Research map / Image generation / Fast diffusion sampling
Efficient Weighted Sampling via Score-based Generative Models
Heasung Kim, Taekyun Lee, Hyeji Kim, Gustavo De Veciana
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use LAGS instead of DAS (particle resampling) for PickScore-weighted sampling | SDXL · SDXL prompts | runtime: 4.722 -> 1.000 (relative runtime) | Improved |
| use LAGS instead of DAS for guidance | SDXL · SDXL prompts | PickScore | Improved |
| use LAGS instead of DAS | SD / SDXL · ImageReward and CLIP Score | ImageReward, CLIP Score: margin on SDXL less than 0.1 and 0.06 | Hurt |
| use LAGS instead of DAS | SDXL · SDXL prompts | BRISQUE: 19.55 -> 26.40 | Hurt |
| use LAGS instead of DAS | SDXL · SDXL prompts | MANIQA: 0.720 -> 0.720 | No clear effect |
| use LAGS instead of FreeDoM | SDXL · SDXL prompts | BRISQUE: 34.49 -> 26.40 | Improved |
| apply guidance (any method) on top of default SDXL | SDXL · SDXL prompts | BRISQUE: 21.48 -> 26.40 (LAGS) | Hurt |
| vary scheduler confidence constant c (about 0.1 to 20) (ablation) | LAGS · weighted sampling tasks | task performance | No clear effect |
| use LAGS instead of FreeDoM and DAS on 2D multimodal heart target | 25-Gaussian mixture base · 2D heart-shaped weighted sampling | Wasserstein distance | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- FreeDoM — FreeDoM: Training-Free Energy-Guided Conditional Diffusion Model (ICCV 2023) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Evaluating the design space of diffusion-based generative models (NeurIPS 2024 · similarity 0.89)
- Accelerating Convergence of Score-Based Diffusion Models, Provably (ICML 2024 · similarity 0.89)
- Nearest Neighbour Score Estimators for Diffusion Generative Models (ICML 2024 · similarity 0.89)
- Outsourced Diffusion Sampling: Efficient Posterior Inference in Latent Spaces of Generative Models (ICML 2025 · similarity 0.88)
- Scalable Discrete Diffusion Samplers: Combinatorial Optimization and Statistical Physics (ICLR 2025 · similarity 0.87)
- A Simple Early Exiting Framework for Accelerated Sampling in Diffusion Models (ICML 2024 · similarity 0.87)
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.
