UDM-GRPO is the first RL integration for uniform discrete diffusion models, using final clean samples as actions and forward-process trajectory reconstruction to raise GenEval accuracy from 69% to 96% and OCR accuracy from 8% to 57%.
Flow matching with general discrete paths: A kinetic-optimal perspective.arXiv preprint arXiv:2412.03487
6 Pith papers cite this work. Polarity classification is still indexing.
6
Pith papers citing it
representative citing papers
DoMinO reformulates discrete flow matching sampling as an MDP for unbiased RL fine-tuning with new TV regularizers, yielding better enhancer activity and naturalness on DNA design tasks.
DFM-VLA uses discrete flow matching to iteratively refine action tokens in VLA models, outperforming autoregressive and diffusion baselines with 4.44 average success length on CALVIN and 95.7% success on LIBERO.
Discrete flow matching on Z_m^d achieves non-asymptotic KL bounds for early-stopped targets and explicit TV convergence to the true target under an approximation error assumption, with improved scaling in dimension d and vocabulary size m.
Discrete Flow Maps recast flow map training for discrete domains using simplex geometry to enable single-step text generation from noise and outperform prior discrete flow models.
Flow Matching is a generative modeling framework with mathematical foundations, design choices, extensions, and open-source PyTorch code for applications like image and text generation.
citing papers explorer
-
UDM-GRPO: Stable and Efficient Group Relative Policy Optimization for Uniform Discrete Diffusion Models
UDM-GRPO is the first RL integration for uniform discrete diffusion models, using final clean samples as actions and forward-process trajectory reconstruction to raise GenEval accuracy from 69% to 96% and OCR accuracy from 8% to 57%.
-
Discrete Flow Matching Policy Optimization
DoMinO reformulates discrete flow matching sampling as an MDP for unbiased RL fine-tuning with new TV regularizers, yielding better enhancer activity and naturalness on DNA design tasks.
-
DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching
DFM-VLA uses discrete flow matching to iteratively refine action tokens in VLA models, outperforming autoregressive and diffusion baselines with 4.44 average success length on CALVIN and 95.7% success on LIBERO.
-
Discrete Flow Matching: Convergence Guarantees Under Minimal Assumptions
Discrete flow matching on Z_m^d achieves non-asymptotic KL bounds for early-stopped targets and explicit TV convergence to the true target under an approximation error assumption, with improved scaling in dimension d and vocabulary size m.
-
Discrete Flow Maps
Discrete Flow Maps recast flow map training for discrete domains using simplex geometry to enable single-step text generation from noise and outperform prior discrete flow models.
-
Flow Matching Guide and Code
Flow Matching is a generative modeling framework with mathematical foundations, design choices, extensions, and open-source PyTorch code for applications like image and text generation.