English
Related papers

Related papers: On Teacher Hacking in Language Model Distillation

200 papers

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…

Computation and Language · Computer Science 2026-01-06 Zishun Yu , Shangzhe Li , Xinhua Zhang

Knowledge distillation is a popular machine learning technique that aims to transfer knowledge from a large 'teacher' network to a smaller 'student' network and improve the student's performance by training it to emulate the teacher. In…

Machine Learning · Computer Science 2022-10-19 Sushil Thapa

Distilling reasoning traces from strong large language models into smaller ones is a promising route to improve intelligence in resource-constrained settings. Existing approaches face a fundamental trade-off: offline distillation from…

Computation and Language · Computer Science 2026-05-15 Yumeng Zhang , Zhengbang Yang , Yevin Nikhel Goonatilake , Zhuangdi Zhu

Distillation is the technique of training a "student" model based on examples that are labeled by a separate "teacher" model, which itself is trained on a labeled dataset. The most common explanations for why distillation "works" are…

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…

Computation and Language · Computer Science 2025-05-21 Somin Wadhwa , Chantal Shaib , Silvio Amir , Byron C. Wallace

The push to compress and impart the proficiency of Large Language Models (LLMs) into more deployable and efficient Small Language Models (SLMs) has benefited from improvements in knowledge distillation (KD) techniques. These techniques…

Artificial Intelligence · Computer Science 2025-07-02 Shreyansh Padarha

As Large Language Models (LLMs) continue to grow in both capability and cost, transferring frontier capabilities into smaller, deployable students has become a central engineering problem, and knowledge distillation remains the dominant…

Machine Learning · Computer Science 2026-05-19 Mingyang Song , Mao Zheng

It has been commonly observed that a teacher model with superior performance does not necessarily result in a stronger student, highlighting a discrepancy between current teacher training practices and effective knowledge transfer. In order…

Computation and Language · Computer Science 2024-05-16 Yuxin Ren , Zihan Zhong , Xingjian Shi , Yi Zhu , Chun Yuan , Mu Li

Knowledge distillation (KD) is a popular method to train efficient networks ("student") with the help of high-capacity networks ("teacher"). Traditional methods use the teacher's soft logits as extra supervision to train the student…

Computer Vision and Pattern Recognition · Computer Science 2021-08-17 Guo-Hua Wang , Yifan Ge , Jianxin Wu

Knowledge distillation is a popular technique for training a small student network to emulate a larger teacher model, such as an ensemble of networks. We show that while knowledge distillation can improve student generalization, it does not…

Machine Learning · Computer Science 2021-12-07 Samuel Stanton , Pavel Izmailov , Polina Kirichenko , Alexander A. Alemi , Andrew Gordon Wilson

This work studies knowledge distillation (KD) for large language models (LLMs) through preference optimization. We propose a reward-guided imitation learning framework for sequential KD, formulating a min-max optimization problem between…

Machine Learning · Computer Science 2025-05-27 Chen Jia

Reinforcement learning (RL) post-training has recently driven major gains in long chain-of-thought reasoning large language models (LLMs), but the high inference cost of such models motivates distillation into smaller students. Most…

Machine Learning · Computer Science 2026-04-13 Zhaoyang Zhang , Shuli Jiang , Yantao Shen , Yuting Zhang , Dhananjay Ram , Shuo Yang , Zhuowen Tu , Wei Xia , Stefano Soatto

Knowledge distillation is widely used as a means of improving the performance of a relatively simple student model using the predictions from a complex teacher model. Several works have shown that distillation significantly boosts the…

Machine Learning · Computer Science 2021-07-09 Michal Lukasik , Srinadh Bhojanapalli , Aditya Krishna Menon , Sanjiv Kumar

Knowledge distillation (KD) is an effective model compression method that can transfer the internal capabilities of large language models (LLMs) to smaller ones. However, the multi-modal probability distribution predicted by teacher LLMs…

Computation and Language · Computer Science 2024-12-19 Tianyu Peng , Jiajun Zhang

Knowledge distillation is a widely applicable technique for training a student neural network under the guidance of a trained teacher network. For example, in neural network compression, a high-capacity teacher is distilled to train a…

Computer Vision and Pattern Recognition · Computer Science 2019-08-05 Frederick Tung , Greg Mori

Knowledge distillation is a technique used to train a small student network using the output generated by a large teacher network, and has many empirical advantages~\citep{Hinton2015DistillingTK}. While the standard one-shot approach to…

Machine Learning · Computer Science 2025-03-25 Shivam Gupta , Sushrut Karmalkar

Distribution Matching Distillation (DMD) facilitates efficient inference by distilling multi-step diffusion models into few-step variants. Concurrently, Reinforcement Learning (RL) has emerged as a vital tool for aligning generative models…

Computer Vision and Pattern Recognition · Computer Science 2026-03-26 Dengyang Jiang , Dongyang Liu , Zanyi Wang , Qilong Wu , Liuzhuozheng Li , Hengzhuang Li , Xin Jin , David Liu , Changsheng Lu , Zhen Li , Bo Zhang , Mengmeng Wang , Steven Hoi , Peng Gao , Harry Yang

The performance of a distillation-based compressed network is governed by the quality of distillation. The reason for the suboptimal distillation of a large network (teacher) to a smaller network (student) is largely attributed to the gap…

Computer Vision and Pattern Recognition · Computer Science 2021-12-21 Vibhas Vats , David Crandall

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…

Cryptography and Security · Computer Science 2023-03-08 Matthew Jagielski , Milad Nasr , Christopher Choquette-Choo , Katherine Lee , Nicholas Carlini

Knowledge distillation optimises a smaller student model to behave similarly to a larger teacher model, retaining some of the performance benefits. While this method can improve results on in-distribution examples, it does not necessarily…

Computation and Language · Computer Science 2024-07-26 Joe Stacey , Marek Rei