Stuart Russell

6

Papers

34

Total Citations

Papers (6)

Monitoring Latent World States in Language Models with Propositional Probes

Diffusion On Syntax Trees For Program Synthesis

AssistanceZero: Scalably Solving Assistance Games

Image Hijacks: Adversarial Images can Control Generative Models at Runtime

AI Alignment with Changing and Influenceable Reward Functions

Position: Social Choice Should Guide AI Alignment in Dealing with Diverse Human Feedback