相关论文: Attributional Safety Failures in Large Language Mo…

Diffusion language models (DLMs) have recently emerged as competitive alternatives to autoregressive (AR) language models, yet differences in their activation dynamics remain poorly understood. We characterize these dynamics in LLaDA-8B and…

As large language models (LLMs) become increasingly integrated into real-world applications, scalable and rigorous safety evaluation is essential. This paper introduces Aymara AI, a programmatic platform for generating and administering…

Vision-Language adaptation (VL adaptation) transforms Large Language Models (LLMs) into Large Vision-Language Models (LVLMs) for multimodal tasks, but this process often compromises the inherent safety capabilities embedded in the original…

This research addresses the complex challenge of automated repair of code vulnerabilities, vital for enhancing digital security in an increasingly technology-driven world. The study introduces a novel and efficient format for the…

Despite recent progress, it has been difficult to prevent semantic hallucinations in generative Large Language Models. One common solution to this is augmenting LLMs with a retrieval system and making sure that the generated output is…

Despite the remarkable versatility of Large Language Models (LLMs) and Multimodal LLMs (MLLMs) to generalize across both language and vision tasks, LLMs and MLLMs have shown vulnerability to jailbreaking, generating textual outputs that…

Large Language Models (LLMs) have rapidly become integral to numerous applications in critical domains where reliability is paramount. Despite significant advances in safety frameworks and guardrails, current protective measures exhibit…

Large vision-language models (LVLMs) are vulnerable to harmful input compared to their language-only backbones. We investigated this vulnerability by exploring LVLMs internal dynamics, framing their inherent safety understanding in terms of…

Recent advancements in Large Language Models (LLMs) have sparked widespread concerns about their safety. Recent work demonstrates that safety alignment of LLMs can be easily removed by fine-tuning with a few adversarially chosen…

Large Language Models (LLMs) are increasingly integrated into high-stakes applications, making robust safety guarantees a central practical and commercial concern. Existing safety evaluations predominantly rely on fixed collections of…

The rapid adoption of Mixture-of-Experts (MoE) architectures marks a major shift in the deployment of Large Language Models (LLMs). MoE LLMs improve scaling efficiency by activating only a small subset of parameters per token, but their…

Recent advancements in large language models (LLMs) have underscored their vulnerability to safety alignment jailbreaks, particularly when subjected to downstream fine-tuning. However, existing mitigation strategies primarily focus on…

Large language models (LLMs) show inherent brittleness in their safety mechanisms, as evidenced by their susceptibility to jailbreaking and even non-malicious fine-tuning. This study explores this brittleness of safety alignment by…

Multimodal Large Language Models (MLLMs) bridge the gap between visual and textual data, enabling a range of advanced applications. However, complex internal interactions among visual elements and their alignment with text can introduce…

Safety-aligned language models systematically refuse harmful requests. While activation steering can modulate refusal, ablating the raw "refusal vector" calculated from contrastive harmful and harmless prompts often causes collateral damage…

Recent advancements in visual context compression enable MLLMs to process ultra-long contexts efficiently by rendering text into images. However, we identify a critical vulnerability inherent to this paradigm: lowering image resolution…

Code-mixed discourse combines multiple languages in a single text. It is commonly used in informal discourse in countries with several official languages, but also in many other countries in combination with English or neighboring…

Safety-aligned language models often refuse cybersecurity requests whose wording resembles misuse, even when the task is authorized and defensive. This makes security evaluation ambiguous: a failed answer may reflect missing capability or…

Large Language Models (LLMs) are acquiring a wider range of capabilities, including understanding and responding in multiple languages. While they undergo safety training to prevent them from answering illegal questions, imbalances in…

Modern large language models (LLMs) are typically secured by auditing data, prompts, and refusal policies, while treating the forward pass as an implementation detail. We show that intermediate activations in decoder-only LLMs form a…