English
Related papers

Related papers: Confronting Reward Overoptimization for Diffusion …

200 papers

Learning from human feedback has been shown to improve text-to-image models. These techniques first learn a reward function that captures what humans care about in the task and then improve the models based on the learned reward function.…

We propose Diffusion Noise Optimization (DNO), a new method that effectively leverages existing motion diffusion models as motion priors for a wide range of motion-related tasks. Instead of training a task-specific diffusion model for each…

Computer Vision and Pattern Recognition · Computer Science 2024-04-04 Korrawe Karunratanakul , Konpat Preechakul , Emre Aksan , Thabo Beeler , Supasorn Suwajanakorn , Siyu Tang

Diffusion large language models (dLLMs) are promising alternatives to autoregressive large language models (AR-LLMs), as they potentially allow higher inference throughput. Reinforcement learning (RL) is a crucial component for dLLMs to…

Machine Learning · Computer Science 2026-02-24 Yuchen Zhu , Wei Guo , Jaemoo Choi , Petr Molodyk , Bo Yuan , Molei Tao , Yongxin Chen

Recent advancements adopt online reinforcement learning (RL) from LLMs to text-to-image rectified flow diffusion models for reward alignment. The use of group-level rewards successfully aligns the model with the targeted reward. However, it…

Machine Learning · Computer Science 2026-01-06 Yiyang Wang , Xi Chen , Xiaogang Xu , Yu Liu , Hengshuang Zhao

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

Diffusion models have demonstrated empirical successes in various applications and can be adapted to task-specific needs via guidance. This paper studies a form of gradient guidance for adapting a pre-trained diffusion model towards…

Machine Learning · Statistics 2024-10-17 Yingqing Guo , Hui Yuan , Yukang Yang , Minshuo Chen , Mengdi Wang

Preference alignment in diffusion models has primarily focused on benign human preferences (e.g., aesthetic). In this paper, we propose a novel perspective: framing unrestricted adversarial example generation as a problem of aligning with…

Computer Vision and Pattern Recognition · Computer Science 2025-06-03 Kaixun Jiang , Zhaoyu Chen , Haijing Guo , Jinglun Li , Jiyuan Fu , Pinxue Guo , Hao Tang , Bo Li , Wenqiang Zhang

Fine-tuning diffusion policies with reinforcement learning (RL) presents significant challenges. The long denoising sequence for each action prediction impedes effective reward propagation. Moreover, standard RL methods require millions of…

The widespread application of large language models (LLMs) raises increasing demands on ensuring safety or imposing constraints, such as reducing harmful content and adhering to predefined rules. While there have been several works studying…

Machine Learning · Computer Science 2026-02-13 Yihan Du , Seo Taek Kong , R. Srikant

Diffusion models and flow matching have demonstrated remarkable success in text-to-image generation. While many existing alignment methods primarily focus on fine-tuning pre-trained generative models to maximize a given reward function,…

Machine Learning · Statistics 2026-02-03 Yidong Ouyang , Liyan Xie , Hongyuan Zha , Guang Cheng

Diffusion models have been successfully adapted to text generation tasks by mapping the discrete text into the continuous space. However, there exist nonnegligible gaps between training and inference, owing to the absence of the forward…

Computation and Language · Computer Science 2023-05-09 Zecheng Tang , Pinzheng Wang , Keyan Zhou , Juntao Li , Ziqiang Cao , Min Zhang

As a typical open-ended generation task, creative writing lacks verifiable reference answers, which has long constrained reward modeling and automatic evaluation due to high human annotation costs, evaluative bias, and coarse feedback…

Computation and Language · Computer Science 2026-03-17 Jihao Zhao , Shuaishuai Zu , Zhiyuan Ji , Chunlai Zhou , Biao Qin

Fine-tuning diffusion models via online reinforcement learning (RL) has shown great potential for enhancing text-to-image alignment. However, since precisely specifying a ground-truth objective for visual tasks remains challenging, the…

Machine Learning · Computer Science 2026-01-01 Haoran He , Yuxiao Ye , Jie Liu , Jiajun Liang , Zhiyong Wang , Ziyang Yuan , Xintao Wang , Hangyu Mao , Pengfei Wan , Ling Pan

Traditional preference tuning methods for LLMs/Visual Generative Models often rely solely on reward model labeling, which can be opaque, offer limited insights into the rationale behind preferences, and are prone to issues such as reward…

Machine Learning · Computer Science 2026-01-13 Hanyang Zhao , Haoxian Chen , Yucheng Guo , Genta Indra Winata , Tingting Ou , Ziyu Huang , David D. Yao , Wenpin Tang

Diffusion models have been widely studied for removing unsafe content learned during pre-training. Existing methods require expensive supervised data, either unsafe-text paired with safe-image groundtruth or negative/positive image pairs,…

Computer Vision and Pattern Recognition · Computer Science 2026-05-19 Komal Kumar , Ankan Deria , Abhishek Basu , Fahad Shamshad , Hisham Cholakkal , Karthik Nandakumar

The increasing prominence of temporal networks in online social platforms and dynamic communication systems has made influence maximization a critical research area. Various diffusion models have been proposed to capture the spread of…

Social and Information Networks · Computer Science 2025-07-31 Aaqib Zahoor , Iqra Altaf Gillani , Janibul Bashir

Large language models are typically aligned with human preferences by optimizing $\textit{reward models}$ (RMs) fitted to human feedback. However, human preferences are multi-faceted, and it is increasingly common to derive reward from a…

A common paradigm to improve the performance of large language models is optimizing for a reward model. Reward models assign a numerical score to an LLM's output that indicates, for example, how likely it is to align with user preferences…

Machine Learning · Computer Science 2025-11-06 Hadi Khalaf , Claudio Mayrink Verdun , Alex Oesterling , Himabindu Lakkaraju , Flavio du Pin Calmon

Reward-based alignment methods for large language models (LLMs) face two key limitations: vulnerability to reward hacking, where models exploit flaws in the reward signal; and reliance on brittle, labor-intensive prompt engineering when…

Computation and Language · Computer Science 2025-05-20 Zae Myung Kim , Chanwoo Park , Vipul Raheja , Suin Kim , Dongyeop Kang

Reinforcement Learning (RL) models have continually evolved to navigate the exploration - exploitation trade-off in uncertain Markov Decision Processes (MDPs). In this study, I leverage the principles of stochastic thermodynamics and system…

Machine Learning · Computer Science 2023-06-22 Peeyush Kumar