compression methods

4 papers

• 4DGCPro: Efficient Hierarchical 4D Gaussian Compression for Progressive Volumetric Video Streaming
• ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
• Homogeneous Keys, Heterogeneous Values: Exploiting Local KV Cache Asymmetry for Long-Context LLMs
• Inference-Time Hyper-Scaling with KV Cache Compression