AI RESEARCH
SPA-Cache: Singular Proxies for Adaptive Caching in Diffusion Language Models
arXiv CS.AI
•
ArXi:2602.02544v2 Announce Type: replace-cross While Diffusion Language Models (DLMs) offer a flexible, arbitrary-order alternative to the autoregressive paradigm, their non-causal nature precludes standard KV caching, forcing costly hidden state recomputation at every decoding step. Existing DLM caching approaches reduce this cost by selective hidden state updates; however, they are still limited by (i) costly token-wise update identification heuristics and (ii) rigid, uniform budget allocation that fails to account for heterogeneous hidden state dynamics.