Yuancheng Xu

Yuancheng Xu, Yanchao Sun, Micah Goldblum et al.

The robustness of a deep classifier can be characterized by its margins: the decision boundary's distances to natural data points. However, it is unclear whether existing robust training methods effectively increase the margin for each vulnerable point during training. To understand this, we propose a continuous-time framework for quantifying the relative speed of the decision boundary with respect to each individual point. Through visualizing the moving speed of the decision boundary under Adversarial Training, one of the most effective robust training algorithms, a surprising moving-behavior is revealed: the decision boundary moves away from some vulnerable points but simultaneously moves closer to others, decreasing their margins. To alleviate these conflicting dynamics of the decision boundary, we propose Dynamics-aware Robust Training (DyART), which encourages the decision boundary to engage in movement that prioritizes increasing smaller margins. In contrast to prior works, DyART directly operates on the margins rather than their indirect approximations, allowing for more targeted and effective robustness improvement. Experiments on the CIFAR-10 and Tiny-ImageNet datasets verify that DyART alleviates the conflicting dynamics of the decision boundary and obtains improved robustness under various perturbation sizes compared to the state-of-the-art defenses. Our code is available at https://github.com/Yuancheng-Xu/Dynamics-Aware-Robust-Training.

Bang An, Sicheng Zhu, Ruiyi Zhang et al.

Safety-aligned large language models (LLMs) sometimes falsely refuse pseudo-harmful prompts, like "how to kill a mosquito," which are actually harmless. Frequent false refusals not only frustrate users but also provoke a public backlash against the very values alignment seeks to protect. In this paper, we propose the first method to auto-generate diverse, content-controlled, and model-dependent pseudo-harmful prompts. Using this method, we construct an evaluation dataset called PHTest, which is ten times larger than existing datasets, covers more false refusal patterns, and separately labels controversial prompts. We evaluate 20 LLMs on PHTest, uncovering new insights due to its scale and labeling. Our findings reveal a trade-off between minimizing false refusals and improving safety against jailbreak attacks. Moreover, we show that many jailbreak defenses significantly increase the false refusal rates, thereby undermining usability. Our method and dataset can help developers evaluate and fine-tune safer and more usable LLMs. Our code and dataset are available at https://github.com/umd-huang-lab/FalseRefusal

Yuancheng Xu, Chenghao Deng, Yanchao Sun et al.

Decisions made by machine learning models can have lasting impacts, making long-term fairness a critical consideration. It has been observed that ignoring the long-term effect and directly applying fairness criterion in static settings can actually worsen bias over time. To address biases in sequential decision-making, we introduce a long-term fairness concept named Equal Long-term Benefit Rate (ELBERT). This concept is seamlessly integrated into a Markov Decision Process (MDP) to consider the future effects of actions on long-term fairness, thus providing a unified framework for fair sequential decision-making problems. ELBERT effectively addresses the temporal discrimination issues found in previous long-term fairness notions. Additionally, we demonstrate that the policy gradient of Long-term Benefit Rate can be analytically simplified to standard policy gradients. This simplification makes conventional policy optimization methods viable for reducing bias, leading to our bias mitigation approach ELBERT-PO. Extensive experiments across various diverse sequential decision-making environments consistently reveal that ELBERT-PO significantly diminishes bias while maintaining high utility. Code is available at https://github.com/umd-huang-lab/ELBERT.

Xiaoyu Liu, Jiaxin Yuan, Bang An et al.

Representation learning assumes that real-world data is generated by a few semantically meaningful generative factors (i.e., sources of variation) and aims to discover them in the latent space. These factors are expected to be causally disentangled, meaning that distinct factors are encoded into separate latent variables, and changes in one factor will not affect the values of the others. Compared to statistical independence, causal disentanglement allows more controllable data generation, improved robustness, and better generalization. However, most existing work assumes unconfoundedness in the discovery process, that there are no common causes to the generative factors and thus obtain only statistical independence. In this paper, we recognize the importance of modeling confounders in discovering causal generative factors. Unfortunately, such factors are not identifiable without proper inductive bias. We fill the gap by introducing a framework entitled Confounded-Disentanglement (C-Disentanglement), the first framework that explicitly introduces the inductive bias of confounder via labels from domain expertise. In addition, we accordingly propose an approach to sufficiently identify the causally disentangled factors under any inductive bias of the confounder. We conduct extensive experiments on both synthetic and real-world datasets. Our method demonstrates competitive results compared to various SOTA baselines in obtaining causally disentangled features and downstream tasks under domain shifts.

Yuancheng Xu, Jiarui Yao, Manli Shu et al.

