Skip to content
conftrace_

conceptual shortcut

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

controllable debiasing (1) pre-trained language model (1) representation disentanglement (1) bias mitigation (1) model unlearning (1) contrastive learning (1)

Papers

CURE: Controlled Unlearning for Robust Embeddings — Mitigating Conceptual Shortcuts in Pre-Trained Language Models EMNLP 2025