Running summarizing tools on a new article

Check the paper below, then start its synthesis.


Global Importance Estimation for KV Cache Eviction

Seojin Kim, Noseong Park

The 40th Conference on Neural Information Processing Systems (NeurIPS 2026)
License: CC BY 4.0

Abstract: Memory consumption of the key-value (KV) cache remains a central bottleneck for long-context LLM inference. We introduce PRKV, a training-free KV eviction method that estimates multi-hop global token importance by applying Personalized PageRank over attention-induced transition graphs. This enables importance propagation beyond recency or instantaneous attention, without modifying the model. We benchmark PRKV on long-context understanding, sparse retrieval, and chain-of-thought reasoning tasks using multiple LLM families. Under moderate to aggressive compression ratios, including up to 90%, PRKV achieves competitive task performance against established eviction baselines while reducing peak memory usage with low end-to-end overhead. These results demonstrate that multi-hop global influence estimation yields a favorable accuracy--memory trade-off for scalable long-context inference.

Submitted to arXiv on 16 Nov. 2026