相关论文: Mitigating Memorization in LLMs using Activation S…

Large language models (LLMs) are known to memorize parts of their training data, raising important concerns around privacy and security. While previous research has focused on studying memorization in pre-trained models, much less is known…

Large language models are shown to memorize privacy information such as social security numbers in training data. Given the sheer scale of the training corpus, it is challenging to screen and filter these privacy data, either manually or…

Recent work has demonstrated the potential of contrastive steering for jailbreaking Large Language Models (LLMs). However, existing methods rely on limited and inherently biased contrastive prompts and require laborious manual tuning of…

Recent large-scale natural language processing (NLP) systems use a pre-trained Large Language Model (LLM) on massive and diverse corpora as a headstart. In practice, the pre-trained model is adapted to a wide array of tasks via fine-tuning…

Memorization in large language models (LLMs) is a growing concern. LLMs have been shown to easily reproduce parts of their training data, including copyrighted work. This is an important problem to solve, as it may violate existing…

Activation engineering is becoming increasingly popular as a means of online control of large language models (LLMs). In this work, we extend the idea of inference-time steering with vectors that represent a behavioral direction of interest…

Recently, large language models (LLMs) have emerged as a notable field, attracting significant attention for its ability to automatically generate intelligent contents for various application domains. However, LLMs still suffer from…

Extensive efforts have been made before the public release of Large language models (LLMs) to align their behaviors with human values. However, even meticulously aligned LLMs remain vulnerable to malicious manipulations such as…

Large Language Models (LLMs) frequently memorize long sequences verbatim, often with serious legal and privacy implications. Much prior work has studied such verbatim memorization using observational data. To complement such work, we…

Controlling multiple behavioral attributes in large language models (LLMs) at inference time is a challenging problem due to interference between attributes and the limitations of linear steering methods, which assume additive behavior in…

Machine learning models are known to leak sensitive information, as they inevitably memorize (parts of) their training data. More alarmingly, large language models (LLMs) are now trained on nearly all available data, which amplifies the…

Large language models (LLMs) exhibit remarkable generative capabilities but raise ethical and security concerns by memorizing sensitive data, reinforcing biases, and producing harmful content. These risks have spurred interest in LLM…

Large Language Models (LLMs) are foundational to AI advancements, facilitating applications like predictive text generation. Nonetheless, they pose risks by potentially memorizing and disseminating sensitive, biased, or copyrighted…

Recent advances in Knowledge Distillation (KD) aim to mitigate the high computational demands of Large Language Models (LLMs) by transferring knowledge from a large ''teacher'' to a smaller ''student'' model. However, students may inherit…

Modern large language models (LLMs) have established state-of-the-art performance through architectural improvements, but still require significant computational cost for inference. In an effort to reduce the inference cost, post-training…

While current large language models (LLMs) perform well on many knowledge-related tasks, they are limited by relying on their parameters as an implicit storage mechanism. As a result, they struggle with memorizing rare events and with…

Adapting language models (LMs) to new tasks via post-training carries the risk of degrading existing capabilities -- a phenomenon classically known as catastrophic forgetting. In this paper, toward identifying guidelines for mitigating this…

Large Language Models exhibit implicit personalities in their generation, but reliably controlling or aligning these traits to meet specific needs remains an open challenge. The need for effective mechanisms for behavioural manipulation of…

Effective and reliable control over large language model (LLM) behavior is a significant challenge. While activation steering methods, which add steering vectors to a model's hidden states, are a promising approach, existing techniques…

Activation steering methods control large language model (LLM) behavior by modifying internal activations at inference time. However, most existing activation steering methods rely on a fixed steering strength, leading to either…