Cover Image

Rethinking the Design Space of Reinforcement Learning for Diffusion Models: On the Importance of Likelihood Estimation Beyond Loss Design

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Choi, Jaemoo, Zhu, Yuchen, Guo, Wei, Molodyk, Petr, Yuan, Bo, Bai, Jinbin, Xin, Yi, Tao, Molei, Chen, Yongxin
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Enhancing Reasoning for Diffusion LLMs via Distribution Matching Policy Optimization
by: Zhu, Yuchen, et al.
Published: (2025)

Proximal Diffusion Neural Sampler
by: Guo, Wei, et al.
Published: (2025)

MDNS: Masked Diffusion Neural Sampler via Stochastic Optimal Control
by: Zhu, Yuchen, et al.
Published: (2025)

MFM-point: Multi-scale Flow Matching for Point Cloud Generation
by: Molodyk, Petr, et al.
Published: (2025)

Non-equilibrium Annealed Adjoint Sampler
by: Choi, Jaemoo, et al.
Published: (2025)

Complexity Analysis of Normalizing Constant Estimation: from Jarzynski Equality to Annealed Importance Sampling and beyond
by: Guo, Wei, et al.
Published: (2025)

Clarify Before You Draw: Proactive Agents for Robust Text-to-CAD Generation
by: Yuan, Bo, et al.
Published: (2026)

RMA: an Agentic System for Research-Level Mathematical Problems
by: Zhao, Zelin, et al.
Published: (2026)

Improving Neural Optimal Transport via Displacement Interpolation
by: Choi, Jaemoo, et al.
Published: (2024)

Plug-and-Play Controllable Generation for Discrete Masked Models
by: Guo, Wei, et al.
Published: (2024)

Laplacian Multi-scale Flow Matching for Generative Modeling
by: Zhao, Zelin, et al.
Published: (2026)

Discrete Adjoint Schrödinger Bridge Sampler
by: Guo, Wei, et al.
Published: (2026)

Efficient Adjoint Matching for Fine-tuning Diffusion Models
by: Shin, Jeongwoo, et al.
Published: (2026)

Provable Benefit of Annealed Langevin Monte Carlo for Non-log-concave Sampling
by: Guo, Wei, et al.
Published: (2024)

Coarse-to-Fine Compositional Diffusion for Long-Horizon Planning
by: Park, Byoungwoo, et al.
Published: (2026)

Diffuse Everything: Multimodal Diffusion Models on Arbitrary State Spaces
by: Rojas, Kevin, et al.
Published: (2025)

Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models
by: Bai, Jinbin, et al.
Published: (2026)

MetaDNS: Enhancing Exploration in Discrete Neural Samplers via Well-Tempered Metadynamics
by: Du, Xiaochen, et al.
Published: (2026)

Robust Barycenter Estimation using Semi-Unbalanced Neural Optimal Transport
by: Gazdieva, Milena, et al.
Published: (2024)

Scalable Simulation-free Entropic Unbalanced Optimal Transport
by: Choi, Jaemoo, et al.
Published: (2024)

Fast Solvers for Discrete Diffusion Models: Theory and Applications of High-Order Algorithms
by: Ren, Yinuo, et al.
Published: (2025)

From Masks to Worlds: A Hitchhiker's Guide to World Models
by: Bai, Jinbin, et al.
Published: (2025)

On the Complexity of Bicycle Unitracks
by: Molodyk, Ivan
Published: (2025)

Diffusion Generative Modeling for Spatially Resolved Gene Expression Inference from Histology Images
by: Zhu, Sichen, et al.
Published: (2025)

TR2-D2: Tree Search Guided Trajectory-Aware Fine-Tuning for Discrete Diffusion
by: Tang, Sophia, et al.
Published: (2025)

Analyzing and Improving Optimal-Transport-based Adversarial Networks
by: Choi, Jaemoo, et al.
Published: (2023)

Overcoming Spurious Solutions in Semi-Dual Neural Optimal Transport: A Smoothing Approach for Learning the Optimal Transport Plan
by: Choi, Jaemoo, et al.
Published: (2025)

Generative Modeling through the Semi-dual Formulation of Unbalanced Optimal Transport
by: Choi, Jaemoo, et al.
Published: (2023)

Scalable Wasserstein Gradient Flow for Generative Modeling through Unbalanced Optimal Transport
by: Choi, Jaemoo, et al.
Published: (2024)

LaViDa-R1: Advancing Reasoning for Unified Multimodal Diffusion Language Models
by: Li, Shufan, et al.
Published: (2026)

Quantum State Generation with Structure-Preserving Diffusion Model
by: Zhu, Yuchen, et al.
Published: (2024)

Efficient Generative Modeling beyond Memoryless Diffusion via Adjoint Schrödinger Bridge Matching
by: Shin, Jeongwoo, et al.
Published: (2026)

Trivialized Momentum Facilitates Diffusion Generative Modeling on Lie Groups
by: Zhu, Yuchen, et al.
Published: (2024)

Multi-source Stable Variable Importance Measure via Adversarial Machine Learning
by: Wang, Zitao, et al.
Published: (2024)

Diffusion Reconstruction-based Data Likelihood Estimation for Core-Set Selection
by: Chen, Mingyang, et al.
Published: (2025)

DiC: Rethinking Conv3x3 Designs in Diffusion Models
by: Tian, Yuchuan, et al.
Published: (2024)

An analogue of non-interacting quantum field theory in Riemannian signature
by: Molodyk, Mikhail, et al.
Published: (2024)

The Feynman propagator for massive Klein-Gordon fields on radiative asymptotically flat spacetimes
by: Molodyk, Mikhail, et al.
Published: (2025)

Beyond Reward Margin: Rethinking and Resolving Likelihood Displacement in Diffusion Models via Video Generation
by: Xu, Ruojun, et al.
Published: (2025)

Adjoint Schrödinger Bridge Sampler
by: Liu, Guan-Horng, et al.
Published: (2025)