Vision-Language Models (VLMs) excel in generating textual responses from visual inputs, but their versatility raises security concerns. This study takes the first step in exposing VLMs' susceptibility to data poisoning attacks that can manipulate responses to innocuous, everyday prompts. We introduce Shadowcast, a stealthy data poisoning attack where poison samples are visually indistinguishable from benign images with matching texts. Shadowcast demonstrates effectiveness in two attack types. The first is a traditional Label Attack, tricking VLMs into misidentifying class labels, such as confusing Donald Trump for Joe Biden. The second is a novel Persuasion Attack, leveraging VLMs' text generation capabilities to craft persuasive and seemingly rational narratives for misinformation, such as portraying junk food as healthy. We show that Shadowcast effectively achieves the attacker's intentions using as few as 50 poison samples. Crucially, the poisoned samples demonstrate transferability across different VLM architectures, posing a significant concern in black-box settings. Moreover, Shadowcast remains potent under realistic conditions involving various text prompts, training data augmentation, and image compression techniques. This work reveals how poisoned VLMs can disseminate convincing yet deceptive misinformation to everyday, benign users, emphasizing the importance of data integrity for responsible VLM deployments. Our code is available at: https://github.com/umd-huang-lab/VLM-Poisoning.

Baijiong Lin, Weisen Jiang, Yuancheng Xu et al.

Multi-objective test-time alignment aims to adapt large language models (LLMs) to diverse multi-dimensional user preferences during inference while keeping LLMs frozen. Recently, GenARM (Xu et al., 2025) first independently trains Autoregressive Reward Models (ARMs) for each preference dimension without awareness of each other, then combines their outputs based on user-specific preference vectors during inference to achieve multi-objective test-time alignment, leading to two key limitations: the need for \textit{multiple} ARMs increases the inference cost, and the separate training of ARMs causes the misalignment between the guided generation and the user preferences. To address these issues, we propose Preference-aware ARM (PARM), a single unified ARM trained across all preference dimensions. PARM uses our proposed Preference-Aware Bilinear Low-Rank Adaptation (PBLoRA), which employs a bilinear form to condition the ARM on preference vectors, enabling it to achieve precise control over preference trade-offs during inference. Experiments demonstrate that PARM reduces inference costs and achieves better alignment with preference vectors compared with existing methods. Additionally, PARM enables weak-to-strong guidance, allowing a smaller PARM to guide a larger frozen LLM without expensive training, making multi-objective alignment accessible with limited computing resources. The code is available at https://github.com/Baijiong-Lin/PARM.

Xiyao Wang, Yuhang Zhou, Xiaoyu Liu et al.

Multimodal Large Language Models (MLLMs) have demonstrated proficiency in handling a variety of visual-language tasks. However, current MLLM benchmarks are predominantly designed to evaluate reasoning based on static information about a single image, and the ability of modern MLLMs to extrapolate from image sequences, which is essential for understanding our ever-changing world, has been less investigated. To address this challenge, this paper introduces Mementos, a new benchmark designed to assess MLLMs' sequential image reasoning abilities. Mementos features 4,761 diverse image sequences with varying lengths. We also employ a GPT-4 assisted method to evaluate MLLM reasoning performance. Through a careful evaluation of nine recent MLLMs on Mementos, including GPT-4V and Gemini, we find that they struggle to accurately describe dynamic information about given image sequences, often leading to hallucinations/misrepresentations of objects and their corresponding behaviors. Our quantitative analysis and case studies identify three key factors impacting MLLMs' sequential image reasoning: the correlation between object and behavioral hallucinations, the influence of cooccurring behaviors, and the compounding impact of behavioral hallucinations. Our dataset is available at https://github.com/umd-huang-lab/Mementos.

Bang An, Mucong Ding, Tahseen Rabbani et al.

In the burgeoning age of generative AI, watermarks act as identifiers of provenance and artificial content. We present WAVES (Watermark Analysis Via Enhanced Stress-testing), a benchmark for assessing image watermark robustness, overcoming the limitations of current evaluation methods. WAVES integrates detection and identification tasks and establishes a standardized evaluation protocol comprised of a diverse range of stress tests. The attacks in WAVES range from traditional image distortions to advanced, novel variations of diffusive, and adversarial attacks. Our evaluation examines two pivotal dimensions: the degree of image quality degradation and the efficacy of watermark detection after attacks. Our novel, comprehensive evaluation reveals previously undetected vulnerabilities of several modern watermarking algorithms. We envision WAVES as a toolkit for the future development of robust watermarks. The project is available at https://wavesbench.github.io/

Pengxin Guo, Yuancheng Xu, Baijiong Lin et al.

Deep neural networks have achieved impressive performance in various areas, but they are shown to be vulnerable to adversarial attacks. Previous works on adversarial attacks mainly focused on the single-task setting. However, in real applications, it is often desirable to attack several models for different tasks simultaneously. To this end, we propose Multi-Task adversarial Attack (MTA), a unified framework that can craft adversarial examples for multiple tasks efficiently by leveraging shared knowledge among tasks, which helps enable large-scale applications of adversarial attacks on real-world systems. More specifically, MTA uses a generator for adversarial perturbations which consists of a shared encoder for all tasks and multiple task-specific decoders. Thanks to the shared encoder, MTA reduces the storage cost and speeds up the inference when attacking multiple tasks simultaneously. Moreover, the proposed framework can be used to generate per-instance and universal perturbations for targeted and non-targeted attacks. Experimental results on the Office-31 and NYUv2 datasets demonstrate that MTA can improve the quality of attacks when compared with its single-task counterpart.