mlsanityData to 7 Oct 2026

EntropyCache: Decoded Token Entropy Guided KV Caching for Diffusion Language Models

arXiv 2603.18489 ↗First posted 19 Mar 2026cs.CLcited by 3

On the map: Diffusion models → Discrete and text diffusion

Abstract (arXiv)

Diffusion-based large language models (dLLMs) rely on bidirectional attention, which prevents lossless KV caching and requires a full forward pass at every denoising step. Existing approximate KV caching methods reduce this cost by selectively updating cached states, but their decision overhead scales with context length or model depth. We propose EntropyCache, a training-free KV caching method that uses the maximum entropy of newly decoded token distributions as a constant-cost signal for deciding when to recompute. Our design is grounded in two empirical observations: (1) decoded token entropy correlates with KV cache drift, providing a cheap proxy for cache staleness, and (2) feature volatility of decoded tokens persists for multiple steps after unmasking, motivating recomputation of the k most recently decoded tokens. The skip-or-recompute decision requires only O(V) computation per step, independent of context length and model scale. Experiments on LLaDA-8B-Instruct and Dream-7B-Instruct show that EntropyCache achieves 15.2×-26.4× speedup on standard benchmarks and 22.4×-24.1× on chain-of-thought benchmarks, with competitive accuracy and decision overhead accounting for only 0.5% of inference time. Code is available at https://github.com/mscheong01/EntropyCache.

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 21 of its 24 references that are in the map. First posted on arXiv 19 Mar 2026.

Builds on

21 of 24 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025FlashDLM: Accelerating Diffusion Language Model Inference via Efficient KV Caching and Guided Diffusion27 May
  2. 2025dKV-Cache: The Cache for Diffusion Language Models21 May
  3. 2025Large Language Diffusion Models14 Feb
  4. 2024Simple and Effective Masked Diffusion Language Models11 Jun

Other references · 17

  1. 2017Attention Is All You Needcited by 59,932
  2. 2023LLaMA: Open and Efficient Foundation Language Modelscited by 13,958
  3. 2021Training Verifiers to Solve Math Word Problemscited by 9,743
  4. 2021Evaluating Large Language Models Trained on Codecited by 8,053
  5. 2021Measuring Mathematical Problem Solving With the MATH Datasetcited by 5,203
  6. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  7. 2021Program Synthesis with Large Language Modelscited by 3,509
  8. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,748
  9. 2022Challenging BIG-Bench Tasks and Whether Chain-of-Thought Can Solve Themcited by 1,767
  10. 2024MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmarkcited by 1,758
  11. 2021Structured Denoising Diffusion Models in Discrete State-Spacescited by 1,665
  12. 2022Efficiently Scaling Transformer Inferencecited by 529
5 more
  1. 2025Fast-dLLM: Training-free Acceleration of Diffusion LLM by Enabling KV Cache and Parallel Decodingcited by 404
  2. 2025dLLM-Cache: Accelerating Diffusion Large Language Models with Adaptive Cachingcited by 168
  3. 2025Fast-dLLM v2: Efficient Block-Diffusion LLMcited by 128
  4. 2025Attention Is All You Need for KV Cache in Diffusion LLMscited by 32
  5. 2025d2Cache: Accelerating Diffusion-Based LLMs via Dual Adaptive Cachingcited by 24

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Improved Large Language Diffusion Modelscited by 12
  2. 2026Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMscited by 1
  3. 2026Affix Cache for Diffusion Large Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models
  2. 2026Fixed State, Long Reach: What a Constant-Size Cache Buys Block Diffusion at Scale
  3. 2026Mask-Guided KV Cache Eviction in Block Diffusion Language Models
  4. 2026Archer: Adaptive Reuse of Cached Hidden States for Efficient Rollback in Diffusion Language Models
  5. 2026Window-Diffusion: Accelerating Diffusion Language Model Inference with Windowed Token Pruning and Caching
  6. 2026WaveFilter: Enhancing the Long-Context Capability of Diffusion LLMs via Wavelet-Guided KV Cache Filtering
  7. 2025Mask Tokens as Prophet: Fine-Grained Cache Eviction for Efficient dLLM Inference
  8. 2026Behavior-Preserving KV Cache Compression