AI Paper Atlas

Research map / Reinforcement learning / RL from feedback

Parallel Diffusion Solver via Residual Dirichlet Policy Optimization

Ruoyu Wang, 李資玉, Beier Zhu, Liangyu Yuan, H L Zhang, Xiaomin Yang, Xiaojun Chang, Chi Zhang

TPAMI 2026 · Reinforcement learning › RL from feedback · approach: RL fine-tuning of generative models · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add learned scaling factor o_n (ablation)EPD-Solver · CIFAR10, NFE 3 and 9FID: 13.25 -> 10.40 (NFE 3); 2.79 -> 2.49 (NFE 9)Improved
add learned scaling factors delta_n^k (ablation)EPD-Solver · CIFAR10, NFE 3 and 9FID: 13.02 -> 10.40 (NFE 3); 2.69 -> 2.49 (NFE 9)Improved
add both scaling factors (ablation)EPD-Solver · CIFAR10, NFE 3 and 9FID: 16.01 -> 10.40 (NFE 3); 3.24 -> 2.49 (NFE 9)Improved
use Uniform time schedule instead of LogSNR (ablation)EPD-Solver · CIFAR10, NFE 3FID: 54.07 -> 10.40Improved
use Uniform time schedule instead of Polynomial (ablation)EPD-Solver · CIFAR10, NFE 3FID: 11.10 -> 10.40Improved
use DPM-Solver-2 as teacher instead of Heun (ablation)EPD-Solver · CIFAR10, NFE 3FID: 15.91 -> 10.40Improved
use DPM-Solver-2 as teacher instead of iPNDM (ablation)EPD-Solver · CIFAR10, NFE 3FID: 13.69 -> 10.40Improved
increase parallel directions K from 1 to 2 (ablation)EPD-Solver · FFHQ, NFE 3FID: 26.0 -> 22.7Improved
increase parallel directions K from 2 to 3 (ablation)EPD-Solver · ImageNet, NFE 9FID: 4.20 -> 4.18No clear effect
add parallel directions K=2 vs K=1 for latency (ablation)EPD-Solver · SD1.5, 20 stepslatency (s): 0.6048 -> 0.6252No clear effect
add RL fine-tuning (Dirichlet policy) after distillation (ablation)EPD-Solver · SD v1.5ImageReward: 3.0405 -> 3.1121Improved
use EPD-Solver with 10 steps instead of iPNDM 50 stepsiPNDM · SD v1.5, HPSv2.1HPSv2.1: 0.2474 -> 0.2482Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy