2 papers
cs.CL2026
DeltaKV: Residual-Based KV Cache Compression via Long-Range Similarity
Jitai Hao, Qiang Huang, Yaowei Wang +2
The deployment of efficient long-context LLMs in applications like autonomous agents, long-chain reasoning, and creative writing is fundamentally bottlenecked by the linear growth…
cs.CV2025
Uni-X: Mitigating Modality Conflict with a Two-End-Separated Architecture for Unified Multimodal Models
Jitai Hao, Hao Liu, Xinyan Xiao +2
Unified Multimodal Models (UMMs) built on shared autoregressive (AR) transformers are attractive for their architectural simplicity. However, we identify a critical limitation: whe…