Related papers: When World Models Dream Wrong: Physical-Conditione…
High-definition maps provide precise environmental information essential for prediction and planning in autonomous driving systems. Due to the high cost of labeling and maintenance, recent research has turned to online HD map construction…
Can generative agents be trusted in multimodal environments? Despite advances in large language and vision-language models that enable agents to act autonomously and pursue goals in rich settings, their ability to reason about safety,…
Deep learning-based denoising models have been widely employed in vision tasks, functioning as filters to eliminate noise while retaining crucial semantic information. Additionally, they play a vital role in defending against adversarial…
Additive manufacturing (AM) is rapidly integrating into critical sectors such as aerospace, automotive, and healthcare. However, this cyber-physical convergence introduces new attack surfaces, especially at the interface between…
While adversarial neural networks have been shown successful for static image attacks, very few approaches have been developed for attacking online image streams while taking into account the underlying physical dynamics of autonomous…
This paper proposes a cyber-resilient secure control framework for autonomous vehicles (AVs) subject to false data injection (FDI) threats as actuator attacks. The framework integrates data-driven modeling, event-triggered communication,…
Generative models estimate the underlying distribution of a dataset to generate realistic samples according to that distribution. In this paper, we present the first membership inference attacks against generative models: given a data…
The remarkable recent advances in object-centric generative world models raise a few questions. First, while many of the recent achievements are indispensable for making a general and versatile world model, it is quite unclear how these…
Due to the high cost of training, large model (LM) practitioners commonly use pretrained models downloaded from untrusted sources, which could lead to owning compromised models. In-context learning is the ability of LMs to perform multiple…
To perform adversarial attacks in the physical world, many studies have proposed adversarial camouflage, a method to hide a target object by applying camouflage patterns on 3D object surfaces. For obtaining optimal physical adversarial…
This paper introduces the concept of Language-Guided World Models (LWMs) -- probabilistic models that can simulate environments by reading texts. Agents equipped with these models provide humans with more extensive and efficient control,…
Adversarial perturbations of normal images are usually imperceptible to humans, but they can seriously confuse state-of-the-art machine learning models. What makes them so special in the eyes of image classifiers? In this paper, we show…
We propose X-WAM, a Unified 4D World Model that unifies real-time robotic action execution and high-fidelity 4D world synthesis (video + 3D reconstruction) in a single framework, addressing the critical limitations of prior unified world…
In standard reinforcement learning settings, agents typically assume immediate feedback about the effects of their actions after taking them. However, in practice, this assumption may not hold true due to physical constraints and can…
We introduce DreamerAD, the first latent world model framework that enables efficient reinforcement learning for autonomous driving by compressing diffusion sampling from 100 steps to 1 - achieving 80x speedup while maintaining visual…
Multi-targeted adversarial attacks aim to mislead classifiers toward specific target classes using a single perturbation generator with a conditional input specifying the desired target class. Existing methods face two key limitations: (1)…
Legged locomotion over various terrains is challenging and requires precise perception of the robot and its surroundings from both proprioception and vision. However, learning directly from high-dimensional visual input is often…
Human motion prediction has achieved a brilliant performance with the help of convolution-based neural networks. However, currently, there is no work evaluating the potential risk in human motion prediction when facing adversarial attacks.…
Foundation models (FMs) for computer vision learn rich and robust representations, enabling their adaptation to task/domain-specific deployments with little to no fine-tuning. However, we posit that the very same strength can make…
World models have become increasingly popular in acting as learned traffic simulators. Recent work has explored replacing traditional traffic simulators with world models for policy training. In this work, we explore the robustness of…