EchoKV：基于迭代查询反馈的自校正 KV cache 淘汰方法
EchoKV: Self-Corrective KV Cache Eviction via Iterative Query Feedback
AI 导读
arXiv 论文提出 EchoKV，针对查询驱动的 KV cache 淘汰中查询与工作区耦合导致的错误累积问题，通过迭代校正修复工作区：先用未来 decode 位置的提议初始化工作区，再用早期 decode 生成的查询恢复工作区。
正文 · 原文
Abstract:KV cache eviction is not merely a token-ranking problem: the workspace retained for generating eviction queries also determines the quality of those queries. Consequently, an early workspace error can distort later query observations and reinforce an incorrect final eviction decision, particularly when evidence is dispersed across a long context. We present EchoKV, which discovers and addresses this failure mode with iterative correction. It initializes a feedback-compatible bootstrap workspace using proposals at future decode positions, then uses early decode-generated queries to recover the workspace before the remaining lookahead steps. It uses the same number of workspace forward passes as LAQ while adding a tensor re-scoring operation for workspace repair. On LongBench (16 tasks, Mistral-7B-Instruct-v0.2 and Qwen2.5-7B-Instruct), EchoKV shows notable improvements over LAQ on several tasks requiring dispersed evidence, including multi-hop cross-document QA and long single-document comprehension. Across budgets and backbones, EchoKV selectively improves evidence-intensive tasks while maintaining comparable performance on the remaining LongBench subsets. These results identify query--workspace coupling as a practical failure mode of query-driven KV eviction.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.05903 [cs.AI] |
| (or arXiv:2611.05903v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.05903 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jiankun Zou [view email]
[v1]
Wed, 4 Nov 2026 03:52:59 UTC (619 KB)
来源：arXiv cs.AI · arxiv.org