Related papers: Extracting alignment data in open models
The goal of model distillation is to faithfully transfer teacher model knowledge to a model which is faster, more generalizable, more interpretable, or possesses other desirable characteristics. Human-readability is an important and…
Dataset distillation methods have demonstrated remarkable performance for neural networks trained with very limited training data. However, a significant challenge arises in the form of \textit{architecture overfitting}: the distilled…
LLM alignment ensures that large language models behave safely and effectively by aligning their outputs with human values, goals, and intentions. Aligning LLMs employ huge amounts of data, computation, and time. Moreover, curating data…
Large language models have led to significant progress across many NLP tasks, although their massive sizes often incur substantial computational costs. Distillation has become a common practice to compress these large and highly capable…
Model distillation is frequently proposed as a technique to reduce the privacy leakage of machine learning. These empirical privacy defenses rely on the intuition that distilled ``student'' models protect the privacy of training data, as…
We introduce a low-resource safety enhancement method for aligning large language models (LLMs) without the need for supervised fine-tuning (SFT) or reinforcement learning from human feedback (RLHF). Our main idea is to exploit knowledge…
Many approaches have been proposed to use diffusion models to augment training datasets for downstream tasks, such as classification. However, diffusion models are themselves trained on large datasets, often with noisy annotations, and it…
Large-scale text-to-image diffusion models excel in generating high-quality images from textual inputs, yet concerns arise as research indicates their tendency to memorize and replicate training data, raising We also addressed the issue of…
Model distillation -- using outputs from a large teacher model to teach a small student model -- is a practical means of creating efficient models for a particular task. We ask: Can we identify a students' teacher based on its outputs? Such…
Machine unlearning -- efficiently removing the effect of a small "forget set" of training data on a pre-trained machine learning model -- has recently attracted significant research interest. Despite this interest, however, recent work…
This paper examines the stability of learned explanations for black-box predictions via model distillation with decision trees. One approach to intelligibility in machine learning is to use an understandable `student' model to mimic the…
Large Language Models are typically trained on datasets collected from the web, which may inadvertently contain harmful or sensitive personal information. To address growing privacy concerns, unlearning methods have been proposed to remove…
Large language models have driven significant progress in natural language processing, but their deployment requires substantial compute and memory resources. As models scale, compression techniques become essential for balancing model…
Structured prediction models aim at solving a type of problem where the output is a complex structure, rather than a single variable. Performing knowledge distillation for such models is not trivial due to their exponentially large output…
Recent success of deep learning is largely attributed to the sheer amount of data used for training deep neural networks.Despite the unprecedented success, the massive data, unfortunately, significantly increases the burden on storage and…
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…
Learning the distance metric between pairs of samples has been studied for image retrieval and clustering. With the remarkable success of pair-based metric learning losses, recent works have proposed the use of generated synthetic points on…
The training process of ranking models involves two key data selection decisions: a sampling strategy, and a labeling strategy. Modern ranking systems, especially those for performing semantic search, typically use a ``hard negative''…
Retinal image matching plays a crucial role in monitoring disease progression and treatment response. However, datasets with matched keypoints between temporally separated pairs of images are not available in abundance to train…
We address the challenge of getting efficient yet accurate recognition systems with limited labels. While recognition models improve with model size and amount of data, many specialized applications of computer vision have severe resource…