mlsanityData to 7 Oct 2026

Search or Accelerate: Confidence-Switched Position Beam Search for Diffusion Language Models

arXiv 2602.10953 ↗First posted 11 Feb 2026cs.CL cs.AIcited by 3

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Diffusion Language Models (DLMs) generate text by iteratively denoising a masked sequence, repeatedly deciding which positions to commit at each step. Standard decoding follows a greedy rule: unmask the most confident positions, yet this local choice can lock the model into a suboptimal unmasking order, especially on reasoning-heavy prompts. We present SOAR, a training-free decoding algorithm that adapts its behavior to the model's uncertainty. When confidence is low, SOAR briefly widens the search over alternative unmasking decisions to avoid premature commitments; when confidence is high, it collapses the search and decodes many positions in parallel to reduce the number of denoising iterations. Across mathematical reasoning and code generation benchmarks (GSM8K, MBPP, HumanEval) on Dream-7B and LLaDA-8B, SOAR improves generation quality while maintaining competitive inference speed, offering a practical way to balance quality and efficiency in DLM decoding. Our Code is available at https://github.com/duterscmy/SOAR

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 22 of its 25 references that are in the map. First posted on arXiv 11 Feb 2026.

Builds on

22 of 25 references.

Closest ancestors · 8

Same direction, newest first

  1. 2026DreamOn: Diffusion Language Models For Code Infilling Beyond Fixed-size Canvas1 Feb
  2. 2026Improving Diffusion Language Model Decoding through Joint Search in Generation Order and Token Space28 Jan
  3. 2025Don't Settle Too Early: Self-Reflective Remasking for Diffusion Language Models28 Sep
  4. 2025Diffusion Language Models Know the Answer Before Decoding27 Aug
  5. 2025Dream 7B: Diffusion Large Language Models21 Aug
  6. 2025DiffuCoder: Understanding and Improving Masked Diffusion Models for Code Generation25 Jun
  7. 2025dKV-Cache: The Cache for Diffusion Language Models21 May
  8. 2024Scaling Diffusion Language Models via Adaptation from Autoregressive Models23 Oct

Other references · 14

  1. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  2. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  3. 2021Program Synthesis with Large Language Modelscited by 3,509
  4. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  5. 2025LLaDA 1.5: Variance-Reduced Preference Optimization for Large Language Diffusion Modelscited by 273
  6. 2022Efficient Training of Language Models to Fill in the Middlecited by 248
  7. 2017Beam Search Strategies for Neural Machine Translationcited by 232
  8. 2025dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Cachingcited by 168
  9. 2025Beyond Fixed: Training-Free Variable-Length Denoising for Diffusion Large Language Modelscited by 32
  10. 2025Attention Is All You Need for KV Cache in Diffusion LLMscited by 32
  11. 2025Self Speculative Decoding for Diffusion Large Language Modelscited by 29
  12. 2025Diffusion LLM with Native Variable Generation Lengths: Let [EOS] Lead the Waycited by 22
2 more
  1. 2025Mask Tokens as Prophet: Fine-Grained Cache Eviction for Efficient dLLM Inferencecited by 17
  2. 2025Latent Refinement Decoding: Enhancing Diffusion-Based Language Models by Refining Belief Statescited by 9

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Focus on the Core: Empowering Diffusion Large Language Models by Self-Contrastcited by 0
  2. 2026Constraints Are Graphs, Not Chains: Exact Decoding for Diffusion Language Modelscited by 0
  3. 2026Unmask the State: When Does State Adaptation Matter for Masked Diffusion Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Diffusion Language Model Inference with Monte Carlo Tree Search
  2. 2025Lookahead Unmasking Elicits Accurate Decoding in Diffusion Language Models
  3. 2026Locally Confident, Globally Stuck: The Quality-Exploration Dilemma in Diffusion Language Models
  4. 2026Early-Bird Decoding: Accelerating Diffusion LLMs with Learnable Block Sizes and Parallel Sampling
  5. 2025From Bits to Rounds: Parallel Decoding with Exploration for Diffusion Language Models
  6. 2026CAI-DLLM: Convergence Aware Inference for Diffusion Language Models
  7. 2026Who Should Lead Decoding Now? Tracking Reliable Trajectories for Ensembling Masked Diffusion Language Models
  8. 2026When Confidence Misleads: Suffix Anchoring and Anchor-Proximity Confidence Modulation for Diffusion Language Models