Long-Lived Characters, Local Inference: Incremental Memory Maintenance for Game NPCs
Researchers present incremental KV-cache memory maintenance for long-lived game NPCs running locally on a quantized Qwen hybrid model.
The paper studies incremental memory maintenance for long-lived game NPCs deployed locally with a quantized Qwen hybrid recurrent-attention language model. The runtime removes superseded attention KV entries, computes replacement records at the true sequence tail, and preserves the continuing recurrent state and unchanged KV. Experiments across eight scripted maintenance rounds show true-tail updates preserve current-state and historical bindings, while slot-preserving alternatives repeat a double-subtraction error.
- Removes superseded KV entries while preserving recurrent state and unchanged KV
- True-tail updates maintain semantic bindings across eight scripted maintenance rounds
- Attention-distribution proximity alone does not explain semantic differences between update placements
From the original article
1 min read at arxiv.orgA game character should not have to reread its entire life before every conversation. For locally deployed language-model characters, however, revising a few memories can invalidate a long reusable prefix. The resulting preparation cost competes with both foreground dialogue and the maintenance of other characters.
By Zimu Xu · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.