Sophie Xhonneux

According to our database, Sophie Xhonneux authored at least 8 papers between 2024 and 2025.

Collaborative distances:

• Dijkstra number of five.
• Erdős number of four.

2025

Model Collapse Is Not a Bug but a Feature in Machine Unlearning for LLMs.

Yan Scholten, Sophie Xhonneux, Leo Schwinn, Stephan Günnemann

CoRR, July, 2025

LLM-Safety Evaluations Lack Robustness.

Tim Beyer, Sophie Xhonneux, Simon Geisler, Gauthier Gidel, Leo Schwinn, Stephan Günnemann

CoRR, March, 2025

A generative approach to LLM harmfulness detection with special red flag tokens.

Sophie Xhonneux, David Dobre, Mehrnaz Mohfakhami, Leo Schwinn, Gauthier Gidel

CoRR, February, 2025

Adversarial Alignment for LLMs Requires Simpler, Reproducible, and More Measurable Objectives.

Leo Schwinn, Yan Scholten, Tom Wollschläger, Sophie Xhonneux, Stephen Casper, Stephan Günnemann, Gauthier Gidel

CoRR, February, 2025

Asynchronous RLHF: Faster and More Efficient Off-Policy RL for Language Models.

Michael Noukhovitch, Shengyi Huang, Sophie Xhonneux, Arian Hosseini, Rishabh Agarwal, Aaron C. Courville

Proceedings of the Thirteenth International Conference on Learning Representations, 2025

2024

In-Context Learning Can Re-learn Forbidden Tasks.

Sophie Xhonneux, David Dobre, Jian Tang, Gauthier Gidel, Dhanya Sridhar

CoRR, 2024

Efficient Adversarial Training in LLMs with Continuous Attacks.

Sophie Xhonneux, Alessandro Sordoni, Stephan Günnemann, Gauthier Gidel, Leo Schwinn

Proceedings of the Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, 2024

Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space.

Leo Schwinn, David Dobre, Sophie Xhonneux, Gauthier Gidel, Stephan Günnemann

Proceedings of the Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, 2024