# Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Models

arXiv: https://arxiv.org/abs/2609.33355 · HTML: https://mlsanity.com/paper/2609.33355/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Discrete and text diffusion, one of 14 directions in Diffusion models.
- Builds on 17 of its 41 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- Diffusion models / [Discrete and text diffusion](https://mlsanity.com/diffusion/discrete-and-text-diffusion/)

## Builds on (most cited first, 17 of 17)

- [Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution](https://mlsanity.com/paper/2310.16834/) (2023)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Models](https://mlsanity.com/paper/2505.19223/) (2025)
- [Accelerated Sampling from Masked Diffusion Models via Entropy Bounded Unmasking](https://mlsanity.com/paper/2505.24857/) (2025)
- [Path Planning for Masked Diffusion Model Sampling](https://mlsanity.com/paper/2502.03540/) (2025)
- [Optimal Inference Schedules for Masked Diffusion Models](https://mlsanity.com/paper/2511.04647/) (2025)
- [Lookahead Unmasking Elicits Accurate Decoding in Diffusion Language Models](https://mlsanity.com/paper/2511.05563/) (2025)
- [Deferred Commitment Decoding for Diffusion Language Models](https://mlsanity.com/paper/2601.02076/) (2026)
- [Generation Order and Parallel Decoding in Masked Diffusion Models: An Information-Theoretic Perspective](https://mlsanity.com/paper/2602.00286/) (2026)
- [Improving Diffusion Language Model Decoding through Joint Search in Generation Order and Token Space](https://mlsanity.com/paper/2601.20339/) (2026)
- [Where-to-Unmask: Ground-Truth-Guided Unmasking Order Learning for Masked Diffusion Language Models](https://mlsanity.com/paper/2602.09501/) (2026)
- [Dependency-Guided Parallel Decoding in Discrete Diffusion Language Models](https://mlsanity.com/paper/2604.02560/) (2026)
- [DSB: Dynamic Sliding Block Scheduling for Diffusion LLMs](https://mlsanity.com/paper/2602.05992/) (2026)
- [When to Commit? Towards Variable-Size Self-Contained Blocks for Discrete Diffusion Language Models](https://mlsanity.com/paper/2604.23994/) (2026)
- [Unifying Masked Diffusion Models with Various Generation Orders and Beyond](https://mlsanity.com/paper/2602.02112/) (2026)
- [Swordsman: Entropy-Driven Adaptive Block Partition for Efficient Diffusion Language Models](https://mlsanity.com/paper/2602.04399/) (2026)
- [Search or Accelerate: Confidence-Switched Position Beam Search for Diffusion Language Models](https://mlsanity.com/paper/2602.10953/) (2026)

## Abstract (arXiv)

Masked diffusion language models (MDMs) admit flexible generation orders, making the unmasking strategy an inference decision. Existing methods vary in how they prioritize positions, control parallelism, restrict selection regions, revise predictions, or plan future denoising, yet it remains unclear when these choices should change during generation. We study this question through strategy reversals, where an alternative action becomes preferable to a fixed choice. We organize MDM inference into five axes--score, cardinality, region, commitment, and planning--and define adaptation opportunity as the one-step utility advantage of the best candidate action over a validation-selected fixed action. This view shows that adaptation value depends on both the frequency and magnitude of such reversals. Across three MDMs and ten tasks, adaptation opportunities are highly heterogeneous, with some regimes exhibiting concentrated and predictable one-step gains. This motivates selective adaptation: lightweight detectors calibrated on validation prompts identify high-opportunity states, capturing, for example, 56.9 percent of the candidate-set oracle opportunity by adapting only the top 10 percent of states on LLaDA-8B constrained JSON filling. Our transition-level results suggest that state adaptation is most useful when applied selectively rather than uniformly.