English
Related papers

Related papers: Backtracking When It Strays: Mitigating Dual Expos…

200 papers

While neural end-to-end text-to-speech (TTS) is superior to conventional statistical methods in many ways, the exposure bias problem in the autoregressive models remains an issue to be resolved. The exposure bias problem arises from the…

Computation and Language · Computer Science 2020-02-21 Rui Liu , Berrak Sisman , Jingdong Li , Feilong Bao , Guanglai Gao , Haizhou Li

Large language models are often post-trained with sparse verifier rewards, which indicate whether a sampled trajectory succeeds but provide limited guidance about where reasoning succeeds or fails. On-policy distillation (OPD) offers denser…

Machine Learning · Computer Science 2026-05-14 Weichen Yu , Xiaomin Li , Yizhou Zhao , Xiaoze Liu , Ruowang Zhang , Haixin Wang , Yinyi Luo , Chen Henry Wu , Gaurav Mittal , Matt Fredrikson , Yu Hu

Large Language Models (LLMs) achieve state-of-the-art performance across various NLP tasks but face deployment challenges due to high computational costs and memory constraints. Knowledge distillation (KD) is a promising solution,…

Computation and Language · Computer Science 2025-03-04 Anh Duc Le , Tu Vu , Nam Le Hai , Nguyen Thi Ngoc Diep , Linh Ngo Van , Trung Le , Thien Huu Nguyen

Knowledge distillation is central to LLM post-training, yet its design space remains poorly understood, especially alongside reinforcement learning (RL). We show that the prevailing paradigms, off-policy distillation and on-policy…

Machine Learning · Computer Science 2026-05-19 Anhao Zhao , Haoran Xin , Yingqi Fan , Junlong Tong , Wenjie Li , Xiaoyu Shen

Large Language Models (LLMs) excel in reasoning tasks through Chain-of-Thought (CoT) prompting. However, CoT prompting greatly increases computational demands, which has prompted growing interest in distilling CoT capabilities into Small…

Computation and Language · Computer Science 2025-05-28 Xinghao Chen , Zhijing Sun , Wenjin Guo , Miaoran Zhang , Yanjun Chen , Yirong Sun , Hui Su , Yijie Pan , Dietrich Klakow , Wenjie Li , Xiaoyu Shen

Despite the success of distillation in large language models (LLMs), most prior work applies identical loss functions to both teacher- and student-generated data. These strategies overlook the synergy between loss formulations and data…

Computation and Language · Computer Science 2025-06-02 Jongwoo Ko , Tianyi Chen , Sungnyun Kim , Tianyu Ding , Luming Liang , Ilya Zharkov , Se-Young Yun

Large language models (LLMs) like GPT-4, DeepSeek-R1, and ReasonFlux have shown significant improvements in various reasoning tasks. However, smaller LLMs still struggle with complex mathematical reasoning because they fail to effectively…

Computation and Language · Computer Science 2025-02-27 Ling Yang , Zhaochen Yu , Tianjun Zhang , Minkai Xu , Joseph E. Gonzalez , Bin Cui , Shuicheng Yan

Knowledge distillation (KD) transfers capabilities from large language models (LLMs) to smaller students, yet it can fail unpredictably and also underpins model leakage risks. Our analysis revealed several distillation traps: tail noise,…

Machine Learning · Computer Science 2026-04-22 Weixiao Zhan , Yongcheng Jing , Leszek Rutkowski , Dacheng Tao

While chain-of-thought (CoT) distillation from advanced large language models (LLMs) has proven effective in general reasoning tasks, it struggles in scientific domains where even advanced models often produce incorrect or superficial…

Computation and Language · Computer Science 2025-10-17 Kehua Feng , Keyan Ding , Zhihui Zhu , Lei Liang , Qiang Zhang , Huajun Chen

On-policy distillation (OPD) is increasingly used in LLM post-training because it can leverage a teacher model to provide dense supervision on student rollouts. The standard implementation, however, usually reduces distribution matching to…

Machine Learning · Computer Science 2026-04-28 Yuqian Fu , Haohuan Huang , Kaiwen Jiang , Jiacai Liu , Zhuo Jiang , Yuanheng Zhu , Dongbin Zhao

Large Language Models (LLMs) have made significant strides in problem-solving by incorporating reasoning processes. However, this enhanced reasoning capability results in an increased number of output tokens during inference, leading to…

Computation and Language · Computer Science 2025-11-05 Jingxian Xu , Mengyu Zhou , Weichang Liu , Hanbing Liu , Shi Han , Dongmei Zhang

Backtracking, an emergent behavior elicited by reasoning fine-tuning, has been shown to be a key mechanism in reasoning models' enhanced capabilities. Prior work has succeeded in manipulating this behavior via steering vectors, but the…

Machine Learning · Computer Science 2025-07-18 Jake Ward , Chuqiao Lin , Constantin Venhoff , Neel Nanda

Previous chain-of-thought (CoT) distillation methods primarily focused on enhancing the reasoning capabilities of Small Language Models (SLMs) by utilizing high-quality rationales generated by powerful Large Language Models (LLMs, e.g.,…

Computation and Language · Computer Science 2025-08-18 Ziyang Ma , Qingyue Yuan , Linhai Zhang , Deyu Zhou

Eliciting "chain of thought" (CoT) rationales -- sequences of token that convey a "reasoning" process -- has been shown to consistently improve LLM performance on tasks like question answering. More recent efforts have shown that such…

Computation and Language · Computer Science 2024-10-01 Somin Wadhwa , Silvio Amir , Byron C. Wallace

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

On-policy distillation is a promising approach for transferring knowledge between language models, where a student learns from dense token-level signals along its own trajectories. This framework typically uses reverse KL divergence,…

Machine Learning · Computer Science 2026-05-25 Woogyeol Jin , Taywon Min , Yongjin Yang , Swanand Ravindra Kadhe , Yi Zhou , Dennis Wei , Nathalie Baracaldo , Kimin Lee

Large language models (LMs) beyond a certain scale, demonstrate the emergent capability of generating free-text rationales for their predictions via chain-of-thought (CoT) prompting. While CoT can yield dramatically improved performance,…

Computation and Language · Computer Science 2023-09-01 Peifeng Wang , Zhengyang Wang , Zheng Li , Yifan Gao , Bing Yin , Xiang Ren

Lifelong imitation learning for manipulation tasks poses significant challenges due to distribution shifts that occur in incremental learning steps. Existing methods often focus on unsupervised skill discovery to construct an ever-growing…

Machine Learning · Computer Science 2025-03-10 Kaushik Roy , Akila Dissanayake , Brendan Tidd , Peyman Moghadam

Reasoning distillation has emerged as an effective approach to enhance the reasoning capabilities of smaller language models. However, the impact of large-scale reasoning distillation on other critical abilities, particularly in-context…

Computation and Language · Computer Science 2025-07-22 Yifei Wang

Dataset distillation creates a small distilled set that enables efficient training by capturing key information from the full dataset. While existing dataset distillation methods perform well on balanced datasets, they struggle under…

Computer Vision and Pattern Recognition · Computer Science 2025-12-16 Xiao Cui , Yulei Qin , Xinyue Li , Wengang Zhou , Hongsheng Li , Houqiang Li