arXiv cs.CL· Han Luo, Zhan Zhuang, Xu He, Jujia Zhao, Yu Li, Suqin Yuan, Dong Li, Lei Feng·· 6 天前AI 评分27
超越最终嵌入向量：从 token 级证据中检索记忆
Beyond the Final Embedding: Retrieving Memories from Token-Level Evidence
AI 导读
一种无需额外训练的稀疏 token 匹配方法，利用嵌入模型已计算的上下文 token 表示，让每个查询 token 在候选记忆中检索少量匹配 token，聚合得分后与整体记忆相似度结合排序。
正文 · 原文
Abstract:Memory-augmented assistants retrieve evidence from stored conversations and documents to answer queries. Most embedding-based retrievers compare single-vector representations of queries and memories, even when relevance depends on a small part of the content. Such local evidence can be matched to queries using the contextual token representations already computed by the same embedding models. Our method uses sparse token matching to improve retrieval without additional training. Each query token retrieves a small set of matching tokens across candidate memories. The matching scores are aggregated for each memory and combined with whole-memory similarity to rank memories. An optional low-rank query adapter can be trained with supervision while keeping the encoder and stored memory representations fixed. Our method improves retrieval across eight benchmarks and multiple model families without additional training. Query adaptation further increases the gains to 13.3 and 14.3 percentage points in average Recall@10 and NDCG@10 on Qwen3-Embedding-4B. These retrieval gains translate into higher question-answering accuracy on long-term memory benchmarks.
| Subjects: | Information Retrieval (cs.IR); Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.13277 [cs.IR] |
| (or arXiv:2610.13277v1 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2610.13277 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Han Luo [view email]
[v1]
Fri, 9 Oct 2026 09:27:55 UTC (777 KB)
来源：arXiv cs.CL · arxiv.org