English
Related papers

Related papers: Directly Aligning the Full Diffusion Trajectory wi…

200 papers

Generating visually appealing images is fundamental to modern text-to-image generation models. A potential solution to better aesthetics is direct preference optimization (DPO), which has been applied to diffusion models to improve general…

Computer Vision and Pattern Recognition · Computer Science 2025-03-26 Zhanhao Liang , Yuhui Yuan , Shuyang Gu , Bohan Chen , Tiankai Hang , Mingxi Cheng , Ji Li , Liang Zheng

Diffusion models have achieved remarkable progress in text-to-image generation, yet aligning them with human preference remains challenging due to the presence of multiple, sometimes conflicting, evaluation metrics (e.g., semantic…

Computer Vision and Pattern Recognition · Computer Science 2026-04-07 Dipesh Tamboli , Souradip Chakraborty , Aditya Malusare , Biplab Banerjee , Amrit Singh Bedi , Vaneet Aggarwal

Text-to-image diffusion models deliver high-quality images, yet aligning them with human preferences remains challenging. We revisit diffusion-based Direct Preference Optimization (DPO) for these models and identify a critical pathology:…

Computer Vision and Pattern Recognition · Computer Science 2025-12-03 Minghao Fu , Guo-Hua Wang , Tianyu Cui , Qing-Guo Chen , Zhao Xu , Weihua Luo , Kaifu Zhang

Post-training alignment of diffusion models relies on simplified signals, such as scalar rewards or binary preferences. This limits alignment with complex human expertise, which is hierarchical and fine-grained. To address this, we first…

Computer Vision and Pattern Recognition · Computer Science 2026-01-09 Chenye Meng , Zejian Li , Zhongni Liu , Yize Li , Changle Xie , Kaixin Jia , Ling Yang , Huanghuang Deng , Shiying Ding , Shengyuan Zhang , Jiayi Li , Lingyun Sun

Latent diffusion models are the state-of-the-art for synthetic image generation. To align these models with human preferences, training the models using reinforcement learning on human feedback is crucial. Black et. al 2024 introduced…

Machine Learning · Computer Science 2024-04-09 Mo Kordzanganeh , Danial Keshvary , Nariman Arian

Direct Preference Optimization (DPO) has been successfully used to align large language models (LLMs) according to human preferences, and more recently it has also been applied to improving the quality of text-to-image diffusion models.…

Computer Vision and Pattern Recognition · Computer Science 2025-10-07 Shivanshu Shekhar , Shreyas Singh , Tong Zhang

The incorporation of online reinforcement learning (RL) into diffusion and flow-based generative models has recently gained attention as a powerful paradigm for aligning model behavior with human preferences. By leveraging stochastic…

Machine Learning · Computer Science 2025-11-25 Yujie Zhou , Pengyang Ling , Jiazi Bu , Yibin Wang , Yuhang Zang , Jiaqi Wang , Li Niu , Guangtao Zhai

Bridging the gap between diffusion models and human preferences is crucial for their integration into practical generative workflows. While optimizing downstream reward models has emerged as a promising alignment strategy, concerns arise…

Machine Learning · Computer Science 2026-03-02 Ziyi Zhang , Sen Zhang , Yibing Zhan , Yong Luo , Yonggang Wen , Dacheng Tao

Language model (LM) post-training (or alignment) involves maximizing a reward function that is derived from preference annotations. Direct Preference Optimization (DPO) is a popular offline alignment method that trains a policy directly on…

Machine Learning · Computer Science 2025-03-04 Adam Fisch , Jacob Eisenstein , Vicky Zayats , Alekh Agarwal , Ahmad Beirami , Chirag Nagpal , Pete Shaw , Jonathan Berant

We present Direct Reward Fine-Tuning (DRaFT), a simple and effective method for fine-tuning diffusion models to maximize differentiable reward functions, such as scores from human preference models. We first show that it is possible to…

Computer Vision and Pattern Recognition · Computer Science 2024-06-24 Kevin Clark , Paul Vicol , Kevin Swersky , David J Fleet

