Related papers: Narrow Fine-Tuning Erodes Safety Alignment in Visi…

Recent work has shown that narrow finetuning can produce broadly misaligned LLMs, a phenomenon termed emergent misalignment (EM). While concerning, these findings were limited to finetuning and activation steering, leaving out in-context…

As Large Language Model (LLM) agents become more widespread, associated misalignment risks increase. While prior research has studied agents' ability to produce harmful outputs or follow malicious instructions, it remains unclear how likely…

Multimodal large language models (MLLMs) face safety misalignment, where visual inputs enable harmful outputs. To address this, existing methods require explicit safety labels or contrastive data; yet, threat-related concepts are concrete…

Safety alignment is critical for deploying large language models (LLMs) in real-world applications, yet most existing approaches rely on large human-annotated datasets and static red-teaming benchmarks that are costly, difficult to scale,…

The safety of large language models (LLMs) has increasingly emerged as a fundamental aspect of their development. Existing safety alignment for LLMs is predominantly achieved through post-training methods, which are computationally…

Vision-Language Models (VLMs) achieve strong cross-modal performance, yet recent evidence suggests they over-rely on textual descriptions while under-utilizing visual evidence -- a phenomenon termed ``text shortcut learning.'' We propose an…

Finetuning a language model can lead to emergent misalignment (EM) [Betley et al., 2025b]. Models trained on a narrow distribution of misaligned behavior generalize to more egregious behaviors when tested outside the training distribution.…

Fine-tuning LLMs on narrow harmful datasets can induce Emergent Misalignment (EM), where models exhibit misaligned behavior far beyond the fine-tuning distribution. We argue that emergent misalignment can be better understood as a…

Fine-tuning large language models (LLMs) improves performance but introduces critical safety vulnerabilities: even minimal harmful data can severely compromise safety measures. We observe that perturbations orthogonal to the alignment…

Emergent misalignment (EM), where fine-tuning on a narrow task (like insecure code) causes broad misalignment across unrelated domains, was first demonstrated by Betley et al. (2025). We conduct the most comprehensive EM study to date,…

Fine-tuning LLMs on benign data can still degrade alignment and adversarial robustness, yet direct analysis of the role of fine-tuning objectives in shaping these safety outcomes remain limited. We present a controlled comparison of six…

Safety alignment is a key requirement for building reliable Artificial General Intelligence. Despite significant advances in safety alignment, we observe that minor latent shifts can still trigger unsafe responses in aligned models. We…

The emergence of finetuning-as-a-service has revealed a new vulnerability in large language models (LLMs). A mere handful of malicious data uploaded by users can subtly manipulate the finetuning process, resulting in an alignment-broken…

Safety alignment is essential for the responsible deployment of large language models (LLMs). Yet, existing approaches often rely on heavyweight fine-tuning that is costly to update, audit, and maintain across model families. Full…

Misalignment in Large Language Models (LLMs) arises when model behavior diverges from human expectations and fails to simultaneously satisfy safety, value, and cultural dimensions, which must co-occur in real-world settings to solve a…

Safety alignment -- training large language models (LLMs) to refuse harmful requests while remaining helpful -- is critical for responsible deployment. Prior work established that safety behaviors are governed by low-rank structures,…

AI developers often apply safety alignment procedures to prevent the misuse of their AI systems. For example, before Meta released Llama 2-Chat - a collection of instruction fine-tuned large language models - they invested heavily in safety…

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…

Cross-modal alignment aims to map heterogeneous modalities into a shared latent space, as exemplified by models like CLIP, which benefit from large-scale image-text pretraining for strong recognition capabilities. However, when operating in…

Instruction-following language models are trained to be helpful and safe, yet their safety behavior can deteriorate under benign fine-tuning and worsen under adversarial updates. Existing defenses often offer limited protection or force a…