AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

ReGen: Hierarchical Multi-Prompt Representation Generation for Efficient Waveform Diffusion Models

Sang-Hoon Lee, Ha-Yeong Choi

ICML 2026 · Image generation › Autoregressive and latent generation · approach: Diffusion audio synthesis · 0 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 REPA with Mel-spectrogram alignment (ablation)waveform diffusion codec · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 136.57 -> 15.10Improved
add REPA with SSL representation only (ablation)waveform diffusion codec · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 136.57 -> 130.64No clear effect
use hierarchical REPA (SSL, Mel) instead of Mel-only REPA (ablation)waveform diffusion codec · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsSPK-SIM: 0.32 -> 0.38Improved
replace REPA with representation generation (ReGen-H, SSL and Mel) (ablation)waveform diffusion codec · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 16.53 -> 11.99Improved
replace REPA with representation generation (ReGen-H, SSL and Mel) (ablation)waveform diffusion codec · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsSPK-SIM: 0.38 -> 0.54Improved
use unified generation at same layer instead of hierarchical (ablation)ReGen · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 11.99 -> 89.94Hurt
lower representation generation weight to 0.01 in unified ReGen (ablation)ReGen-U · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 89.94 -> 19.31Improved
add generalized flow matching (repulsive term) (ablation)ReGen-H · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsSPK-SIM: 0.54 -> 0.59Improved
add generalized flow matching (repulsive term) (ablation)ReGen-H · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 11.99 -> 12.59No clear effect
add adversarial post-training (ablation)ReGenVAE · LibriTTS reconstructionWER: 3.63 -> 2.88Improved
add auxiliary decoder (CosyVoice2 token distillation on FSQ) (ablation)ReGenTokenizer · LibriTTS reconstruction, waveform diffusion codec, 0.5M stepsWER: 12.59 -> 6.01Improved
add GFM objective to LDM-based TTS (ablation)ReGenVoice · Seed-en zero-shot TTSWER: 1.60 -> 1.46Improved

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