Recent GRPO-based approaches built on flow matching models have shown remarkable improvements in human preference alignment for text-to-image generation. Nevertheless, they still suffer from the sparse reward problem: the terminal reward of…

Computer Vision and Pattern Recognition · Computer Science 2026-02-26 Haoyou Deng , Keyu Yan , Chaojie Mao , Xiang Wang , Yu Liu , Changxin Gao , Nong Sang

Diffusion models are a class of flexible generative models trained with an approximation to the log-likelihood objective. However, most use cases of diffusion models are not concerned with likelihoods, but instead with downstream objectives…

Machine Learning · Computer Science 2024-01-08 Kevin Black , Michael Janner , Yilun Du , Ilya Kostrikov , Sergey Levine

Reinforcement learning from human feedback (RLHF) has proven effectiveness for aligning text-to-image (T2I) diffusion models with human preferences. Although Direct Preference Optimization (DPO) is widely adopted for its computational…

Computer Vision and Pattern Recognition · Computer Science 2025-10-03 Jiamu Bai , Xin Yu , Meilong Xu , Weitao Lu , Xin Pan , Kiwan Maeng , Daniel Kifer , Jian Wang , Yu Wang

Direct Preference Optimization (DPO) has been proposed as an effective and efficient alternative to reinforcement learning from human feedback (RLHF). In this paper, we propose a novel and enhanced version of DPO based on curriculum…

Computer Vision and Pattern Recognition · Computer Science 2025-05-12 Florinel-Alin Croitoru , Vlad Hondru , Radu Tudor Ionescu , Nicu Sebe , Mubarak Shah

Current mainstream methods of aligning diffusion models with human preferences typically employ VLM-based reward models. However, these reward models, pre-trained for semantic alignment, struggle to capture the essential perceptual…

Computer Vision and Pattern Recognition · Computer Science 2026-05-26 Jaxon Zhang , Binxin Yang , Hubery Yin , Chen Li , Jing Lyu

Direct Preference Optimization (DPO) has recently been applied as a post-training technique for text-to-video diffusion models. To obtain training data, annotators are asked to provide preferences between two videos generated from…

Computer Vision and Pattern Recognition · Computer Science 2025-10-13 Ziyi Wu , Anil Kag , Ivan Skorokhodov , Willi Menapace , Ashkan Mirzaei , Igor Gilitschenski , Sergey Tulyakov , Aliaksandr Siarohin

Preference learning has garnered extensive attention as an effective technique for aligning diffusion models with human preferences in visual generation. However, existing alignment approaches such as Diffusion-DPO suffer from two…

Machine Learning · Computer Science 2026-05-19 Xiaomeng Yang , Mengping Yang , Junyan Wang , Zhijian Zhou , Zhiyu Tan , Hao Li

Recent research has shown that fine-tuning diffusion models (DMs) with arbitrary rewards, including non-differentiable ones, is feasible with reinforcement learning (RL) techniques, enabling flexible model alignment. However, applying…

Machine Learning · Computer Science 2025-03-12 Zhiwei Jia , Yuesong Nan , Huixi Zhao , Gengdai Liu

Aligning text-to-image (T2I) diffusion models with preference optimization is valuable for human-annotated datasets, but the heavy cost of manual data collection limits scalability. Using reward models offers an alternative, however,…

Computer Vision and Pattern Recognition · Computer Science 2025-09-29 Kyungmin Lee , Xiaohang Li , Qifei Wang , Junfeng He , Junjie Ke , Ming-Hsuan Yang , Irfan Essa , Jinwoo Shin , Feng Yang , Yinxiao Li

In recent years, motion generative models have undergone significant advancement, yet pose challenges in aligning with downstream objectives. Recent studies have shown that using differentiable rewards to directly align the preference of…

Computer Vision and Pattern Recognition · Computer Science 2026-02-10 Xiaofeng Tan , Wanjiang Weng , Haodong Lei , Hongsong Wang