# LESS Is More: Mutual-Stability Sampling for Diffusion Language Models

arXiv: https://arxiv.org/abs/2606.16908 · HTML: https://mlsanity.com/paper/2606.16908/

## 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 21 of its 39 references that are in the map.
- First posted on arXiv 15 Jun 2026.

## Placed in

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

## Builds on (most cited first, 21 of 21)

- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Dream 7B: Diffusion Large Language Models](https://mlsanity.com/paper/2508.15487/) (2025)
- [Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decoding](https://mlsanity.com/paper/2505.22618/) (2025)
- [LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Models](https://mlsanity.com/paper/2505.19223/) (2025)
- [Mercury: Ultra-Fast Language Models Based on Diffusion](https://mlsanity.com/paper/2506.17298/) (2025)
- [The Diffusion Duality](https://mlsanity.com/paper/2506.10892/) (2025)
- [Diffusion Language Models Know the Answer Before Decoding](https://mlsanity.com/paper/2508.19982/) (2025)
- [A Survey on Parallel Text Generation: From Parallel Decoding to Diffusion Language Models](https://mlsanity.com/paper/2508.08712/) (2025)
- [Self Speculative Decoding for Diffusion Large Language Models](https://mlsanity.com/paper/2510.04147/) (2025)
- [d2Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Caching](https://mlsanity.com/paper/2509.23094/) (2025)
- [Structuring The Future: Diffusion LLM Speculative Decoding via Calibrated Draft Graphs](https://mlsanity.com/paper/2509.18085/) (2025)
- [DLM-One: Diffusion Language Models for One-Step Sequence Generation](https://mlsanity.com/paper/2506.00290/) (2025)
- [FS-DFM: Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models](https://mlsanity.com/paper/2509.20624/) (2025)
- [A Survey of Diffusion Models in Natural Language Processing](https://mlsanity.com/paper/2305.14671/) (2023)
- [UltraLLaDA: Scaling the Context Length to 128K for Diffusion Large Language Models](https://mlsanity.com/paper/2510.10481/) (2025)
- [Fast-Decoding Diffusion Language Models via Progress-Aware Confidence Schedules](https://mlsanity.com/paper/2512.02892/) (2025)
- [Beyond I-Con: Exploring New Dimension of Distance Measures in Representation Learning](https://mlsanity.com/paper/2509.04734/) (2025)

## Abstract (arXiv)

Diffusion large language models (dLLMs) offer a promising alternative to autoregressive decoding by iteratively refining masked sequences, enabling parallel token updates and bidirectional conditioning. Their practical efficiency, however, is limited by sampling procedures that execute a fixed number of reverse denoising steps selected before decoding, spending computation on already-stable positions and sometimes committing unstable ones too early. We present textscLESS, a training-free, model-agnostic adaptive sampler that treats token commitment as an online stopping problem. textscLESS implements mutual-stability sampling through a joint stability rule that makes a masked position eligible for unmasking only when its top-1 prediction has high confidence, its top-1 token persists across recent reverse steps, and its predictive distribution is stable under top-K inter-step Jensen--Shannon divergence. We evaluate textscLESS on Dream-7B, LLaDA-8B, and LLaDA-1.5-8B, covering full-sequence diffusion and semi-autoregressive blockwise sampling regimes, across seven benchmarks spanning general knowledge, math, and code. textscLESS improves average accuracy over strong training-free adaptive samplers while using 72.1% fewer reverse steps than fixed-budget decoding. Since each reverse step requires a Transformer forward pass, these step-count reductions translate into fewer forward evaluations, lower measured wall-clock latency, and lower estimated inference compute.