A Spectral Theory of Distortion in LLM Graph Reconstruction: Sharp Bounds and Empirical Characterization
Paper recorded by Signals 4 on 2026-09-29 in cs.LG. Abstract reproduced from arXiv; link to the original below.
Published 2026-09-29 on arXiv · recorded by Signals 4 on 2026-09-30
Category: cs.LG · 机器学习 · first seen 2026-09-30
Abstract
Evaluations of graph reconstruction by language models typically report a single aggregate distance between the original and the reconstructed graph. We prove that for the Wasserstein distance between Laplacian spectra such a summary is bracketed by two edge counts, the net change in edge number from below and the symmetric difference from above, each scaled by $2/n$ where $n$ is the number of ver
Read on arXiv →
Cite this page: A Spectral Theory of Distortion in LLM Graph Reconstruction: Sharp Bounds and Empirical Characterization: the #3 most recent of 334 cs.LG papers we have recorded (as of 2026-09-29). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/a-spectral-theory-of-distortion-in-llm-graph-reconstruction-sharp-bounds-and-emp.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable:
papers.json
Get 4 AI signals a day by email — free.
Get 4 AI signals a day by email — free