bicache enables shared-prefix KV caching in diffusion language models by dynamically selecting reusable shallow-layer depths based on prefix fraction, delivering 36.3-98.3% throughput gains with 0-1.8% accuracy difference.
Reward-Weighted Sampling: Enhancing Non-Autoregressive Characteristics in Masked Diffusion LLM s
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Enabling KV Caching of Shared Prefix for Diffusion Language Models
bicache enables shared-prefix KV caching in diffusion language models by dynamically selecting reusable shallow-layer depths based on prefix fraction, delivering 36.3-98.3% throughput gains with 0-1.8% accuracy difference.