相关论文: Surgery: Mitigating Harmful Fine-Tuning for Large …

Currently, open-sourced large language models (OSLLMs) have demonstrated remarkable generative performance. However, as their structure and weights are made public, they are exposed to jailbreak attacks even after alignment. Existing…

Current safety alignment for large language models(LLMs) continues to present vulnerabilities, given that adversarial prompting can effectively bypass their safety measures.Our investigation shows that these safety mechanisms predominantly…

Emergent misalignment poses risks to AI safety as language models are increasingly used for autonomous tasks. In this paper, we present a population of large language models (LLMs) fine-tuned on insecure datasets spanning 11 diverse…

Although Large Language Models (LLMs) have demonstrated impressive capabilities across a wide range of tasks, growing concerns have emerged over the misuse of sensitive, copyrighted, or harmful data during training. To address these…

Large language models often lose previously aligned safety behaviors when fine-tuned on benign data, a phenomenon known as catastrophic forgetting. Prior work shows that adding random safety examples can mitigate this effect, but it remains…

From the perspective of content safety issues, alignment has shown to limit large language models' (LLMs) harmful content generation. This intentional method of reinforcing models to not respond to certain user inputs seem to be present in…

We investigate the functional role of emergent outliers in large language models, specifically attention sinks (a few tokens that consistently receive large attention logits) and residual sinks (a few fixed dimensions with persistently…

Large language models (LLMs) achieve state-of-the-art performance on multiple language tasks, yet their safety guardrails can be circumvented, leading to harmful generations. In light of this, recent research on safety mechanisms has…

Large language models (LLMs) undergo alignment training to avoid harmful behaviors, yet the resulting safeguards remain brittle: jailbreaks routinely bypass them, and fine-tuning on narrow domains can induce ``emergent misalignment'' that…

Large Language Models (LLMs) show remarkable proficiency in natural language tasks, yet their frequent overconfidence-misalignment between predicted confidence and true correctness-poses significant risks in critical decision-making…

Open-source Large Language Models (LLMs) often employ safety alignment methods to resist harmful instructions. However, recent research shows that maliciously fine-tuning these LLMs on harmful data can easily bypass these safeguards. To…

Releasing open-source large language models (LLMs) presents a dual-use risk since bad actors can easily fine-tune these models for harmful purposes. Even without the open release of weights, weight stealing and fine-tuning APIs make closed…

Pre-trained large models for multimodal contrastive learning, such as CLIP, have been widely recognized in the industry as highly susceptible to data-poisoned backdoor attacks. This poses significant risks to downstream model training. In…

Language models, while capable of generating remarkably coherent and seemingly accurate text, can occasionally produce undesirable content, including harmful or toxic outputs. In this paper, we present a new two-stage approach to detect and…

The safety alignment of large language models (LLMs) is becoming increasingly important with their democratization. In this paper, we study the safety degradation that comes with adapting LLMs to new tasks. We attribute this safety…

Transformer-based large language models exhibit in-context learning, enabling adaptation to downstream tasks via few-shot prompting with demonstrations. In practice, such models are often fine-tuned to improve zero-shot performance on…

A binary decision task, like yes-no questions or answer verification, reflects a significant real-world scenario such as where users look for confirmation about the correctness of their decisions on specific issues. In this work, we observe…

Training large language models to follow instructions makes them perform better on a wide range of tasks and generally become more helpful. However, a perfectly helpful model will follow even the most malicious instructions and readily…

Large language models (LLMs) have emerged as powerful tools for addressing a wide range of general inquiries and tasks. Despite this, fine-tuning aligned LLMs on smaller, domain-specific datasets, critical to adapting them to specialized…

Large language model (LLM) unlearning has become a critical mechanism for removing undesired data, knowledge, or behaviors from pre-trained models while retaining their general utility. Yet, with the rise of open-weight LLMs, we ask: can…