Progressive UnMAsking
No mentions found
This entity hasn't been tracked yet, or Iris is still building its knowledge base.
Related Articles from SNS
Stop Training for the Worst: Progressive Unmasking Accelerates Masked Diffusion Training
arXiv:2602.10314v2 Announce Type: replace Abstract: Masked Diffusion Models (MDMs) have emerged as a promising approach for generative modeling in discrete spaces. By generating sequences in any order and allowing for parallel decoding, they enable fast inference and strong performance on non-causal tasks. However, this flexibility comes with a training complexity trade-off: MDMs train on an exponentially large set of masking patterns, which is not only computationally expensive, but also...
ViewMask-1-to-3: Multi-View Consistent Image Generation via Multimodal Discrete Diffusion Models
arXiv:2512.14099v3 Announce Type: replace Abstract: Motivated by discrete diffusion's success in language-vision modeling, we explore its potential for multi-view generation, a task dominated by continuous approaches. We introduce ViewMask-1-to-3, formulating multi-view generation as a discrete sequence modeling problem where each viewpoint is represented as visual tokens from MAGVIT-v2. Through discrete diffusion via masked token prediction, our approach enables progressive multi-view...
Read the Trace, Steer the Path: Trajectory-Aware Reinforcement Learning for Diffusion Language Models
arXiv:2606.04396v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) generate responses by iteratively unmasking and revising many positions in parallel. This process leaves a rich denoising trace depicting which tokens become confident, which remain unstable, and when commitments form. Existing dLLM reinforcement learning methods use this signal only weakly.
Stop the Flip-Flop: Context-Preserving Verification for Fast Revocable Diffusion Decoding
arXiv:2602.06161v2 Announce Type: replace Abstract: Parallel diffusion decoding can accelerate diffusion language model inference by unmasking multiple tokens per step, but aggressive parallelism often harms quality. Revocable decoding mitigates this by rechecking earlier tokens, yet we observe that existing verification schemes frequently trigger flip-flop oscillations, where tokens are remasked and later restored unchanged. This behaviour slows inference in two ways: remasking verified...
How to Tell the American Story
Illustrations by Tyler ComrieOn a July afternoon in 2019, I found myself in a large, sun-dappled room within one of America’s great estates. An assemblage of distinguished jurists, Ivy League professors, nonprofit leaders, journalists, and theologians sat around me in a half circle. I was trying to be on my best behavior, but I blurted out a word dirty enough to make them blanch.
Algal blooms explained: How scientists are helping spot them sooner
Algal blooms explained: How scientists are helping spot them sooner Lisa Lock Scientific Editor Andrew Zinin Lead Editor Algal blooms can seem to appear overnight. A stretch of ocean that looked clear days earlier can suddenly appear discolored and sometimes pose risks to ecosystems and human health. But scientists say blooms are rarely sudden—understanding what happens before they appear is key to managing them.
U.S. and Iran launch fresh strikes and the House incumbents that could be unseated: Morning Rundown
In today’s newsletter: The U.S. and Iran trade attacks in the latest escalation since agreeing to a ceasefire. Democratic incumbents are at risk of losing their seats as they face strong challengers. And why shoppers paying in cash are bearing the brunt of credit card fees.