English
Related papers

Related papers: On Teacher Hacking in Language Model Distillation

200 papers

Knowledge distillation involves transferring the predictive capabilities of large, high-performing AI models (teachers) to smaller models (students) that can operate in environments with limited computing power. In this paper, we address…

Machine Learning · Computer Science 2026-01-12 Pattarawat Chormai , Ali Hashemi , Klaus-Robert Müller , Grégoire Montavon

Knowledge distillation (KD) aims to transfer knowledge from a large teacher model to a smaller student model. Previous work applying KD in the field of large language models (LLMs) typically focused on the post-training phase, where the…

Computation and Language · Computer Science 2024-10-22 Hao Peng , Xin Lv , Yushi Bai , Zijun Yao , Jiajie Zhang , Lei Hou , Juanzi Li

Model distillation has emerged as a prominent technique to improve neural search models. To date, distillation taken an offline approach, wherein a new neural model is trained to predict relevance scores between arbitrary queries and…

Information Retrieval · Computer Science 2023-06-19 Sean MacAvaney , Xi Wang

After a large "teacher" neural network has been trained on labeled data, the probabilities that the teacher assigns to incorrect classes reveal a lot of information about the way in which the teacher generalizes. By training a small…

Machine Learning · Computer Science 2020-06-12 Rafael Müller , Simon Kornblith , Geoffrey Hinton

Knowledge distillation methods have recently shown to be a promising direction to speedup the synthesis of large-scale diffusion models by requiring only a few inference steps. While several powerful distillation methods were recently…

Computer Vision and Pattern Recognition · Computer Science 2024-04-08 Nikita Starodubcev , Artem Fedorov , Artem Babenko , Dmitry Baranchuk

Knowledge distillation (KD) has shown great promise in transferring knowledge from larger teacher models to smaller student models. However, existing KD strategies for large language models often minimize output distributions between…

Computation and Language · Computer Science 2024-12-23 Yuncheng Song , Liang Ding , Changtong Zan , Shujian Huang

Large Language Models (LLMs) have displayed remarkable performances across various complex tasks by leveraging Chain-of-Thought (CoT) prompting. Recently, studies have proposed a Knowledge Distillation (KD) approach, reasoning distillation,…

Computation and Language · Computer Science 2024-10-14 Hojae Lee , Junho Kim , SangKeun Lee

Knowledge distillation is widely used to improve generalization in practice, yet its theoretical understanding remains elusive. In the standard distillation setting, a teacher model provides soft predictions to guide the training of a…

Information Theory · Computer Science 2026-05-18 Bingying Li , Haiyun He

Large Language models (LLMs) are achieving state-of-the-art performance in many different downstream tasks. However, the increasing urgency of data privacy puts pressure on practitioners to train LLMs with Differential Privacy (DP) on…

Machine Learning · Computer Science 2025-12-17 James Flemings , Murali Annavaram

Offline distillation is a two-stage pipeline that requires expensive resources to train a teacher network and then distill the knowledge to a student for deployment. Online knowledge distillation, on the other hand, is a one-stage strategy…

Computer Vision and Pattern Recognition · Computer Science 2023-12-19 Baitan Shao , Ying Chen

Knowledge distillation aims at transferring knowledge acquired in one model (a teacher) to another model (a student) that is typically smaller. Previous approaches can be expressed as a form of training the student to mimic output…

Computer Vision and Pattern Recognition · Computer Science 2019-05-02 Wonpyo Park , Dongju Kim , Yan Lu , Minsu Cho

Distillation addresses the slow sampling problem in diffusion models by creating models with smaller size or fewer steps that approximate the behavior of high-step teachers. In this work, we propose a reinforcement learning based…

Machine Learning · Computer Science 2025-12-30 Amirhossein Tighkhorshid , Zahra Dehghanian , Gholamali Aminian , Chengchun Shi , Hamid R. Rabiee

Techniques such as ensembling and distillation promise model quality improvements when paired with almost any base model. However, due to increased test-time cost (for ensembles) and increased complexity of the training pipeline (for…

Machine Learning · Computer Science 2020-08-24 Rohan Anil , Gabriel Pereyra , Alexandre Passos , Robert Ormandi , George E. Dahl , Geoffrey E. Hinton

Recent research has explored distilling knowledge from large language models (LLMs) to optimize retriever models, especially within the retrieval-augmented generation (RAG) framework. However, most existing training methods rely on…

Information Retrieval · Computer Science 2024-06-19 Zizhong Li , Haopeng Zhang , Jiawei Zhang

Knowledge distillation is the technique of compressing a larger neural network, known as the teacher, into a smaller neural network, known as the student, while still trying to maintain the performance of the larger neural network as much…

Machine Learning · Computer Science 2023-05-11 Tianxun Zhou , Keng-Hwee Chiam

Pretrained language models have led to significant performance gains in many NLP tasks. However, the intensive computing resources to train such models remain an issue. Knowledge distillation alleviates this problem by learning a…

Computation and Language · Computer Science 2020-05-04 Linqing Liu , Huan Wang , Jimmy Lin , Richard Socher , Caiming Xiong

In the past decade, there has been substantial progress at training increasingly deep neural networks. Recent advances within the teacher--student training paradigm have established that information about past training updates show promise…

Computer Vision and Pattern Recognition · Computer Science 2020-03-03 Liang Jiang , Zujie Wen , Zhongping Liang , Yafang Wang , Gerard de Melo , Zhe Li , Liangzhuang Ma , Jiaxing Zhang , Xiaolong Li , Yuan Qi

Knowledge distillation (KD) is widely used to train small, high-performing student language models (LMs) using large teacher LMs. While effective in fine-tuning, KD during pre-training faces efficiency, flexibility, and effectiveness…

Computation and Language · Computer Science 2025-03-20 Yuxian Gu , Hao Zhou , Fandong Meng , Jie Zhou , Minlie Huang

On-policy distillation (OPD) has become a popular training paradigm in the LLM community. This paradigm selects a larger model as the teacher to provide dense, fine-grained signals for each sampled trajectory, in contrast to reinforcement…

Machine Learning · Computer Science 2026-04-09 Chenxu Yang , Chuanyu Qin , Qingyi Si , Minghui Chen , Naibin Gu , Dingyu Yao , Zheng Lin , Weiping Wang , Jiaqi Wang , Nan Duan

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…

Machine Learning · Computer Science 2025-10-21 Pingzhi Li , Zhen Tan , Mohan Zhang , Huaizhi Qu , Huan Liu , Tianlong Chen