GEAR: An Efficient KV Cache Compression Recipe for Near-Lossless Generative Inference of LLM

Documenti analoghi

• KeepKV: Achieving Periodic Lossless KV Cache Compression for Efficient LLM Inference
  di: Tian, Yuxuan, et al.
  Pubblicazione: (2025)
• Lossless KV Cache Compression to 2%
  di: Yang, Zhen, et al.
  Pubblicazione: (2024)
• KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference
  di: Li, Xing, et al.
  Pubblicazione: (2025)
• ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
  di: Liu, Xiang, et al.
  Pubblicazione: (2025)
• EvolKV: Evolutionary KV Cache Compression for LLM Inference
  di: Yu, Bohan, et al.
  Pubblicazione: (2025)