4 papers
Distribution-Specific Curvature Control with Finite-Sample Guarantees for Open-Weight Safety
Domenic Rosati, Ali Dadsetan, Hong Huang +5
A short fine-tuning run can undo the safety guards of an open-weight model---retraining a refusal-trained assistant to aid weapons development or produce hate speech. Preventing su…
Voluntary Collusion with Secret Tools in Competing LLM Agents
Xijie Zeng, Frank Rudzicz
Even when a tool is explicitly described as unfair and harmful to others, ostensibly safety-aligned LLM agents still voluntarily engage in secret collusion whenever doing so confer…
Limits of Convergence-Rate Control for Open-Weight Safety
Domenic Rosati, Xijie Zeng, Hong Huang +4
Open-weight foundation models can be fine-tuned for harmful purposes after release, yet no existing training resistance methods provide theoretical guarantees. Treating these inter…
Can large language models be privacy preserving and fair medical coders?
Ali Dadsetan, Dorsa Soleymani, Xijie Zeng +1
Protecting patient data privacy is a critical concern when deploying machine learning algorithms in healthcare. Differential privacy (DP) is a common method for preserving privacy…