A survey on symbolic knowledge distillation of large language models

K Acharya, A Velasquez… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
This survey article delves into the emerging and critical area of symbolic knowledge
distillation in large language models (LLMs). As LLMs such as generative pretrained …

Survey on knowledge distillation for large language models: methods, evaluation, and application

C Yang, Y Zhu, W Lu, Y Wang, Q Chen, C Gao… - ACM Transactions on …, 2024 - dl.acm.org
Large Language Models (LLMs) have showcased exceptional capabilities in various
domains, attracting significant interest from both academia and industry. Despite their …

A survey on knowledge distillation of large language models

X Xu, M Li, C Tao, T Shen, R Cheng, J Li, C Xu… - arXiv preprint arXiv …, 2024 - arxiv.org
This survey presents an in-depth exploration of knowledge distillation (KD) techniques
within the realm of Large Language Models (LLMs), spotlighting the pivotal role of KD in …

Distilling Monolingual Models from Large Multilingual Transformers

P Singh, O De Clercq, E Lefever - Electronics, 2023 - mdpi.com
Although language modeling has been trending upwards steadily, models available for low-
resourced languages are limited to large multilingual models such as mBERT and XLM …

Knowledge bases and language models: Complementing forces

F Suchanek, AT Luu - International Joint Conference on Rules and …, 2023 - Springer
Large language models (LLMs), as a particular instance of generative artificial intelligence,
have revolutionized natural language processing. In this invited paper, we argue that LLMs …

Knowledge distillation of transformer-based language models revisited

C Lu, J Zhang, Y Chu, Z Chen, J Zhou, F Wu… - arXiv preprint arXiv …, 2022 - arxiv.org
In the past few years, transformer-based pre-trained language models have achieved
astounding success in both industry and academia. However, the large model size and high …

A Principled Framework for Knowledge-enhanced Large Language Model

S Wang, Z Liu, Z Wang, J Guo - arXiv preprint arXiv:2311.11135, 2023 - arxiv.org
Large Language Models (LLMs) are versatile, yet they often falter in tasks requiring deep
and reliable reasoning due to issues like hallucinations, limiting their applicability in critical …

Xai-driven knowledge distillation of large language models for efficient deployment on low-resource devices

R Cantini, A Orsino, D Talia - Journal of Big Data, 2024 - Springer
Abstract Large Language Models (LLMs) are characterized by their inherent memory
inefficiency and compute-intensive nature, making them impractical to run on low-resource …

Knowledge Circuits in Pretrained Transformers

Y Yao, N Zhang, Z Xi, M Wang, Z Xu, S Deng… - arXiv preprint arXiv …, 2024 - arxiv.org
The remarkable capabilities of modern large language models are rooted in their vast
repositories of knowledge encoded within their parameters, enabling them to perceive the …

Reasoning about concepts with LLMs: Inconsistencies abound

RU Sosa, KN Ramamurthy, M Chang… - First Conference on …, 2024 - openreview.net
The ability to summarize and organize knowledge into abstract concepts is key to learning
and reasoning. Many industrial applications rely on the consistent and systematic use of …