English
Related papers

Related papers: Skill-Conditioned Gated Self-Distillation for LLM …

200 papers

Conventional knowledge distillation (KD) methods require access to the internal information of teachers, e.g., logits. However, such information may not always be accessible for large pre-trained language models (PLMs). In this work, we…

Computation and Language · Computer Science 2023-06-16 Qinhong Zhou , Zonghan Yang , Peng Li , Yang Liu

Training vision-language models (VLMs) for complex reasoning remains a challenging task, i.a. due to the scarcity of high-quality image-text reasoning data. Conversely, text-based reasoning resources are abundant and scalable, but it is…

Computer Vision and Pattern Recognition · Computer Science 2025-10-29 Walid Bousselham , Hilde Kuehne , Cordelia Schmid

Structured data question answering (QA), including table QA, Knowledge Graph (KG) QA, and temporal KG QA, is a pivotal research area. Advances in large language models (LLMs) have driven significant progress in unified structural QA…

Computation and Language · Computer Science 2025-11-18 Yushan Zhu , Wen Zhang , Long Jin , Mengshu Sun , Ling Zhong , Zhiqiang Liu , Juan Li , Lei Liang , Chong Long , Chao Deng , Junlan Feng

Multivariate time series forecasting (MTSF) endeavors to predict future observations given historical data, playing a crucial role in time series data management systems. With advancements in large language models (LLMs), recent studies…

Machine Learning · Computer Science 2025-05-07 Chenxi Liu , Hao Miao , Qianxiong Xu , Shaowen Zhou , Cheng Long , Yan Zhao , Ziyue Li , Rui Zhao

Knowledge Distillation (KD) aims at improving the performance of a low-capacity student model by inheriting knowledge from a high-capacity teacher model. Previous KD methods typically train a student by minimizing a task-related loss and…

Computer Vision and Pattern Recognition · Computer Science 2019-09-10 Mengya Gao , Yujun Shen , Quanquan Li , Junjie Yan , Liang Wan , Dahua Lin , Chen Change Loy , Xiaoou Tang

Knowledge distillation is a technique to enhance the generalization ability of a student model by exploiting outputs from a teacher model. Recently, feature-map based variants explore knowledge transfer between manually assigned…

Computer Vision and Pattern Recognition · Computer Science 2021-08-31 Defang Chen , Jian-Ping Mei , Yuan Zhang , Can Wang , Yan Feng , Chun Chen

Knowledge distillation (KD) is a key technique for compressing large language models into smaller ones while preserving performance. Despite the recent traction of KD research, its effectiveness for smaller language models (LMs) and the…

Computation and Language · Computer Science 2025-08-05 Suhas Kamasetty Ramesh , Ayan Sengupta , Tanmoy Chakraborty

Self-distillation has emerged as a powerful framework for post-training LLMs, where a teacher conditioned on extra information guides a student without it, both from the same model. While this guidance is useful when the student has failed,…

Machine Learning · Computer Science 2026-05-12 Jeonghye Kim , Jiwon Jeon , Dongsheng Li , Yuqing Yang

Large language models (LLMs) have garnered increasing attention owing to their powerful logical reasoning capabilities. Generally, larger LLMs (L-LLMs) that require paid interfaces exhibit significantly superior performance compared to…

Artificial Intelligence · Computer Science 2025-11-11 Dong Chen , Shilin Zhang , Fei Gao , Yueting Zhuang , Siliang Tang , Qidong Liu , Mingliang Xu

State-of-the-art frameworks in self-supervised learning have recently shown that fully utilizing transformer-based models can lead to performance boost compared to conventional CNN models. Striving to maximize the mutual information of two…

Computer Vision and Pattern Recognition · Computer Science 2022-11-29 Jiho Jang , Seonhoon Kim , Kiyoon Yoo , Chaerin Kong , Jangho Kim , Nojun Kwak

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

Large language models trained with reinforcement learning (RL) for mathematical reasoning face a fundamental challenge: on problems the model cannot solve at all - "cliff" prompts - the RL gradient vanishes entirely, preventing any learning…

Machine Learning · Computer Science 2026-03-26 Ken Ding

Existing knowledge distillation methods mostly focus on distillation of teacher's prediction and intermediate activation. However, the structured representation, which arguably is one of the most critical ingredients of deep models, is…

Computer Vision and Pattern Recognition · Computer Science 2024-07-16 Jing Yang , Xiatian Zhu , Adrian Bulat , Brais Martinez , Georgios Tzimiropoulos

Adapting large language models (LLMs) to long-context tasks requires post-training methods that remain accurate and coherent over thousands of tokens. Existing approaches are limited in several ways: 1) off-policy methods such as supervised…

Computation and Language · Computer Science 2026-05-13 Miguel Moura Ramos , Duarte M. Alves , André F. T. Martins

Black-box knowledge distillation for large language models presents a strict trade-off. Simple off-policy methods (e.g., sequence-level knowledge distillation) struggle to correct the student's inherent errors. Fully on-policy methods…

Machine Learning · Computer Science 2026-04-24 Xiwen Chen , Jingjing Wang , Wenhui Zhu , Peijie Qiu , Xuanzhao Dong , Hejian Sang , Zhipeng Wang , Alborz Geramifard , Feng Luo

Capability distillation applies knowledge distillation to selected model capabilities, aiming to compress a large language model (LLM) into a smaller one while preserving the abilities needed for a downstream task. However, most existing…

Computation and Language · Computer Science 2026-05-13 Xueqi Cheng , Xugui Zhou , Tyler Derr , Yushun Dong

Knowledge distillation aims to transfer knowledge from a large teacher model to a compact student counterpart, often coming with a significant performance gap between them. We find that a too-large performance gap can hamper the training…

Computer Vision and Pattern Recognition · Computer Science 2024-10-08 Yong Guo , Shulian Zhang , Haolin Pan , Jing Liu , Yulun Zhang , Jian Chen

In natural language processing (NLP) tasks, slow inference speed and huge footprints in GPU usage remain the bottleneck of applying pre-trained deep models in production. As a popular method for model compression, knowledge distillation…

Computation and Language · Computer Science 2020-12-15 Fei Yuan , Linjun Shou , Jian Pei , Wutao Lin , Ming Gong , Yan Fu , Daxin Jiang

Knowledge Distillation (KD) is a central paradigm for transferring knowledge from a large teacher network to a typically smaller student model, often by leveraging soft probabilistic outputs. While KD has shown strong empirical success in…

Machine Learning · Computer Science 2026-01-06 Itai Morad , Nir Shlezinger , Yonina C. Eldar

Knowledge Distillation (KD) is one of the approaches to reduce the size of Large Language Models (LLMs). A LLM with smaller number of model parameters (student) is trained to mimic the performance of a LLM of a larger size (teacher model)…

Computation and Language · Computer Science 2025-04-29 Rishika Sen , Sujoy Roychowdhury , Sumit Soman , H. G. Ranjani , Srikhetra Mohanty