Summary
Large language models can remember past conversations using a special memory called KV cache, but when many users share this memory, it's hard to make sure only the right user accesses their own data. The authors propose Bio-MemArt, a system that attaches biometric data like face or palmprint templates to each memory piece, allowing the model to check if the current user is authorized before using that memory. Their tests show this method accurately trusts the rightful user while blocking others, and it keeps the memory use very efficient by reducing how much conversation has to be reprocessed. This approach adds physical user verification into AI memory sharing without slowing down the system.
What this means in practice
- •For cloud platform operators: Secure shared memory for multi-user AI services by enforcing biometric-based access to reused context data.
- •For mobile device developers: Incorporate biometric gating to improve privacy and efficiency of on-device LLM memory caching.
Authors
Yanhong Qian, Xuanying He, Qingguo Meng, Shihao Ding, Xingbo Dong, Zhe Jin
Abstract
KV cache is evolving from a serving optimization into an external memory substrate for long-term LLM agents. In a shared multi-user deployment, however, reusable KV blocks introduce a missing access-control question: semantic relevance alone cannot determine whether a memory block is authorized for the current physical user. We propose Bio-MemArt, a biometric-aware KV-cache memory framework for multi-user LLM agents. Bio-MemArt attaches a normalized biometric template to each stored KV memory block, filters the shared memory pool with the current user's biometric probe, and then runs the original MemArt retrieval and KV reuse pipeline only inside the authorized candidate pool. This design preserves latent-space retrieval, direct cache reuse, and decoupled position encoding while adding physical-user access control to shared KV memory. We evaluate Bio-MemArt under Owner and Non-owner query conditions on long-term dialogue QA with face and palmprint benchmarks. Across face benchmarks, the average owner and non-owner biometric success rates are 95.71% and 0.86%; across palmprint benchmarks, they are 97.60% and 2.00%. In the efficiency study, average prefill tokens drop from 18,781.96 under full-context prompting to 28.57 with Bio-MemArt, showing that biometric gating preserves the low-token operating regime of KV-cache memory.