Related papers: Dynamically Scaled Activation Steering

Finetuning large language models (LLMs) enables user-specific customization but introduces critical safety risks: even a few harmful examples can compromise safety alignment. A common mitigation strategy is to update the model more strongly…

Generating safety-critical scenarios in high-fidelity simulations offers a promising and cost-effective approach for efficient testing of autonomous vehicles. Existing methods typically rely on manipulating a single vehicle's trajectory…

Vision Language Models (VLMs) have demonstrated impressive capabilities in integrating visual and textual information for understanding and reasoning, but remain highly vulnerable to adversarial attacks. While activation steering has…

In the hardware design space exploration process, it is critical to optimize both hardware parameters and algorithm-to-hardware mappings. Previous work has largely approached this simultaneous optimization problem by separately exploring…

Test-time optimization remains impractical at scale due to prohibitive inference costs--techniques like iterative refinement and multi-step verification can require $10-100\times$ more compute per query than standard decoding. Latent space…

Active nematics, formed from a liquid crystalline suspension of active force dipoles, are a paradigmatic active matter system whose study provides insights into how chemical driving produces the cellular mechanical forces essential for…

Large pre-trained language models contain societal biases and carry along these biases to downstream tasks. Current in-processing bias mitigation approaches (like adversarial training) impose debiasing by updating a model's parameters,…

Adaptive control approaches yield high-performance controllers when a precise system model or suitable parametrizations of the controller are available. Existing data-driven approaches for adaptive control mostly augment standard…

Obtaining human-interpretable explanations of large, general-purpose language models is an urgent goal for AI safety. However, it is just as important that our interpretability methods are faithful to the causal dynamics underlying model…

In this paper, we apply the attention mechanism to autonomous driving for steering angle prediction. We propose the first model, applying the recently introduced sparse attention mechanism to visual domain, as well as the aggregated…

The complexity of the real world demands robotic systems that can intelligently adapt to unseen situations. We present STEER, a robot learning framework that bridges high-level, commonsense reasoning with precise, flexible low-level…

When humans physically interact with robots, we need the robots to be both safe and performant. Series elastic actuators (SEAs) fundamentally advance safety by introducing compliant actuation. On the one hand, adding a spring mitigates the…

Real-world artificial intelligence (AI) systems are increasingly required to operate autonomously in dynamic, uncertain, and continuously changing environments. However, most existing AI models rely on predefined objectives, static training…

Extensive simulation-based testing is important for assuring the safety of autonomous driving systems (ADS). However, generating safety-critical traffic scenarios remains challenging because failures often arise from rare, complex…

In this paper, we introduce Tolerant Discrete Barrier States (T-DBaS), a novel safety-embedding technique for trajectory optimization with enhanced exploratory capabilities. The proposed approach generalizes the standard discrete barrier…

Data assimilation (DA) integrates observations with a dynamical model to estimate states of PDE-governed systems. Model-driven methods (e.g., Kalman, particle) presuppose full knowledge of the true dynamics, which is not always satisfied in…

This paper investigates a novel active-sensing-based obstacle avoidance paradigm for flying robots in dynamic environments. Instead of fusing multiple sensors to enlarge the field of view (FOV), we introduce an alternative approach that…

Nighttime environments pose significant challenges for camera-based perception, as existing methods passively rely on the scene lighting. We introduce Lighting-driven Dynamic Active Sensing (LiDAS), a closed-loop active illumination system…

This study proposes the application of a backcasting approach to a mobility model with the aim of defining an optimal decarbonization roadmap. The selected decision variable is the introduction of a fleet of shared autonomous vehicles. The…

We introduce SteeringDiffusion, a bottlenecked activation-level control interface for diffusion models that exposes a smooth, monotonic, and runtime-adjustable control surface over the content--style trade-off. Our method keeps the U-Net…