6 papers
Steering Robustness into World Action Models via Mechanistic Interpretability and Optimal Control
Jihoon Hong, Julian Skifstad, Qiyue Dai +2
World Action Models (WAMs) enable semantically- and physically-informed control but are brittle under distribution shift. In this work, we use mechanistic interpretability to study…
Activation Steering of Video Generation Models via Reduced-Order Linear Optimal Control
Jihoon Hong, Alice Chan, Qiyue Dai +2
Text-to-video (T2V) models trained on large-scale web data can generate undesired content, motivating interventions that reduce harmful outputs without sacrificing visual quality.…
ATLAS: A Large-Scale Evaluation Benchmark for Adversarial LiDAR Perception
Mellon M. Zhang, Siddhant Panse, Zimo Fan +3
Autonomous driving perception is typically evaluated on clean benchmark data, yet real-world deployment requires robustness to rare, structured, and potentially adversarial sensor…
Local Linearity of LLMs Enables Activation Steering via Model-Based Linear Optimal Control
Julian Skifstad, Xinyue Annie Yang, Glen Chou
Inference-time LLM alignment methods, particularly activation steering, offer an alternative to fine-tuning by directly modifying activations during generation. Existing methods, h…
MAPS: Preserving Vision-Language Representations via Module-Wise Proximity Scheduling for Better Vision-Language-Action Generalization
Chengyue Huang, Mellon M. Zhang, Robert Azarcon +2
Vision-Language-Action (VLA) models inherit strong priors from pretrained Vision-Language Models (VLMs), but naive fine-tuning often disrupts these representations and harms genera…
Towards Streaming LiDAR Object Detection with Point Clouds as Egocentric Sequences
Mellon M. Zhang, Glen Chou, Saibal Mukhopadhyay
Accurate and low-latency 3D object detection is essential for autonomous driving, where safety hinges on both rapid response and reliable perception. While rotating LiDAR sensors a…