研究人员开发了增强流匹配模型（一种生成式AI）的新方法。一种名为“Precise”的方法通过使用与SDE一致的随机采样来改进训练后强化学习，以实现更好的对齐和更快的优化。另一篇论文探讨了用于具身AI轨迹的“稀疏组合流匹配”，直接在物理空间中组合运动原语以提高准确性。一项调查还回顾了用于表格数据的扩散模型和流匹配模型，强调了挑战和未来方向，而其他工作则研究了“转换匹配”作为某些分布的潜在优于流匹配的替代方案，并引入了用于无监督异常检测的“流不匹配”。
AI
arXiv:2605.08063v5 Announce Type: replace-cross Abstract: Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scalar-valued rewards, and the gradient interference arising from jointly opt…
arXiv cs.LG
TIER_1English(EN)·Xuyang Li, Rui Li, Teng Man, Yimin Lu·
arXiv:2604.04453v2 Announce Type: replace-cross Abstract: Granular flows govern many natural and industrial processes, yet their interior kinematics and mechanics remain largely unobservable, as experiments access only boundaries or free surfaces. Conventional numerical simulatio…
arXiv:2512.23956v2 Announce Type: replace-cross Abstract: Flow Matching (FM) has emerged as a powerful paradigm for continuous normalizing flows, yet standard FM implicitly performs an unweighted $L^2$ regression over the entire ambient space. In high dimensions, this leads to a …
arXiv cs.LG
TIER_1English(EN)·Pedro M. P. Curvo, Maksim Zhdanov, Floor Eijkelboom, Jan-Willem van de Meent·
arXiv:2605.10302v3 Announce Type: replace Abstract: Existing approaches to controllable generation typically rely on fine-tuning, auxiliary networks, or test-time search. We show that flow matching admits a different control interface: adaptation through examples. For determinist…
arXiv:2512.20063v3 Announce Type: replace Abstract: We introduce $\texttt{PairFlow}$, a lightweight preprocessing step for training Discrete Flow Models (DFMs) to achieve few-step sampling without requiring a pretrained teacher. DFMs have recently emerged as a new class of genera…
arXiv cs.LG
TIER_1English(EN)·Xifeng Zhang, Jin Zhao·
arXiv:2605.25509v1 Announce Type: cross Abstract: Reconstructing PDE solutions from sparse observations is a core challenge in scientific computing. We present FM4PDE, a flow-matching generative framework that learns the joint distribution of PDE coefficients (or initial states) …
arXiv:2605.26013v1 Announce Type: cross Abstract: We introduce AdvantageFlow, a forward-process reinforcement learning algorithm for rectified flow models. Unlike Flow-GRPO, which optimizes the reverse process, we optimize an advantage-weighted forward-process prediction loss. Th…
We introduce AdvantageFlow, a forward-process reinforcement learning algorithm for rectified flow models. Unlike Flow-GRPO, which optimizes the reverse process, we optimize an advantage-weighted forward-process prediction loss. This optimization problem is unstable when advantage…
arXiv cs.AI
TIER_1English(EN)·Zhong Li, Qi Huang, Lincen Yang, Jiayang Shi, Zhao Yang, Niki van Stein, Thomas B\"ack, Matthijs van Leeuwen·
arXiv:2502.17119v2 Announce Type: replace-cross Abstract: Deep generative models have made rapid progress in image, text, audio, and video generation, and are increasingly being applied to structured records. For tabular data, however, generative modeling remains difficult: a dat…
arXiv:2605.23522v1 Announce Type: cross Abstract: Reinforcement learning (RL) has become an effective way to improve prompt alignment and perceptual quality in diffusion and flow-matching generators. A critical step for applying online RL to flow matching is turning the determini…
arXiv:2605.23341v1 Announce Type: cross Abstract: Embodied trajectories, such as the executable motion sequences of robotic manipulators, underwater vehicles, and mobile robots, are a fundamental output of embodied AI. Modern generative models often treat them as a dense, monolit…
Geometry-aware generative models leveraging spherical manifolds and optimal transport techniques outperform traditional Euclidean approaches for natural image synthesis.
Embodied trajectories, such as the executable motion sequences of robotic manipulators, underwater vehicles, and mobile robots, are a fundamental output of embodied AI. Modern generative models often treat them as a dense, monolithic signal generated point by point, fitting an in…
arXiv cs.AI
TIER_1English(EN)·Jingxuan Wu, Zhenglin Wan, Xingrui Yu, Yuzhe Yang, Bo An, Ivor Tsang, Yang You·
arXiv:2510.09060v2 Announce Type: replace Abstract: Flow-based text-to-image models follow deterministic trajectories, making it costly to explore diverse modes under limited sampling budgets. Existing approaches to improving diversity often rely on retraining or degrade image fi…
arXiv cs.AI
TIER_1English(EN)·Yifu Luo, Haoyuan Sun, Xinhao Hu, Penghui Du, Keyu Fan, Bo Li, Sinan Du, Xu Wan, Zhiyu Chen, Bo Xia, Tiantian Zhang, Yongzhe Chang, Changqian Yu, Kun Gai, Xueqian Wang·
arXiv:2510.21583v2 Announce Type: replace-cross Abstract: Recent Progress in post-training flow matching for text-to-image (T2I) generation with Group Relative Policy Optimization (GRPO) has demonstrated strong potential. However, it is hindered by a critical limitation: inaccura…
arXiv cs.LG
TIER_1English(EN)·Jaihoon Kim, Rajarshi Saha, Minhyuk Sung, Youngsuk Park·
arXiv:2510.17991v3 Announce Type: replace Abstract: Flow Matching (FM) underpins many state-of-the-art generative models, yet recent results indicate that Transition Matching (TM) can achieve higher quality with fewer sampling steps. This work answers the question of when and why…
This paper tackles the task of learning to generate signals over triangle meshes in a triangulation-agnostic manner, meaning the trained model can be applied to different meshes and triangulations effectively. Practically, the paper adapts the flow matching (FM) paradigm to a mes…
Reconstructing PDE solutions from sparse observations is a core challenge in scientific computing. We present FM4PDE, a flow-matching generative framework that learns the joint distribution of PDE coefficients (or initial states) and solutions (or final states), enabling both for…
arXiv:2605.23070v1 Announce Type: new Abstract: We propose Flow Mismatching, an unsupervised anomaly detection method that deliberately avoids reconstruction-based paradigms. Instead, we treat flow matching as geometric dynamics and leverage a key insight: anomalies occur at plac…
Reinforcement learning (RL) has become an effective way to improve prompt alignment and perceptual quality in diffusion and flow-matching generators. A critical step for applying online RL to flow matching is turning the deterministic sampling trajectory into a stochastic policy,…
arXiv:2605.20068v1 Announce Type: new Abstract: Standard generative models struggle with heavy-tailed data: Lipschitz architectures cannot produce power-law tails from Gaussian noise, and interpolating between heavy-tailed data and Gaussians is ill-posed. We propose a simple fix:…
Standard generative models struggle with heavy-tailed data: Lipschitz architectures cannot produce power-law tails from Gaussian noise, and interpolating between heavy-tailed data and Gaussians is ill-posed. We propose a simple fix: apply the soft-log transform $φ(x) = \mathrm{si…
