.1 of [albrechtsen2025excluding] implies that G contains no 3K-fat \Theta_t-minor. Hence, Lemma 3.4 of [albrechtsen2025excluding] provides an honest, R(t,3K)-bounded H-partition satisfying properties (i)-(iv) of [albrechtsen2025excluding].
We root H at the node s used in the construction of [albrechtsen2025excluding]. Property ? of our Definition 11 is exactly (i) of [albrechtsen2025excluding]. Property ? follows from (ii) of [albrechtsen2025excluding] : the root class is a ball around the distinguished root vertex, while every other class is level.
It remains to check ?, consider two nodes h_A,h_B\in L^{n+1} that belong to the same cluster with respect to s. By the construction provided by Lemma 3.4 of [albrechtsen2025excluding], the partition \mathcal{P} is the union of the partitions \mathcal{P}_D, where \mathcal{D} ranges over the classes of the partition \mathcal R of the components of G-G^n. Moreover, for every component C\in \mathcal{P}_D, the height R_C of C depends on the component D such that C\in \mathcal{P}_D. Thus, it is enough to prove A and B belong to \mathcal{P}_D for the same D\in \mathcal{R}.
Suppose, for a contradiction, that D_A\neq D_B. Recall that h_A,h_B\in L^{n+1}. If at least one of D_A,D_B is a singleton \{C\}, then, by the definition of \mathcal{R}, C has neighbors in exactly one partition class V_x for some vertex x\in L^n. In this case, one of V_{h_A}, V_{h_B} is contained in C, and V_x separates V_{h_A} from V_{h_B} in G. Consequently, every path between h_A and h_B in H contains x, and hence h_A and h_B lie in distinct components of H-B_H(s,n).
Otherwise, both D_A and D_B are of the form D_Z for some components Z of G-G_{n-1}. Since D_A\neq D_B, the corresponding components Z_A and Z_B are distinct. Moreover, V_{h_A}\subseteq Z_A and V_{h_B}\subseteq Z_B. Hence any path in H between h_A and h_B must contain a node of L^n, otherwise, by the definition of an H-partition, the corresponding bags would yield a path in G-G_{n-1} joining Z_A and Z_B, a contradiction. Thus again h_A and h_B lie in distinct components of H-B_H(s,n).
In both cases, h_A and h_B cannot belong to the same cluster with respect to s, since they both lie in L^{n+1}. This contradiction shows that D_A=D_B. Therefore
Hence, ? follows.
Finally, if H contained \Theta_t as a minor, then, since \Theta_t is 2-connected, Lemma 3.3 of [albrechtsen2025excluding] would yield a 3K-fat \Theta_t-minor in G, a contradiction. Thus H is \Theta_t-minor-free. Taking f(K,t)=R(t,3K) proves the lemma.
Is it true that if a graph X has no K-fat U_t-minor for some K \in \mathbb{N}, t\geq 3, then X admits a map of bounded additive distortion onto a graph with no U_t-minor?
The text on this page is from arXiv:2609.36186v1 by Dibyayan Chakraborty and Yann Vaxès, under the licence CC BY 4.0. The corpus made these changes: it read the items out of the LaTeX source, it expanded the macros of the paper, it put each proof after its statement, and it converted the text to HTML. The formulas are the TeX of the paper, and KaTeX renders them. A formula or a reference can look different from the PDF. The page is published under CC BY 4.0.
The record of each item is in analysis/objects/2609/2609.36186.jsonl.