● ResearchNov 3, 2026 · 12:001 source

Interpretable structural temperature scaling for calibrating graph neural networks and point-cloud segmentation

From arxiv.org

AI;DR

Interpretable Structural Temperature Scaling (ISTS) is a twelve-parameter linear model for graph neural networks. ISTS sets each node's temperature from named descriptors of its confidence, neighbourhood, and closeness to labelled nodes. On seven graphs with two backbones and eleven baselines, ISTS has the best average rank in negative log-likelihood (NLL) and Brier score.

Read originalSign in to suggest

Sources

#inferenceMarkdownScore 0.13.3k tokens
Read original