Research map / Image generation / Fast diffusion sampling
CREPE: Controlling diffusion with REPlica Exchange
Jiajun He, Paul Jeha, Peter Potaptchik, Leo Zhang, José Miguel Hernández-Lobato, Yuanqi Du, Saifuddin Syed, Francisco Vargas
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 CREPE (replica exchange) instead of SMC for sentiment-controlled CFG debiasing in discrete diffusion | CTMC text diffusion · sentiment text generation, target strength 1.8 | perplexity: up to a factor of 5 | Improved |
| use CREPE instead of FKC (SMC) for debiasing CFG with large sample counts | class-conditional image diffusion · ImageNet CFG debiasing | FID | Improved |
| use CREPE instead of SMC when only 8 samples are generated | class-conditional image diffusion · ImageNet CFG debiasing, 8 samples | ImageReward, CLIP, FID | Hurt |
Closest papers by meaning
- A Mean-Field Framework for Inference-Time Distributional Control of Diffusion Models (NeurIPS 2026 · similarity 0.88)
- On scalable and efficient training of diffusion samplers (NeurIPS 2025 · similarity 0.87)
- Test-time Alignment of Diffusion Models without Reward Over-optimization (ICLR 2025 · similarity 0.87)
- Progressive Inference-Time Annealing of Diffusion Models for Sampling from Boltzmann Densities (NeurIPS 2025 · similarity 0.87)
- Transport meets Variational Inference: Controlled Monte Carlo Diffusions (ICLR 2024 · similarity 0.87)
- Diffusion Model is Effectively Its Own Teacher (CVPR 2025 · 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.
