Woosang Lim, Sungjoo Yoo, Donghyun Son, Minsoo Cheong
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
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
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2603.18489")
get_code_for_paper("2603.18489")
have("2603.18489")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2603.18489.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2603.18489)
Connect an agent — have() is free.