相关论文: The Devil in the Details: Emergent Misalignment, F…

Increasing model size has unlocked a dazzling array of capabilities in modern language models. At the same time, even frontier models remain vulnerable to jailbreaks and prompt injections, despite concerted efforts to make them robust. As…

The scaling of large language models (LLMs) emphasizes increasing depth, yet performance gains diminish with added layers. Prior work introduces the concept of "effective depth", arguing that deeper models fail to fully utilize their layers…

Large language models (LLMs) achieve remarkable performance through ever-increasing parameter counts, but scaling incurs steep computational costs. To better understand LLM scaling, we study representational differences between LLMs and…

A pervasive phenomenon in machine learning applications is distribution shift, where training and deployment conditions for a machine learning model differ. As distribution shift typically results in a degradation in performance, much…

Recent breakthroughs in Large Language Models (LLMs) have led to their adoption across a wide range of tasks, ranging from code generation to machine translation and sentiment analysis, etc. Red teaming/Safety alignment efforts show that…

Large language models (LLMs) have demonstrated revolutionary capabilities in understanding complex contexts and performing a wide range of tasks. However, LLMs can also answer questions that are unethical or harmful, raising concerns about…

Reasoning in large language models has long been a central research focus, and recent studies employing reinforcement learning (RL) have introduced diverse methods that yield substantial performance gains with minimal or even no external…

Fine-tuning pretrained models is a standard and effective workflow in modern machine learning. However, robust fine-tuning (RFT), which aims to simultaneously achieve adaptation to a downstream task and robustness to adversarial examples,…

As increasingly capable open-weight large language models (LLMs) are deployed, improving their tamper resistance against unsafe modifications, whether accidental or intentional, becomes critical to minimize risks. However, there is no…

Progress in making neural networks more robust against adversarial attacks is mostly marginal, despite the great efforts of the research community. Moreover, the robustness evaluation is often imprecise, making it difficult to identify…

Despite the remarkable progress of deep neural networks (DNNs) in various visual tasks, their vulnerability to adversarial examples raises significant security concerns. Recent adversarial training methods leverage inverse adversarial…

Large Language Models (LLMs) have made significant advances in natural language processing, but their underlying mechanisms are often misunderstood. Despite exhibiting coherent answers and apparent reasoning behaviors, LLMs rely on…

Adversarial factuality refers to the deliberate insertion of misinformation into input prompts by an adversary, characterized by varying levels of expressed confidence. In this study, we systematically evaluate the performance of several…

Adversarial Training is proved to be an efficient method to defend against adversarial examples, being one of the few defenses that withstand strong attacks. However, traditional defense mechanisms assume a uniform attack over the examples…

Advances in deep generative and density models have shown impressive capacity to model complex probability density functions in lower-dimensional space. Also, applying such models to high-dimensional image data to model the PDF has shown…

Recent advances in the development of large language models have resulted in public access to state-of-the-art pre-trained language models (PLMs), including Generative Pre-trained Transformer 3 (GPT-3) and Bidirectional Encoder…

Adversarial perturbations are imperceptible changes to input pixels that can change the prediction of deep learning models. Learned weights of models robust to such perturbations are previously found to be transferable across different…

We study the model robustness against adversarial examples, referred to as small perturbed input data that may however fool many state-of-the-art deep learning models. Unlike previous research, we establish a novel theory addressing the…

Large Language Models (LLMs) excel in various natural language processing tasks but remain vulnerable to generating harmful content or being exploited for malicious purposes. Although safety alignment datasets have been introduced to…

While Large Language Models (LLMs) are widely documented to be sensitive to minor prompt perturbations and prone to sycophantic alignment, their robustness in consequential, rule-bound decision-making remains under-explored. We uncover a…