Research map / Image generation / Fast diffusion sampling
Step-unrolled Denoising Autoencoders for Text Generation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add a second, unrolled denoising loss term (ablation) | SUNDAE · WMT'14 EN→DE test | BLEU: 11.19 -> 26.57 | Helped |
| Add a third unrolled loss term (ablation) | SUNDAE with L^(1:2) · WMT'14 EN→DE test | BLEU: 26.57 -> 26.25 | No clear effect |
| Remove target length prediction (ablation) | SUNDAE · WMT'14 EN→DE | BLEU: degrades on average by 2 BLEU score points | Hurt |
| Use stochastic sampling rather than deterministic decoding | SUNDAE · WMT'14 EN→DE raw | BLEU: 25.54 -> 26.25 | Mixed |
Closest papers by meaning
- Denoising Time Matters: Diverse Generation in Diffusion Language Models (NeurIPS 2026 · similarity 0.87)
- Fast and Fluent Diffusion Language Models via Convolutional Decoding and Rejective Fine-tuning (NeurIPS 2025 · similarity 0.86)
- Compressed and Smooth Latent Space for Text Diffusion Modeling (NeurIPS 2025 · similarity 0.86)
- Error Norm Truncation: Robust Training in the Presence of Data Noise for Text Generation Models (ICLR 2024 · similarity 0.86)
- Reflection-Window Decoding: Text Generation with Selective Refinement (ICML 2025 · similarity 0.86)
- Energy-Based Diffusion Language Models for Text Generation (ICLR 2025 · similarity 0.86)
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.