UnMaskFork: Test-Time Scaling for Masked Diffusion Models
SakanaAILabs · x · 2026-07-04
Sakana AI introduced UnMaskFork, a test-time scaling framework for Masked Diffusion Language Models (MMDLM), which will be presented at ICML 2026.
Leveraging Monte Carlo Tree Search (MCTS), the method dynamically switches between multiple pretrained MDLMs for collaborative text generation, exploring diverse generation paths. It consistently outperforms standard Best-of-N and other tree search baselines on coding and mathematical reasoning tasks.
More from Research
- 'Vibe verification': ~60% of AI-written formal specs prove the wrong property — basedjensen · 2026-10-26
- Ito: a browser that detects browsing intent and auto-organizes tabs into Flows — andrewkuznet · 2026-10-26
- Cornell study says brains process language as linear sequences, challenging Chomsky's hierarchy — anirbanbandyo · 2026-10-26
- Hi3D V1.0 open-sourced with 2M training assets and 90% slicing success rate — kimmonismus · 2026-10-26
- Ortet evaluates 34 open-weight models on 50 health tasks, finds general leaderboards insufficient — elmanmansimov · 2026-10-26
- HackBench injects reward hacks into agent trajectories to stress-test monitors — DanielKhashabi · 2026-10-26