Trendoday

Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking

Diffusion language models (DLMs) generate text through iterative denoising, where decisions about which tokens to commit or remask strongly affect output quality. Existing certainty-aware strategies r

  • Upvotes4
–Rank now
#18Peak rank
3Days on list
176Score · 7d
Nov 11, 2026First seen
Rank · 7 days

About this paper

Diffusion language models (DLMs) generate text through iterative denoising, where decisions about which tokens to commit or remask strongly affect output quality. Existing certainty-aware strategies rely on instantaneous confidence or entropy, but these local signals can be misleading on difficult reasoning problems. We formulate DLM test-time scaling as adaptive allocation of both inference compute and decoding strategies. Our two-stage approach uses a fast base sampler to solve easy questions, then applies frequency-aware remasking to the remaining hard questions. This strategy tracks how often each token's prediction changes across denoising steps and preferentially commits frequently…

Upvotes
4
Authors
4
Published
Nov 7, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking was first seen on Hugging Face · Daily papers on Nov 11, 2026. Since then it has appeared in 23 checks on 3 different days. Its best position was #18 of 30, reached on Nov 11, 23:28 UTC. It dropped off the list after Nov 13, 15:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 19:2806:2817:2804:2815:28 Nov 11, 19:28 UTC · #25Nov 11, 21:28 UTC · #21Nov 11, 23:28 UTC · #18Nov 12, 01:28 UTC · #20Nov 12, 03:28 UTC · #19Nov 12, 05:28 UTC · #20Nov 12, 07:28 UTC · #21Nov 12, 09:28 UTC · #21Nov 12, 11:28 UTC · #20Nov 12, 13:28 UTC · #20Nov 12, 15:28 UTC · #20Nov 12, 17:28 UTC · #21Nov 12, 19:28 UTC · #21Nov 12, 21:28 UTC · #20Nov 12, 23:28 UTC · #21Nov 13, 01:28 UTC · #20Nov 13, 03:28 UTC · #20Nov 13, 05:28 UTC · #20Nov 13, 07:28 UTC · #22Nov 13, 09:28 UTC · #23Nov 13, 11:28 UTC · #22Nov 13, 13:28 UTC · #23Nov 13, 15:28 UTC · #23

Last 14 days

Best rank per day (UTC).

Nov 3today

Metric history

Platform metric as reported by Hugging Face.

0369 19:2806:2817:2804:2815:28 Nov 11, 19:28 UTC · 4Nov 11, 21:28 UTC · 7Nov 11, 23:28 UTC · 8Nov 12, 01:28 UTC · 8Nov 12, 03:28 UTC · 8Nov 12, 05:28 UTC · 8Nov 12, 07:28 UTC · 8Nov 12, 09:28 UTC · 8Nov 12, 11:28 UTC · 8Nov 12, 13:28 UTC · 8Nov 12, 15:28 UTC · 8Nov 12, 17:28 UTC · 8Nov 12, 19:28 UTC · 8Nov 12, 21:28 UTC · 8Nov 12, 23:28 UTC · 8Nov 13, 01:28 UTC · 8Nov 13, 03:28 UTC · 9Nov 13, 05:28 UTC · 9Nov 13, 07:28 UTC · 9Nov 13, 09:28 UTC · 9Nov 13, 11:28 UTC · 9Nov 13, 13:28 UTC · 9Nov 13, 15:28 UTC · 9

Questions

Is Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 13, 15:28 UTC, after 3 days of trending.

What is the highest rank Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking reached?

#18 on Hugging Face · Daily papers, on Nov 11, 23:28 UTC. Trendoday checks this list every 120 minutes.

How long has Test-Time Scaling for Diffusion Language Models with Frequency-Aware Remasking been trending?

It was first seen on Nov 11, 2026 and has been on the list on 3 days (23 checks) since.