English
Related papers

Related papers: Reasoning Compression with Mixed-Policy Distillati…

200 papers

Policy distillation, which transfers a teacher policy to a student policy has achieved great success in challenging tasks of deep reinforcement learning. This teacher-student framework requires a well-trained teacher model which is…

Machine Learning · Computer Science 2020-06-09 Kwei-Herng Lai , Daochen Zha , Yuening Li , Xia Hu

Large language models (LLMs) have demonstrated remarkable performance across a wide range of natural language processing (NLP) tasks. However, these models are often difficult to deploy due to significant computational requirements and…

Computation and Language · Computer Science 2024-12-25 Vijay Goyal , Mustafa Khan , Aprameya Tirupati , Harveer Saini , Michael Lam , Kevin Zhu

Knowledge Distillation (KD) compresses computationally expensive pre-trained language models (PLMs) by transferring their knowledge to smaller models, allowing their use in resource-constrained or real-time settings. However, most smaller…

Computation and Language · Computer Science 2023-11-08 Hayeon Lee , Rui Hou , Jongpil Kim , Davis Liang , Hongbo Zhang , Sung Ju Hwang , Alexander Min

Knowledge distillation (KD) is a common approach to compress a teacher model to reduce its inference cost and memory footprint, by training a smaller student model. However, in the context of autoregressive language models (LMs), we…

Computation and Language · Computer Science 2024-06-18 Qihuang Zhong , Liang Ding , Li Shen , Juhua Liu , Bo Du , Dacheng Tao

The widespread deployment of Large Language Models (LLMs) is hindered by the high computational demands, making knowledge distillation (KD) crucial for developing compact smaller ones. However, the conventional KD methods endure the…

Computation and Language · Computer Science 2025-02-18 Zengkui Sun , Yijin Liu , Fandong Meng , Yufeng Chen , Jinan Xu , Jie Zhou

The deployment and application of Large Language Models (LLMs) is hindered by their memory inefficiency, computational demands, and the high costs of API inferences. Traditional distillation methods, which transfer the capabilities of LLMs…

Computation and Language · Computer Science 2024-11-21 Yifei Zhang , Bo Pan , Chen Ling , Yuntong Hu , Liang Zhao

Large language models (LLMs) excel at reasoning tasks but are expensive to deploy. Thus small language models (SLMs) are fine-tuned on CoT data generated by LLMs to copy LLMs' abilities. However, these CoT data may include noisy rationales…

Computation and Language · Computer Science 2025-09-10 Hongyan Xie , Yitong Yao , Yikun Ban , Zixuan Huang , Deqing Wang , Zhenhe Wu , Haoxiang Su , Chao Wang , Shuangyong Song

The significant computational demands of large language models have increased interest in distilling reasoning abilities into smaller models via Chain-of-Thought (CoT) distillation. Current CoT distillation methods mainly focus on…

Computation and Language · Computer Science 2026-04-20 Yao Chen , Jiawei Sheng , Wenyuan Zhang , Tingwen Liu

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

Advances in large language models (LLMs) significantly enhance reasoning capabilities but their deployment is restricted in resource-constrained scenarios. Knowledge distillation addresses this by transferring knowledge from powerful…

Computation and Language · Computer Science 2025-08-26 Zhenyu Lei , Zhen Tan , Song Wang , Yaochen Zhu , Zihan Chen , Yushun Dong , Jundong Li

On-policy self-distillation (OPSD) improves the reasoning performance of large language models (LLMs) by providing dense token-level supervision for on-policy rollouts. However, existing OPSD methods often yield limited gains on in-domain…

Computation and Language · Computer Science 2026-05-28 Ziqi Zhao , Xinyu Ma , Liu Yang , Yujie Feng , Daiting Shi , Jingzhou He , Xin Xin , Zhaochun Ren , Xiao-Ming Wu

Distilling reasoning paths from teacher to student models via supervised fine-tuning (SFT) provides a shortcut for improving the reasoning ability of smaller Large Language Models (LLMs). However, the reasoning paths generated by teacher…

Computation and Language · Computer Science 2026-02-03 Shicheng Xu , Liang Pang , Yunchang Zhu , Jia Gu , Zihao Wei , Jingcheng Deng , Feiyang Pan , Huawei Shen , Xueqi Cheng

On-policy distillation (OPD) has become a core technique in the post-training of large language models, yet its training dynamics remain poorly understood. This paper provides a systematic investigation of OPD dynamics and mechanisms. We…

Machine Learning · Computer Science 2026-04-16 Yaxuan Li , Yuxin Zuo , Bingxiang He , Jinqian Zhang , Chaojun Xiao , Cheng Qian , Tianyu Yu , Huan-ang Gao , Wenkai Yang , Zhiyuan Liu , Ning Ding

Large language models (LLMs) exhibit enhanced reasoning at larger scales, driving efforts to distill these capabilities into smaller models via teacher-student learning. Previous works simply fine-tune student models on teachers' generated…

Computation and Language · Computer Science 2024-05-31 Chengwei Dai , Kun Li , Wei Zhou , Songlin Hu

We introduce Proximal Policy Distillation (PPD), a novel policy distillation method that integrates student-driven distillation and Proximal Policy Optimization (PPO) to increase sample efficiency and to leverage the additional rewards that…

Machine Learning · Computer Science 2026-05-11 Giacomo Spigler

Large Language Models (LLMs) can transfer their reasoning skills to smaller models by teaching them to generate the intermediate reasoning process required to solve multistep reasoning tasks. While LLMs can accurately solve reasoning tasks…

Artificial Intelligence · Computer Science 2024-10-25 Shivam Adarsh , Kumar Shridhar , Caglar Gulcehre , Nicholas Monath , Mrinmaya Sachan

Knowledge Distillation (KD) is a promising technique for reducing the high computational demand of large language models (LLMs). However, previous KD methods are primarily applied to white-box classification models or training small models…

Computation and Language · Computer Science 2026-02-03 Yuxian Gu , Li Dong , Furu Wei , Minlie Huang

Recent advances in large language model (LLM) post-training have leveraged two distinct paradigms to enhance reasoning capabilities: reinforcement learning (RL) and knowledge distillation (KD). While RL enables the emergence of complex…

Machine Learning · Computer Science 2025-06-04 Hongling Xu , Qi Zhu , Heyuan Deng , Jinpeng Li , Lu Hou , Yasheng Wang , Lifeng Shang , Ruifeng Xu , Fei Mi

Compressing Large Language Models (LLMs) often leads to reduced performance, especially for knowledge-intensive tasks. In this work, we dive into how compression damages LLMs' inherent knowledge and the possible remedies. We start by…

Computation and Language · Computer Science 2024-02-19 Duc N. M Hoang , Minsik Cho , Thomas Merth , Mohammad Rastegari , Zhangyang Wang

End-to-end motion planning models equipped with deep neural networks have shown great potential for enabling full autonomous driving. However, the oversized neural networks render them impractical for deployment on resource-constrained…

Computer Vision and Pattern Recognition · Computer Science 2024-04-16 Kaituo Feng , Changsheng Li , Dongchun Ren , Ye Yuan , Guoren Wang