Contiguity, Not Importance: Budgeted Repair of Stale KV Caches After Document Edits

arXiv:2609.17983 · cs.AI · Submitted 2026-09-16 · Read on arXiv

cs.AI

Submitted: 2026-09-16

Updated: 2026-09-16

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• KV cache — 10× in this paper · explained in 3 episodes
• Benchmark — 9× in this paper · explained in 2 episodes
• RAG — 6× in this paper · explained in 3 episodes
• Accuracy — 3× in this paper · explained in 1 episode
• Consistency — 3× in this paper · explained in 3 episodes
• Agentic System — 2× in this paper · explained in 2 episodes
• KV cache reuse — 2× in this paper · explained in 2 episodes
• Provenance — 2× in this paper · explained in 4 episodes
• Steering — 2× in this paper · explained in 2 episodes
• Multi-Agent System — 1× in this paper · explained in 6 episodes

Sources

• Models Take Notes at Prefill: KV Cache Can Be Editable and Composable
• KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
• Leyline: KV Cache Directives for Agentic Inference
• MemGPT: Towards LLMs as Operating Systems
• ProphetKV: User-Query-Driven Selective Recomputation for Efficient KV Cache Reuse in Retrieval-Augmented Generation
• KVShare: An LLM Service System with Efficient and Effective Multi-Tenant KV Cache Reuse

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection