Continuous Diffusion LMs: One-hot Flow Jumps From 4% to 50% on TinyGSM
brandondamos · x · 2026-10-14
New research on continuous diffusion language models (CDLMs) improves one-hot flow from 4% to 50% on TinyGSM without self-conditioning, and systematically explores the design space — questioning why self-conditioning is used and how tokens should be embedded. The authors note CDLMs went from niche to contender this year.
More from Research
- Dev trains boolean-logic LM with microsecond forward passes on a single CPU thread — cephaloform · 2026-10-23
- AdaDex turns seconds of human hand motion into generalizable dexterous robot skills — mangahomanga · 2026-10-23
- ReAPO paper: conditioning retries on failures beats GRPO in RLVR reasoning training — jasondeanlee · 2026-10-23
- Quantum-JEPA: first JEPA world model deployable on real quantum circuits (QPU) — DrMorganLevine · 2026-10-23
- Omnii Genome: Single Generalist Model Outperforms AlphaGenome Across All Benchmarks — JosephJacks_ · 2026-10-23
- Domingos Cites New Result Claiming RL Is 'Basically Useless' vs Supervised Learning — pmddomingos · 2026-10-23