works on
inference optimization 1kv cache eviction 1large language models 1structural bias mitigation 1token role filtering 1

From the 1 of 1 linked paper with an AI index.

collaborators

1 paper

cs.CL2026

Adaptive Filtering of the KV Cache: Diagnosing and Correcting Structural-Role Bias in LLM Inference

Soumil Mandal

The paper identifies a structural-role bias in attention‑based KV‑cache eviction methods for long‑context LLMs, where delimiter and key tokens are over‑retained, and introduces a l…