相关论文: Asking Back: Interaction-Layer Antidistillation Wa…

Knowledge distillation is a widely adopted technique for transferring capabilities from LLMs to smaller, more efficient student models. However, unauthorized use of knowledge distillation takes unfair advantage of the considerable effort…

The radioactive nature of Large Language Model (LLM) watermarking enables the detection of watermarks inherited by student models when trained on the outputs of watermarked teacher models, making it a promising tool for preventing…

Knowledge distillation from proprietary LLM APIs poses a growing threat to model providers, yet defenses against this attack remain fragmented and unevaluated. We present DistillGuard, a framework for systematically evaluating output-level…

Large Language Models (LLMs) represent substantial intellectual and economic investments, yet their effectiveness can inadvertently facilitate model imitation via knowledge distillation (KD). In practical scenarios, competitors can distill…

Large language models (LLMs) are trained on massive corpora that may contain sensitive information, creating privacy risks under membership inference attacks (MIAs). Knowledge distillation is widely used to compress LLMs into smaller…

The promise of LLM watermarking rests on a core assumption that a specific watermark proves authorship by a specific model. We demonstrate that this assumption is dangerously flawed. We introduce the threat of watermark spoofing, a…

Model distillation enables efficient emulation of frontier large language models (LLMs), creating a need for robust mechanisms to detect when a third-party student model has trained on a teacher model's outputs. However, existing…

Knowledge distillation from large language models (LLMs) assumes that the teacher's output distribution is a high-quality training signal. On reasoning tasks, this assumption is frequently violated. A model's intermediate representations…

Adversarial attacks pose a significant threat to the security and safety of deep neural networks being applied to modern applications. More specifically, in computer vision-based tasks, experts can use the knowledge of model architecture to…

Proprietary large language models (LLMs) embody substantial economic value and are generally exposed only as black-box APIs, yet adversaries can still exploit their outputs to extract knowledge via distillation. Existing defenses focus…

Distillation via sampling reasoning traces exposes closed-source frontier models to adversarial third parties who can bypass their guardrails and misappropriate their capabilities. Antidistillation methods aim to address this by poisoning…

Post-training of language models (LMs) increasingly relies on the following two stages: (i) knowledge distillation, where the LM is trained to imitate a larger teacher LM, and (ii) reinforcement learning from human feedback (RLHF), where…

Adversarial training is a widely adopted strategy to bolster the robustness of neural network models against adversarial attacks. This paper revisits the fundamental assumptions underlying image classification and suggests that representing…

Large language models (LLMs) are increasingly deployed worldwide, yet their safety alignment remains predominantly English-centric. This allows for vulnerabilities in non-English contexts, especially with low-resource languages. We…

Knowledge distillation is considered as a training and compression strategy in which two neural networks, namely a teacher and a student, are coupled together during training. The teacher network is supposed to be a trustworthy predictor…

Watermarking of language model outputs enables statistical detection of model-generated text, which can mitigate harms and misuses of language models. Existing watermarking strategies operate by altering the decoder of an existing language…

Despite their accuracy, neural network-based classifiers are still prone to manipulation through adversarial perturbations. Those perturbations are designed to be misclassified by the neural network, while being perceptually identical to…

Knowledge distillation (KD) transfers capabilities from large language models (LLMs) to smaller students, yet it can fail unpredictably and also underpins model leakage risks. Our analysis revealed several distillation traps: tail noise,…

Self-Supervised Learning (SSL) has become a prominent paradigm for pre-training encoders to learning general-purpose representations from unlabeled data and releasing them on third-party platforms for broad downstream deep learning tasks.…

To mitigate the potential harms of Large Language Models (LLMs)generated text, researchers have proposed watermarking, a process of embedding detectable signals within text. With watermarking, we can always accurately detect LLM-generated…