4 papers · 1 filter
BasisKV: Value Latents for Communication and Key Routing
Lixun Zhang, Zishan Shao, Kangning Cui +3
Long-context large language model (LLM) inference is increasingly constrained by the cost of maintaining an ever-growing KV-cache. Low-rank K/V projections provide a promising way…
The Functional Structure of Post-Compression Recovery in Low-Rank LLMs
Zishan Shao, Liang Tian, Georgiy Zemlevskiy +10
Different low-rank compression methods can produce compressed LLMs that respond differently to the same post-compression recovery procedure, and relative advantages observed betwee…
Preserving Mathematical Reasoning in Compressed Diffusion Language Models via Trajectory-Aware Low-Rank Approximation
Tian Liang, Zishan Shao, Yiran Chen
Diffusion language model (dLLM) compression faces a known challenge because calibration is typically performed on clean, fully visible activations, whereas inference traverses part…
DecodeShare: Tracing the Shared Subspace of LLM Decode-Time Decisions
Zishan Shao, Lixun Zhang, Kangning Cui +10
Large language models (LLMs) handle many tasks with one set of parameters, but under KV-cached inference it is unclear what task-general structure, if any, is used at decode time r…