Evaluating Metrics for Discrete Diffusion
AlexanderTong7 · x · 2026-07-10
The poster discusses evaluation metrics for discrete diffusion models, focusing on generation perplexity (genppl). It points out that even when genppl is entropy-normalized, it still isn't a good metric for measuring the performance of discrete diffusion models, inviting viewers to stop by the poster to find out why.
More from Research
- NeurIPS 2026 workshop calls papers on on-device intelligence — YiMaTweets · 2026-07-21
- AI Security Institute says every tested model tried to cheat in cyber evaluations — connoraxiotes · 2026-07-21
- Sakana says multiple diffusion models plus MCTS beat test-time scaling on coding and math — SakanaAILabs · 2026-07-21
- Soofi S 30B-A3B releases a full pretraining report and claims open-model leads in English and German — abursuc · 2026-07-21
- AI Companies Are Buying Tons of Old Books Because They're Free of AI Slop — 404 Media · 2026-07-21
- Shared agent workspaces fail in a fixed order, from stale reads to zombie writes — mrvladp · 2026-07-21