English
Related papers

Related papers: ORPO-Distill: Mixed-Policy Preference Optimization…

200 papers

Distilling Chain-of-Thought (CoT) reasoning from large language models into compact student models presents a fundamental challenge: teacher rationales are often too verbose for smaller models to faithfully reproduce. Existing approaches…

Machine Learning · Computer Science 2026-03-06 Bowen Yu , Maolin Wang , Sheng Zhang , Binhao Wang , Yi Wen , Jingtong Gao , Bowen Liu , Zimo Zhao , Wanyu Wang , Xiangyu Zhao

Offline reinforcement learning struggles with distributional shift and constrained performance due to static dataset limitations, while online RL demands prohibitive environment interactions. The recent advent of hybrid offline-to-online…

Machine Learning · Computer Science 2026-05-19 Qisai Liu , Zhanhong Jiang , Joshua Russell Waite , Aditya Balu , Cody Fleming , Soumik Sarkar

Knowledge Distillation (KD) aims to transfer knowledge from a large teacher model to a smaller student model. While contrastive learning has shown promise in self-supervised learning by creating discriminative representations, its…

Computer Vision and Pattern Recognition · Computer Science 2025-05-14 Nikolaos Giakoumoglou , Tania Stathaki

Knowledge distillation involves transferring knowledge from large, cumbersome teacher models to more compact student models. The standard approach minimizes the Kullback-Leibler (KL) divergence between the probabilistic outputs of a teacher…

Computer Vision and Pattern Recognition · Computer Science 2025-05-14 Nikolaos Giakoumoglou , Tania Stathaki

Growing efforts to improve knowledge distillation (KD) in large language models (LLMs) replace dense teacher supervision with selective distillation, which uses a subset of token positions, vocabulary classes, or training samples for…

Computation and Language · Computer Science 2026-02-03 Almog Tavor , Itay Ebenspanger , Neil Cnaan , Mor Geva

In the field of large language models (LLMs), aligning models with the diverse preferences of users is a critical challenge. Direct Preference Optimization (DPO) has played a key role in this area. It works by using pairs of preferences…

Computation and Language · Computer Science 2024-05-29 Yueqin Yin , Zhendong Wang , Yi Gu , Hai Huang , Weizhu Chen , Mingyuan Zhou

Knowledge distillation (KD) has shown potential for learning compact models in dense object detection. However, the commonly used softmax-based distillation ignores the absolute classification scores for individual categories. Thus, the…

Computer Vision and Pattern Recognition · Computer Science 2024-03-13 Longrong Yang , Xianpan Zhou , Xuewei Li , Liang Qiao , Zheyang Li , Ziwei Yang , Gaoang Wang , Xi Li

Knowledge Distillation (KD) has emerged as a promising approach for transferring knowledge from a larger, more complex teacher model to a smaller student model. Traditionally, KD involves training the student to mimic the teacher's output…

Machine Learning · Computer Science 2024-10-03 Noel Loo , Fotis Iliopoulos , Wei Hu , Erik Vee

Vision-based policies are widely applied in robotics for tasks such as manipulation and locomotion. On lightweight mobile robots, however, they face a trilemma of limited scene transferability, restricted onboard computation resources, and…

Robotics · Computer Science 2026-03-24 Kai Li , Shiyu Zhao

Solving imaging inverse problems has usually been addressed by designing proper prior models of the underlying signal. However, minimizing the data fidelity term poses significant challenges due to the ill-conditioned sensing matrix caused…

Image and Video Processing · Electrical Eng. & Systems 2026-05-18 Romario Gualdrón-Hurtado , Roman Jacome , Leon Suarez , Henry Arguello

On-policy self-distillation (self-OPD) densifies reinforcement learning with verifiable rewards (RLVR) by letting a policy teach itself under privileged context. We find that when this guidance spans the full response, all-token KL spends…

Artificial Intelligence · Computer Science 2026-05-12 Jiaxuan Wang , Xuan Ouyang , Zhiyu Chen , Yulan Hu , Zheng Pan , Xin Li , Lan-Zhe Guo

Diffusion large language models (dLLMs) offer parallel decoding and bidirectional context, but state-of-the-art dLLMs require billions of parameters for competitive performance. While existing distillation methods for dLLMs reduce inference…

Computation and Language · Computer Science 2026-04-30 Gongbo Zhang , Wen Wang , Ye Tian , Li Yuan

Recent research on knowledge distillation has increasingly focused on logit distillation because of its simplicity, effectiveness, and versatility in model compression. In this paper, we introduce Refined Logit Distillation (RLD) to address…

Computer Vision and Pattern Recognition · Computer Science 2025-07-29 Wujie Sun , Defang Chen , Siwei Lyu , Genlang Chen , Chun Chen , Can Wang

Knowledge distillation has become widely recognized for its ability to transfer knowledge from a large teacher network to a compact and more streamlined student network. Traditional knowledge distillation methods primarily follow a…

Computer Vision and Pattern Recognition · Computer Science 2024-09-30 Chaomin Shen , Yaomin Huang , Haokun Zhu , Jinsong Fan , Guixu Zhang

In modern large language models (LLMs), LLM alignment is of crucial importance and is typically achieved through methods such as reinforcement learning from human feedback (RLHF) and direct preference optimization (DPO). However, in most…

Computation and Language · Computer Science 2025-07-24 Songming Zhang , Xue Zhang , Tong Zhang , Bojie Hu , Yufeng Chen , Jinan Xu

Previous knowledge distillation (KD) methods for object detection mostly focus on feature imitation instead of mimicking the prediction logits due to its inefficiency in distilling the localization information. In this paper, we investigate…

Computer Vision and Pattern Recognition · Computer Science 2022-12-09 Zhaohui Zheng , Rongguang Ye , Qibin Hou , Dongwei Ren , Ping Wang , Wangmeng Zuo , Ming-Ming Cheng

Heterogeneous distillation is an effective way to transfer knowledge from cross-architecture teacher models to student models. However, existing heterogeneous distillation methods do not take full advantage of the dark knowledge hidden in…

Computer Vision and Pattern Recognition · Computer Science 2025-02-11 Yaoxin Yang , Peng Ye , Weihao Lin , Kangcong Li , Yan Wen , Jia Hao , Tao Chen

Alignment techniques enable Large Language Models (LLMs) to generate outputs that align with human preferences and play a crucial role in their effectiveness. However, their impact often diminishes when applied to Small Language Models…

Computation and Language · Computer Science 2025-03-06 Shiping Gao , Fanqi Wan , Jiajian Guo , Xiaojun Quan , Qifan Wang

Post-training of language models, either through reinforcement learning, preference optimization or supervised finetuning, tends to sharpen the output probability distribution and reduce the diversity of generated responses. This is…

Computation and Language · Computer Science 2025-05-23 Jack Lanchantin , Angelica Chen , Shehzaad Dhuliawala , Ping Yu , Jason Weston , Sainbayar Sukhbaatar , Ilia Kulikov

Trust Region Policy Optimization (TRPO) and Proximal Policy Optimization (PPO), as the widely employed policy based reinforcement learning (RL) methods, are prone to converge to a sub-optimal solution as they limit the policy representation…

Machine Learning · Computer Science 2020-06-16 Jun Song , Chaoyue Zhao