English
Related papers

Related papers: Causal Distillation for Language Models

200 papers

Knowledge distillation is typically conducted by training a small model (the student) to mimic a large and cumbersome model (the teacher). The idea is to compress the knowledge from the teacher by using its output probabilities as…

Computation and Language · Computer Science 2020-01-17 Gustavo Aguilar , Yuan Ling , Yu Zhang , Benjamin Yao , Xing Fan , Chenlei Guo

Deep models are susceptible to learning spurious correlations, even during the post-processing. We take a closer look at the knowledge distillation -- a popular post-processing technique for model compression -- and find that distilling…

Machine Learning · Computer Science 2023-02-23 Jiwoon Lee , Jaeho Lee

Language models rely on semantic priors to perform in-context learning, which leads to poor performance on tasks involving inductive reasoning. Instruction-tuning methods based on imitation learning can superficially enhance the in-context…

Computation and Language · Computer Science 2025-04-16 Nafis Sadeq , Xin Xu , Zhouhang Xie , Julian McAuley , Byungkyu Kang , Prarit Lamba , Xiang Gao

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…

Algorithmic efficiency techniques such as distillation (\cite{hinton2015distillation}) are useful in improving model quality without increasing serving costs, provided a larger teacher model is available for a smaller student model to learn…

Machine Learning · Computer Science 2025-10-09 Khoa Trinh , Gaurav Menghani , Erik Vee

Knowledge distillation leverages a teacher model to improve the training of a student model. A persistent challenge is that a better teacher does not always yield a better student, to which a common mitigation is to use additional…

Machine Learning · Computer Science 2024-10-10 Abhishek Panigrahi , Bingbin Liu , Sadhika Malladi , Andrej Risteski , Surbhi Goel

A popular approach to model compression is to train an inexpensive student model to mimic the class probabilities of a highly accurate but cumbersome teacher model. Surprisingly, this two-step knowledge distillation process often leads to…

Machine Learning · Statistics 2021-04-21 Tri Dao , Govinda M Kamath , Vasilis Syrgkanis , Lester Mackey

Knowledge distillation aims to compress a powerful yet cumbersome teacher model into a lightweight student model without much sacrifice of performance. For this purpose, various approaches have been proposed over the past few years,…

Computer Vision and Pattern Recognition · Computer Science 2022-03-29 Defang Chen , Jian-Ping Mei , Hailin Zhang , Can Wang , Yan Feng , Chun Chen

Language model pre-training, such as BERT, has significantly improved the performances of many natural language processing tasks. However, pre-trained language models are usually computationally expensive, so it is difficult to efficiently…

Computation and Language · Computer Science 2020-10-19 Xiaoqi Jiao , Yichun Yin , Lifeng Shang , Xin Jiang , Xiao Chen , Linlin Li , Fang Wang , Qun Liu

Model distillation has become essential for creating smaller, deployable language models that retain larger system capabilities. However, widespread deployment raises concerns about resilience to adversarial manipulation. This paper…

Machine Learning · Computer Science 2025-10-17 Harsh Chaudhari , Jamie Hayes , Matthew Jagielski , Ilia Shumailov , Milad Nasr , Alina Oprea

Neural machine translation (NMT) offers a novel alternative formulation of translation that is potentially simpler than statistical approaches. However to reach competitive performance, NMT models need to be exceedingly large. In this paper…

Computation and Language · Computer Science 2016-09-23 Yoon Kim , Alexander M. Rush

As Transfer Learning from large-scale pre-trained models becomes more prevalent in Natural Language Processing (NLP), operating these large models in on-the-edge and/or under constrained computational training or inference budgets remains…

Computation and Language · Computer Science 2020-03-03 Victor Sanh , Lysandre Debut , Julien Chaumond , Thomas Wolf

The use of large transformer-based models such as BERT, GPT, and T5 has led to significant advancements in natural language processing. However, these models are computationally expensive, necessitating model compression techniques that…

Computation and Language · Computer Science 2023-08-29 Apoorv Dankar , Adeem Jassani , Kartikaeya Kumar

Knowledge distillation is a technique for improving the performance of a simple "student" model by replacing its one-hot training labels with a distribution over labels obtained from a complex "teacher" model. While this simple approach has…

Machine Learning · Computer Science 2020-05-22 Aditya Krishna Menon , Ankit Singh Rawat , Sashank J. Reddi , Seungyeon Kim , Sanjiv Kumar

Running large-scale pre-trained language models in computationally constrained environments remains a challenging problem yet to be addressed, while transfer learning from these models has become prevalent in Natural Language Processing…

Computation and Language · Computer Science 2022-04-14 Andrei-Marius Avram , Darius Catrina , Dumitru-Clementin Cercel , Mihai Dascălu , Traian Rebedea , Vasile Păiş , Dan Tufiş

Tiny, causal models are crucial for embedded audio machine learning applications. Model compression can be achieved via distilling knowledge from a large teacher into a smaller student model. In this work, we propose a novel two-step…

Sound · Computer Science 2023-09-18 Rayan Daod Nathoo , Mikolaj Kegler , Marko Stamenovic

This paper introduces a novel approach for efficiently distilling LLMs into smaller, application-specific models, significantly reducing operational costs and manual labor. Addressing the challenge of deploying computationally intensive…

Computation and Language · Computer Science 2024-03-26 Lukas Vöge , Vincent Gurgul , Stefan Lessmann

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

As large language models increasingly mediate firm - customer interactions, firms face a tradeoff: the most capable models perform well but are costly and difficult to control at scale. Existing knowledge distillation methods address this…

Computation and Language · Computer Science 2026-02-23 Tong Wang , K. Sudhir

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