mlsanityData to 7 Oct 2026

CAI-DLLM: Convergence Aware Inference for Diffusion Language Models

arXiv 2608.22646 ↗First posted 23 Aug 2026cs.AIcited by 0

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Diffusion language models can generate many tokens in parallel, but they still require repeated denoising steps during inference. This makes generation costly, especially when the model continues to recompute tokens that are already stable. To address these limitations, we propose CAI-DLLM, a training-free inference method that uses first-step confidence to guide denoising and reduce inference time. Specifically, CAI-DLLM commits easy tokens earlier, allocates more denoising steps to harder tokens, and adjusts decoding schedules across output blocks. As it relies only on first-step confidence signals, it does not require retraining, extra predictors, or weight updates. We evaluate CAI-DLLM on LLaDA-8B-Instruct and Dream-7B-Instruct across math, code, reasoning, commonsense, and long-context tasks. CAI-DLLM achieves up to 18.2x wall clock inference speedup on LLaDA GSM8K while improving accuracy from 76.27% to 77.41%, and up to 13.1x speedup on Dream HumanEval while achieving higher pass@1 than no-cache inference, 48.17% compared with 46.95%. On harder reasoning tasks, speedups reach 44.8x, with a largest accuracy drop of 4.4 points, while energy consumption is reduced by up to 95.3%.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Discrete and text diffusion, one of 14 directions in Diffusion models. Builds on 21 of its 26 references that are in the map. First posted on arXiv 23 Aug 2026.

Builds on

21 of 26 references.

Closest ancestors · 5

Same direction, newest first

  1. 2025Dream 7B: Diffusion Large Language Models21 Aug
  2. 2025Diffusion LLMs Can Do Faster-Than-AR Inference via Discrete Diffusion Forcing8 Aug
  3. 2025FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion27 May
  4. 2025Large Language Diffusion Models14 Feb
  5. 2023Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution25 Oct

Other references · 16

  1. 2017Attention Is All You Needcited by 59,932
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  5. 2019PIQA: Reasoning about Physical Commonsense in Natural Languagecited by 2,818
  6. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  7. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  8. 2022Fast Inference from Transformers via Speculative Decodingcited by 1,615
  9. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,378
  10. 2019MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalismscited by 653
  11. 2016Adaptive Computation Time for Recurrent Neural Networkscited by 479
  12. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
4 more
  1. 2018Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinementcited by 194
  2. 2025dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Cachingcited by 168
  3. 2025Sparse-dLLM: Accelerating Diffusion LLMs with Dynamic Cache Evictioncited by 64
  4. 2026ES-dLLM: Efficient Inference for Diffusion Large Language Models by Early-Skippingcited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025dInfer: An Efficient Inference Framework for Diffusion Language Models
  2. 2025EDIT: Early Diffusion Inference Termination for dLLMs Based on Dynamics of Training Gradients
  3. 2026Temporal Self-Distillation: Faster Inference in Discrete Diffusion Language Models
  4. 2026Roll Out and Roll Back: Diffusion LLMs are Their Own Efficiency Teachers
  5. 2026CORA-Diff: Confidence-Oriented Residual Acceptance for Efficient Diffusion Language Model Inference
  6. 2026Window-Diffusion: Accelerating Diffusion Language Model Inference with Windowed Token Pruning and Caching
  7. 2026Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques
  8. 2026Prism: Efficient Test-Time Scaling via Hierarchical Search and Self-Verification for Discrete Diffusion Language Models