Showing cs.AIShow all
2 papers · 1 filter
cs.AI2026
Routing Probes Can Improve Without New Information: An Exact-Null Audit of Uncertainty Beyond Model Outputs
Wenhao Liang, Lin Yue, Wei Emma Zhang +3
Routing signals of modern vision transformers -- expert gates, attention-residual weights and halting scores -- often improve probes that predict whether the model is correct, and…
cs.AI2026
How Much Can Reliability Drift Under a Fixed Confidence Distribution?
Wenhao Liang, Lin Yue, Wei Emma Zhang +3
A classifier's conditional accuracy can change while its confidence distribution stays exactly the same. We study the worst-case movement of the reliability relation under covariat…