Discrete Diffusion Models via Evolving Variational Autoregressive Networks
arXiv:2609.27306v1 Announce Type: cross Abstract: Conventional score-based diffusion models learn scores without representing normalized densities, whereas tractable normalized models support both sampling and direct likelihood evaluation. A recent tensor-network approach provides such a representa…
Read the full story at arXiv stat.ML ↗
Timeline · 1 report
- 2026-09-24 04:00 · arXiv stat.ML
Discrete Diffusion Models via Evolving Variational Autoregressive Networks