4 papers
OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization
Kavin Aravindan, Arihant Rastogi, Aadi Prasad +4
Activation steering provides a lightweight mechanism for controlling large language models at inference time, but steering vectors can have unintended externalities: utility vector…
Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability
Aaditya Vikram Prasad, Connor Watts, Jack Merullo +4
Language models trained on large-scale datasets have been shown to learn features that encode abstract concepts such as factuality or intent. Such features are traditionally used f…
Constraint-Preserving Data Generation for Visuomotor Policy Learning
Kevin Lin, Varun Ragunath, Andrew McAlinden +4
Large-scale demonstration data has powered key breakthroughs in robot manipulation, but collecting that data remains costly and time-consuming. We present Constraint-Preserving Dat…
TidyBot++: An Open-Source Holonomic Mobile Manipulator for Robot Learning
Jimmy Wu, William Chong, Robert Holmberg +6
Exploiting the promise of recent advances in imitation learning for mobile manipulation will require the collection of large numbers of human-guided demonstrations. This paper prop…