English
Related papers

Related papers: Flow-OPD: On-Policy Distillation for Flow Matching…

200 papers

Standard knowledge distillation for autoregressive models often suffers from distribution mismatch. While on-policy methods mitigate this by leveraging student-generated outputs, they rely on computationally expensive Reinforcement Learning…

Machine Learning · Computer Science 2026-05-08 Miao Rang , Zhenni Bi , Hang Zhou , Kai Han , Xuechun Wang , An Xiao , Xinghao Chen , Yunhe Wang , Hanting Chen

On-policy distillation (OPD) is a powerful paradigm for model alignment, yet its reliance on teacher logits restricts its application to white-box scenarios. We contend that structured semantic rubrics can serve as a scalable alternative to…

Machine Learning · Computer Science 2026-05-11 Junfeng Fang , Zhepei Hong , Mao Zheng , Mingyang Song , Gengsheng Li , Houcheng Jiang , Dan Zhang , Haiyun Guo , Xiang Wang , Tat-Seng Chua

Flow matching has emerged as a powerful generative framework, with recent few-step methods achieving remarkable inference acceleration. However, we identify a critical yet overlooked limitation: these models suffer from severe diversity…

Machine Learning · Computer Science 2026-04-15 Yexiong Lin , Jia Shi , Shanshan Ye , Wanyu Wang , Yu Yao , Tongliang Liu

Reinforcement learning for large language models faces a fundamental trade-off between sample efficiency and asymptotic performance: strictly on-policy methods discard trajectories after a single update, while off-policy reuse introduces…

Machine Learning · Computer Science 2026-05-26 Changyu Chen , Xiting Wang , Rui Yan

The performance of flow matching and diffusion models can be greatly improved at inference time using reward alignment algorithms, yet efficiency remains a major limitation. While several algorithms were proposed, we demonstrate that a…

Machine Learning · Computer Science 2026-02-12 Peter Holderrieth , Uriel Singer , Tommi Jaakkola , Ricky T. Q. Chen , Yaron Lipman , Brian Karrer

On-policy distillation is pivotal for transferring reasoning capabilities to capacity-constrained models, yet remains prone to instability and negative transfer. We show that on-policy distillation can be interpreted, both theoretically and…

Machine Learning · Computer Science 2026-03-13 Jongwoo Ko , Sara Abdali , Young Jin Kim , Tianyi Chen , Pashmina Cameron

On-Policy Distillation (OPD) has gained wide attraction as an LLM post-training paradigm due to its effectiveness in improving capabilities without introducing model distribution drift, and consequently, regression in general tasks.…

Artificial Intelligence · Computer Science 2026-05-25 Aristotelis Lazaridis , Dylan Bates , Aman Sharma , Brian King , Vincent Lu , Jack FitzGerald

On-Policy Self-Distillation (OPSD) is a unified learning framework in which a single large language model acts simultaneously as both teacher and student. Unlike conventional knowledge distillation that relies on a separate, often larger…

Human-Computer Interaction · Computer Science 2026-05-22 Fangming Cui , Sunan Li , Jiahong Li

Large-scale flow matching models have achieved strong performance across generative tasks such as text-to-image, video, 3D, and speech synthesis. However, aligning their outputs with human preferences and task-specific objectives remains…

Machine Learning · Computer Science 2026-03-10 Zexiang Liu , Xianglong He , Yangguang Li

Post-training has become central to turning pretrained large language models (LLMs) into aligned, capable, and deployable systems. Recent progress spans supervised fine-tuning (SFT), preference optimization, reinforcement learning (RL),…

Computation and Language · Computer Science 2026-04-17 Shiwan Zhao , Zhihu Wang , Xuyang Zhao , Jiaming Zhou , Caiyue Xu , Chenfei Liu , Liting Zhang , Yuhang Jia , Yanzhe Zhang , Hualong Yu , Zichen Xu , Qicheng Li , Yong Qin

Recent advancements in reinforcement learning (RL) have achieved great success in fine-tuning diffusion-based generative models. However, fine-tuning continuous flow-based generative models to align with arbitrary user-defined reward…

Machine Learning · Computer Science 2025-02-11 Jiajun Fan , Shuaike Shen , Chaoran Cheng , Yuxin Chen , Chumeng Liang , Ge Liu

Diffusion Models have emerged as a leading class of generative models, yet their iterative sampling process remains computationally expensive. Timestep distillation is a promising technique to accelerate generation, but it often requires…

Computer Vision and Pattern Recognition · Computer Science 2025-11-26 Guanjie Chen , Shirui Huang , Kai Liu , Jianchen Zhu , Xiaoye Qu , Peng Chen , Yu Cheng , Yifu Sun

Flow-matching models have enabled high-quality text-to-speech synthesis, but their iterative sampling process during inference incurs substantial computational cost. Although distillation is widely used to reduce the number of inference…

Sound · Computer Science 2026-02-11 Bin Lin , Peng Yang , Chao Yan , Xiaochen Liu , Wei Wang , Boyong Wu , Pengfei Tan , Xuerui Yang

Diffusion- and flow-based models have emerged as state-of-the-art generative modeling approaches, but they require many sampling steps. Consistency models can distill these models into efficient one-step generators; however, unlike flow-…

Computer Vision and Pattern Recognition · Computer Science 2025-06-18 Amirmojtaba Sabour , Sanja Fidler , Karsten Kreis

On-policy distillation (OPD) has become a promising paradigm for reasoning-oriented post-training of large language models (LLMs), especially when combined with reinforcement learning from verifiable rewards (RLVR). Existing OPD methods…

Knowledge distillation improves large language model (LLM) reasoning by compressing the knowledge of a teacher LLM to train smaller LLMs. On-policy distillation advances this approach by having the student sample its own trajectories while…

Machine Learning · Computer Science 2026-03-23 Siyan Zhao , Zhihui Xie , Mengchen Liu , Jing Huang , Guan Pang , Feiyu Chen , Aditya Grover

Generative models, especially diffusion and flow-based models, have been promising in offline multi-agent reinforcement learning. However, integrating powerful generative models into this framework poses unique challenges. In particular,…

Machine Learning · Computer Science 2026-03-02 Zhuoran Li , Xun Wang , Hai Zhong , Qingxin Xia , Lihua Zhang , Longbo Huang

Tool-integrated reasoning (TIR) is difficult to scale to small language models due to instability in long-horizon tool interactions and limited model capacity. While reinforcement learning methods like group relative policy optimization…

Computation and Language · Computer Science 2026-05-11 Qiyong Zhong , Mao Zheng , Mingyang Song , Xin Lin , Jie Sun , Houcheng Jiang , Xiang Wang , Junfeng Fang

Distribution Matching Distillation (DMD) distills score-based generative models into efficient one-step generators, without requiring a one-to-one correspondence with the sampling trajectories of their teachers. Yet, the limited capacity of…

Computer Vision and Pattern Recognition · Computer Science 2026-03-26 Xiangyu Fan , Zesong Qiu , Zhuguanyu Wu , Fanzhou Wang , Zhiqian Lin , Tianxiang Ren , Dahua Lin , Ruihao Gong , Lei Yang

Flow Matching (FM) has shown remarkable ability in modeling complex distributions and achieves strong performance in offline imitation learning for cloning expert behaviors. However, despite its behavioral cloning expressiveness, FM-based…

Machine Learning · Computer Science 2025-10-14 Zhenglin Wan , Jingxuan Wu , Xingrui Yu , Chubin Zhang , Mingcong Lei , Bo An , Ivor Tsang