Poster · Paris

The Structural Bias of ℓ₂-Regularized Cross-Entropy Heads in Class-Incremental Learning: Characterization, Limits, and a Gauge-Anchoring Fix

Name: The Structural Bias of ℓ₂-Regularized Cross-Entropy Heads in Class-Incremental Learning: Characterization, Limits, and a Gauge-Anchoring Fix
Start: 2026-12-10T12:30:00+01:00
End: 2026-12-10T14:30:00+01:00
Location: Paris Poster Hall, Paris

When: Thu 10 Dec 2026, 12:30–14:30 CET
Where: Paris Poster Hall
Session: Paris Poster Session 3
Authors: Milad Khademi Nori, Guanghui Wang
Organisations: Toronto Metropolitan University

Open in schedule Official page