English
Related papers

Related papers: CreFlow: Corrective Reflow for Sparse-Reward Embod…

200 papers

Perceptual studies demonstrate that conditional diffusion models excel at reconstructing video content aligned with human visual perception. Building on this insight, we propose a video compression framework that leverages conditional…

Computer Vision and Pattern Recognition · Computer Science 2025-09-26 Fangqiu Yi , Jingyu Xu , Jiawei Shao , Chi Zhang , Xuelong Li

Recent advancements in reinforcement learning (RL) demonstrate the significant potential in autonomous driving. Despite this promise, challenges such as the manual design of reward functions and low sample efficiency in complex environments…

Robotics · Computer Science 2025-01-10 Zengqi Peng , Yubin Wang , Xu Han , Lei Zheng , Jun Ma

Single-task RL agents are typically trained under a fixed reward function, which limits their robustness to reward misspecification and their ability to adapt to changing preferences. We introduce Reward-Conditioned Reinforcement Learning…

Machine Learning · Computer Science 2026-05-20 Michal Nauman , Marek Cygan , Pieter Abbeel

Advanced reasoning in LLMs on challenging domains like mathematical reasoning can be tackled using verifiable rewards based reinforced fine-tuning (ReFT). In standard ReFT frameworks, a behavior model generates multiple completions with…

Machine Learning · Computer Science 2025-11-25 Maxime Heuillet , Yufei Cui , Boxing Chen , Audrey Durand , Prasanna Parthasarathi

Video generation has recently emerged as a central task in the field of generative AI. However, the substantial computational cost inherent in video synthesis makes model distillation a critical technique for efficient deployment. Despite…

Computer Vision and Pattern Recognition · Computer Science 2026-03-24 Yuyang You , Yongzhi Li , Jiahui Li , Yadong Mu , Quan Chen , Peng Jiang

Machine unlearning in text-to-image diffusion models aims to remove targeted concepts while preserving overall utility. Prior diffusion unlearning methods typically rely on supervised weight edits or global penalties; reinforcement-learning…

Machine Learning · Computer Science 2026-02-17 Mykola Vysotskyi , Zahar Kohut , Mariia Shpir , Taras Rumezhak , Volodymyr Karpiv

Visual generation is dominated by three paradigms: AutoRegressive (AR), diffusion, and Visual AutoRegressive (VAR) models. Unlike AR and diffusion, VARs operate on heterogeneous input structures across their generation steps, which creates…

Computer Vision and Pattern Recognition · Computer Science 2026-01-06 Shikun Sun , Liao Qu , Huichao Zhang , Yiheng Liu , Yangyang Song , Xian Li , Xu Wang , Yi Jiang , Daniel K. Du , Xinglong Wu , Jia Jia

Reward engineering has long been a challenge in Reinforcement Learning (RL) research, as it often requires extensive human effort and iterative processes of trial-and-error to design effective reward functions. In this paper, we propose…

Robotics · Computer Science 2024-06-18 Yufei Wang , Zhanyi Sun , Jesse Zhang , Zhou Xian , Erdem Biyik , David Held , Zackory Erickson

Recent advancements in diffusion models (DMs) have been propelled by alignment methods that post-train models to better conform to human preferences. However, these approaches typically require computation-intensive training of a base model…

Computer Vision and Pattern Recognition · Computer Science 2025-08-05 Zejian Li , Yize Li , Chenye Meng , Zhongni Liu , Yang Ling , Shengyuan Zhang , Guang Yang , Changyuan Yang , Zhiyuan Yang , Lingyun Sun

We study behavior-regularized reinforcement learning (RL), where regularization toward a reference distribution (the dataset in offline RL or the base model in LLM RL finetuning) is essential to prevent value over-optimization caused by…

Machine Learning · Computer Science 2026-04-17 Haoran Xu , Kaiwen Hu , Somayeh Sojoudi , Amy Zhang

Reinforcement Learning (RL) post-training has become the standard for aligning generative models with human preferences, yet most methods rely on a single scalar reward. When multiple criteria matter, the prevailing practice of ``early…

Machine Learning · Computer Science 2026-04-23 Shelly Golan , Michael Finkelson , Ariel Bereslavsky , Yotam Nitzan , Or Patashnik

Diffusion models have demonstrated strong generative capabilities across scientific domains, but often produce outputs that violate physical laws. We propose a new perspective by framing physics-informed generation as a sparse reward…

Machine Learning · Computer Science 2025-09-26 Mingze Yuan , Pengfei Jin , Na Li , Quanzheng Li

Diffusion models have shown remarkable progress in text-to-audio generation. However, text-guided audio editing remains in its early stages. This task focuses on modifying the target content within an audio signal while preserving the rest,…

Sound · Computer Science 2026-04-17 Liting Gao , Yi Yuan , Yaru Chen , Yuelan Cheng , Zhenbo Li , Juan Wen , Shubin Zhang , Wenwu Wang

Instruction-guided image editing has achieved remarkable progress, yet current models still face challenges with complex instructions and often require multiple samples to produce a desired result. Reinforcement Learning (RL) offers a…

Computer Vision and Pattern Recognition · Computer Science 2026-02-03 Xin Luo , Jiahao Wang , Chenyuan Wu , Shitao Xiao , Xiyan Jiang , Defu Lian , Jiajun Zhang , Dong Liu , Zheng liu

Model-based reinforcement learning (MBRL) typically relies on modeling environment dynamics for data efficiency. However, due to the accumulation of model errors over long-horizon rollouts, such methods often face challenges in maintaining…

Machine Learning · Computer Science 2026-01-06 Bin Wang , Boxiang Tao , Haifeng Jing , Hongbo Dou , Zijian Wang

In many RL applications, ensuring an agent's actions adhere to constraints is crucial for safety. Most previous methods in Action-Constrained Reinforcement Learning (ACRL) employ a projection layer after the policy network to correct the…

Machine Learning · Computer Science 2025-02-18 Janaka Chathuranga Brahmanage , Jiajing Ling , Akshat Kumar

Reinforcement learning (RL) post-training is crucial for aligning generative models with human preferences, but its prohibitive computational cost remains a major barrier to widespread adoption. We introduce \textbf{TreeGRPO}, a novel RL…

Machine Learning · Computer Science 2025-12-10 Zheng Ding , Weirui Ye

The post-training pipeline for diffusion models currently has two stages: supervised fine-tuning (SFT) on curated data and reinforcement learning (RL) with reward models. A fundamental gap separates them. SFT optimizes the denoiser only on…

Machine Learning · Computer Science 2026-04-20 You Qin , Linqing Wang , Hao Fei , Roger Zimmermann , Liefeng Bo , Qinglin Lu , Chunyu Wang

Effective tool use is essential for large language models (LLMs) to interact with their environment. However, progress is limited by the lack of efficient reinforcement learning (RL) frameworks specifically designed for tool use, due to…

Computation and Language · Computer Science 2026-04-16 Junjie Ye , Changhao Jiang , Zhengyin Du , Yufei Xu , Xuesong Yao , Zhiheng Xi , Xiaoran Fan , Qi Zhang , Tao Gui , Xuanjing Huang , Jiecao Chen

We address the problem of fine-tuning diffusion models for reward-guided generation in biomolecular design. While diffusion models have proven highly effective in modeling complex, high-dimensional data distributions, real-world…