Beyond Mean Attention: Diversity-Aware, Layer-Wise Scoring for KV Cache Eviction

arXiv:2609.30738 · cs.CL, cs.LG · Submitted 2026-09-25 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-09-25

Updated: 2026-09-25

Terminology

• Redundancy — 12× in this paper · explained in 2 episodes
• KV cache — 11× in this paper · explained in 3 episodes
• Ablation — 6× in this paper · explained in 1 episode
• Benchmark — 3× in this paper · explained in 2 episodes
• Ablation Studies — 2× in this paper · explained in 1 episode
• Ablation study — 1× in this paper · explained in 1 episode
• Cosine Similarity — 1× in this paper · explained in 3 episodes
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Large Language Models (LLMs) — 1× in this paper · explained in 6 episodes
• Vision-language models — 1× in this paper · explained in 2 episodes

Sources

• SnapKV: LLM Knows What You are Looking for Before Generation
• PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
• Efficient Streaming Language Models with Attention Sinks
• Model Tells You What to Discard: Adaptive KV Cache Compression for LLMs
• Small Frequency Corrections Can Change What Survives KV Cache Compression
• Mistral 7B

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering