English
Related papers

Related papers: Reward Score Matching: Unifying Reward-based Fine-…

200 papers

We introduce iterative tilting, a gradient-free method for fine-tuning diffusion models toward reward-tilted distributions. The method decomposes a large reward tilt $\exp(\lambda r)$ into $N$ sequential smaller tilts, each admitting a…

Machine Learning · Statistics 2025-12-04 Jean Pachebat , Giovanni Conforti , Alain Durmus , Yazid Janati

Visual-motor policy learning has advanced with architectures like diffusion-based policies, known for modeling complex robotic trajectories. However, their prolonged inference times hinder high-frequency control tasks requiring real-time…

Robotics · Computer Science 2024-12-20 Bofang Jia , Pengxiang Ding , Can Cui , Mingyang Sun , Pengfang Qian , Siteng Huang , Zhaoxin Fan , Donglin Wang

Fine-tuning Diffusion Models remains an underexplored frontier in generative artificial intelligence (GenAI), especially when compared with the remarkable progress made in fine-tuning Large Language Models (LLMs). While cutting-edge…

Machine Learning · Computer Science 2024-02-16 Huizhuo Yuan , Zixiang Chen , Kaixuan Ji , Quanquan Gu

Learning rewards from expert videos offers an affordable and effective solution to specify the intended behaviors for reinforcement learning (RL) tasks. In this work, we propose Diffusion Reward, a novel framework that learns rewards from…

Machine Learning · Computer Science 2024-08-12 Tao Huang , Guangqi Jiang , Yanjie Ze , Huazhe Xu

We present a mechanism to steer the sampling diversity of denoising diffusion and flow matching models, allowing users to sample from a sharper or broader distribution than the training distribution. We build on the observation that these…

Machine Learning · Computer Science 2026-05-26 Yanbo Xu , Yu Wu , Sungjae Park , Zhizhuo Zhou , Shubham Tulsiani

Recent works on large language models (LLMs) have successfully demonstrated the emergence of reasoning capabilities via reinforcement learning (RL). Although recent efforts leverage group relative policy optimization (GRPO) for MLLMs…

Computation and Language · Computer Science 2025-06-18 Shilin Xu , Yanwei Li , Rui Yang , Tao Zhang , Yueyi Sun , Wei Chow , Linfeng Li , Hang Song , Qi Xu , Yunhai Tong , Xiangtai Li , Hao Fei

Foundation models are first pre-trained on vast unsupervised datasets and then fine-tuned on labeled data. Reinforcement learning, notably from human feedback (RLHF), can further align the network with the intended usage. Yet the…

Recent advances in one-step text-to-image generation have enabled real-time synthesis with remarkable efficiency and quality. Previous reinforcement learning methods for one-step generators combine image-space reward optimization with…

Computer Vision and Pattern Recognition · Computer Science 2026-05-27 Junyi Wu , Weijian Luo , Haoyang Zheng , Ruizhe Zhang , Guang Lin

Controlling generative models is computationally expensive. This is because optimal alignment with a reward function--whether via inference-time steering or fine-tuning--requires estimating the value function. This task demands access to…

Machine Learning · Statistics 2026-01-22 Peter Potaptchik , Adhi Saravanan , Abbas Mammadov , Alvaro Prat , Michael S. Albergo , Yee Whye Teh

While foundation models have been exploited for various expert tasks through fine-tuning, any foundation model will become outdated due to its old knowledge or limited capability. Thus the underlying foundation model should be eventually…

Machine Learning · Computer Science 2025-02-19 Daiki Chijiwa , Taku Hasegawa , Kyosuke Nishida , Kuniko Saito , Susumu Takeuchi

Reinforcement Learning (RL) has proven highly effective in aligning Large Language Models (LLMs) with human preferences. Typical RL methods optimize under an overall sequence reward, which can lead to a suboptimal learning process. This…

Machine Learning · Computer Science 2025-02-26 Yanshi Li , Shaopan Xiong , Gengru Chen , Xiaoyang Li , Yijia Luo , Xingyuan Bu , Yingshui Tan , Wenbo Su , Bo Zheng

The recent success of denoising diffusion models has significantly advanced text-to-image generation. While these large-scale pretrained models show excellent performance in general image synthesis, downstream objectives often require…

Computer Vision and Pattern Recognition · Computer Science 2024-11-27 Maorong Wang , Jiafeng Mao , Xueting Wang , Toshihiko Yamasaki

Reward function is essential in reinforcement learning (RL), serving as the guiding signal to incentivize agents to solve given tasks, however, is also notoriously difficult to design. In many cases, only imperfect rewards are available,…

Machine Learning · Computer Science 2023-02-06 Jianxiong Li , Xiao Hu , Haoran Xu , Jingjing Liu , Xianyuan Zhan , Qing-Shan Jia , Ya-Qin Zhang

Flow Matching (FM) models achieve remarkable results in generative tasks. Building upon diffusion models, FM's simulation-free training paradigm enables simplicity and efficiency but introduces a train-inference gap: model outputs cannot be…

Machine Learning · Computer Science 2026-01-30 Zhaoyi Li , Jingtao Ding , Yong Li , Shihua Li

Fine-tuning text-to-image diffusion models to maximize rewards has proven effective for enhancing model performance. However, reward fine-tuning methods often suffer from slow convergence due to online sample generation. Therefore,…

Computer Vision and Pattern Recognition · Computer Science 2025-02-21 Daewon Chae , June Suk Choi , Jinkyu Kim , Kimin Lee

We introduce Self-supervised Online Reward Shaping (SORS), which aims to improve the sample efficiency of any RL algorithm in sparse-reward environments by automatically densifying rewards. The proposed framework alternates between…

Machine Learning · Computer Science 2021-07-27 Farzan Memarian , Wonjoon Goo , Rudolf Lioutikov , Scott Niekum , Ufuk Topcu

Data heterogeneity hinders clinical deployment of medical image analysis models, and generative data augmentation helps mitigate this issue. However, recent diffusion-based methods that synthesize image-mask pairs often ignore distribution…

Image and Video Processing · Electrical Eng. & Systems 2026-04-06 Jie Yang , Ziqi Ye , Aihua Ke , Jian Luo , Bo Cai , Xiaosong Wang

Reward models are central to aligning large language models (LLMs) with human preferences. Yet most approaches rely on pointwise reward estimates that overlook the epistemic uncertainty in reward models arising from limited human feedback.…

Machine Learning · Computer Science 2026-03-02 Daniel Yang , Samuel Stante , Florian Redhardt , Lena Libon , Parnian Kassraie , Ido Hakimi , Barna Pásztor , Andreas Krause

Recent advances in large language models (LLMs) have significantly improved performance on the Text-to-SQL task by leveraging their powerful reasoning capabilities. To enhance accuracy during the reasoning process, external Process Reward…

Computation and Language · Computer Science 2025-05-20 Yuxin Zhang , Meihao Fan , Ju Fan , Mingyang Yi , Yuyu Luo , Jian Tan , Guoliang Li

Reinforcement fine-tuning played an instrumental role in enhancing the instruction-following and reasoning abilities of large language models. In this work, we employ reinforcement fine-tuning for materials design, in which discriminative…

Materials Science · Physics 2026-01-19 Zhendong Cao , Lei Wang