LoRe: Adaptive Interaction-Evaluation Routing with Per-Step Interaction Budgets for Iterative Graph Solvers
Jintao Li, Yong-Yi Wang, Zheng-An Wang, Heng Fan · 2026
Read LoRe: Adaptive Interaction-Evaluation Routing with Per-Step Interaction Budgets for Iterative Graph Solvers by Jintao Li, Yong-Yi Wang, Zheng-An Wang, Heng Fan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-29005.txt.
Description
Jintao Li and colleagues introduce LoRe, a training-free wrapper budgeting interaction checks for diffusion-based neural solvers. Each round it checks only a fixed share of interactions, routing compute to high-conflict or high-uncertainty ones. On Maximum Independent Set, it pushes feasible runs more than threefold past the memory limit, with about eightfold speedup and twelvefold memory reduction with quality kept. On the Traveling Salesperson Problem it reaches about 15-fold speedup at 1000 nodes with 44-fold memory reduction and good tours.
Catalog record
| CREATOR | Jintao Li, Yong-Yi Wang, Zheng-An Wang, Heng Fan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 46770 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.29005