SAFE-KV: Causal Suffix Repair for Behavior-Preserving Non-Prefix KV Cache Reuse in RAG

arXiv:2610.15186 · cs.CL, cs.LG · Submitted 2026-10-10 · Read on arXiv

cs.CL, cs.LG

Submitted: 2026-10-10

Updated: 2026-10-10

Terminology

• KV cache — 1× in this paper · explained in 3 episodes
• KV cache reuse — 1× in this paper · explained in 2 episodes
• RAG — 1× in this paper · explained in 3 episodes

Sources

• Extending Context Window of Large Language Models via Positional Interpolation
• Retrieval-Augmented Generation for Large Language Models: A Survey
• Mistral 7B
• LMCache: An Efficient KV Cache Layer for Enterprise-Scale LLM Inference
• WebGPT: Browser-assisted question-answering with human feedback
• Passage Re-ranking with BERT
• Fast Transformer Decoding: One Write-Head is All You Need
• Qwen2.5 Technical Report

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