conftrace_

feature suppression

6 papers

Explore in graph

Co-occurring keywords

domain generalization (1522) contrastive learning (4032) representation learning (6206) class collapse (6) token-aware transformation (1) subclass representation (1) degradation disentanglement (1) multi-degradation scenario (1) cross-correlation learning (1) safety-critical unlearning (1)

Papers

Piercing the Fog: Disentangling Key Features for Vision Models in Multi-Degradation Scenarios AAAI 2026
CRISP: Persistent Concept Unlearning via Sparse Autoencoders ACL 2026
Domain Generalization in CLIP via Learning with Diverse Text Prompts CVPR 2025
START: A Generalized State Space Model with Saliency-Driven Token-Aware Transformation NIPS 2024
Which Features are Learnt by Contrastive Learning? On the Role of Simplicity Bias in Class Collapse and Feature Suppression ICML 2023
Intriguing Properties of Contrastive Losses NIPS 2021