相关论文: From Hard Refusals to Safe-Completions: Toward Out…

As vision-language models (VLMs) become increasingly capable, maintaining a balance between safety and usefulness remains a central challenge. Safety mechanisms, while essential, can backfire, causing over-refusal, where models decline…

Large (vision-)language models exhibit remarkable capability but remain highly susceptible to jailbreaking. Existing safety training approaches aim to have the model learn a refusal boundary between safe and unsafe, based on the user's…

Chat-based language models are designed to be helpful, yet they should not comply with every user request. While most existing work primarily focuses on refusal of "unsafe" queries, we posit that the scope of noncompliance should be…

This study addresses a critical gap in safety tuning practices for Large Language Models (LLMs) by identifying and tackling a refusal position bias within safety tuning data, which compromises the models' ability to appropriately refuse…

Safety alignment aims to ensure that large language models (LLMs) refuse harmful requests by post-training on harmful queries paired with refusal answers. Although safety alignment is widely adopted in industry, the overrefusal problem…

Current safety alignment methods encode safe behavior implicitly within model parameters, creating a fundamental opacity: we cannot easily inspect why a model refuses a request, nor intervene when its safety judgments fail. We propose Safe…

Safety-trained language models routinely refuse requests for help circumventing rules. But not all rules deserve compliance. When users ask for help evading rules imposed by an illegitimate authority, rules that are deeply unjust or absurd…

Safety alignment in Large Language Models is critical for healthcare; however, reliance on binary refusal boundaries often results in \emph{over-refusal} of benign queries or \emph{unsafe compliance} with harmful ones. While existing…

Large Language Models (LLMs) are widely used across sectors, yet their alignment with International Humanitarian Law (IHL) is not well understood. This study evaluates eight leading LLMs on their ability to refuse prompts that explicitly…

Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning and text generation. However, these models can inadvertently generate unsafe or biased responses when prompted with problematic inputs, raising…

Refusal on harmful prompts is a key safety behaviour in instruction-tuned large language models (LLMs), yet the internal causes of this behaviour remain poorly understood. We study two public instruction-tuned models, Gemma-2-2B-IT and…

Since the release of OpenAI's ChatGPT, generative language models have attracted extensive public attention. The increased usage has highlighted generative models' broad utility, but also revealed several forms of embedded bias. Some is…

As large language models (LLMs) become easily accessible nowadays, the trade-off between safety and helpfulness can significantly impact user experience. A model that prioritizes safety will cause users to feel less engaged and assisted…

Conversational large language models are fine-tuned for both instruction-following and safety, resulting in models that obey benign requests but refuse harmful ones. While this refusal behavior is widespread across chat models, its…

Recent research shows that fine-tuning on benign instruction-following data can inadvertently undo the safety alignment process and increase a model's propensity to comply with harmful queries. While instruction-following fine-tuning is…

Safety alignment in large language models (LLMs), particularly for cybersecurity tasks, primarily focuses on preventing misuse. While this approach reduces direct harm, it obscures a complementary failure mode: denial of assistance to…

Evaluating aligned large language models' (LLMs) ability to recognize and reject unsafe user requests is crucial for safe, policy-compliant deployments. Existing evaluation efforts, however, face three limitations that we address with…

Safety alignment in large language models (LLMs) is primarily evaluated under open-ended generation, where models can mitigate risk by refusing to respond. In contrast, many real-world applications place LLMs in structured decision-making…

Currently, large pre-trained language models are widely applied in neural code completion systems. Though large code models significantly outperform their smaller counterparts, around 70\% of displayed code completions from Github Copilot…

Recent advancements in large language models (LLMs) have demonstrated that fine-tuning and human alignment can render LLMs harmless. In practice, such "harmlessness" behavior is mainly achieved by training models to reject harmful requests,…