Find a story
Search Spins
Search titles, summaries, and missing voices across published articles — press releases, announcements, and media coverage.
3 results for “diffusion language models”
Commitment Before Realization: When Classifier-Free Guidance Becomes Unnecessary in Masked Diffusion Language Models
Researchers propose a method to dynamically deactivate classifier-free guidance (CFG) during masked diffusion language model decoding once a 'commitment horizon' is reached, improving efficiency without sacrificing constraint satisfaction across 13 subtasks.
Aug 11, 2026
OPTD: On-Policy Transition Distillation with Consistency-Guided Adaptive Compression for Few-Step Diffusion Language Models
A new AI research paper introduces OPTD, a method to improve few-step diffusion language models by using on-policy distillation with adaptive compression, aiming to balance generation quality and decoding speed.
Aug 5, 2026
Adaptive Multi-Step Lookahead Decoding for Diffusion Language Models
Researchers introduced AdaLook, an adaptive multi-step lookahead decoding method for masked diffusion language models that dynamically adjusts rollout depth based on candidate-score variance to improve the accuracy–decoding steps trade-off.
Jul 20, 2026