arXiv cs.CL· Di Lu, Zechen Li, Hao Xue·· 11 小时前AI 评分38
Chain on Demand（CoD）：把长对话记忆的证据组织推迟到查询时
Chain on Demand: Deferred Evidence Organization for Long-Term Conversational Memory
AI 导读
研究者提出 Chain on Demand（CoD），一种在不变的轮级记忆之上、于查询时临时构建证据链的框架，按查询相关性、词汇多样性和新覆盖查询词分配固定锚点预算，再确定性还原锚点所在会话邻域。
正文 · 原文
Abstract:Long-term conversational memory faces a timing mismatch: write-time summaries and graphs organize history before a future query is known, while independent retrieval can allocate a limited evidence budget to redundant or fragmented turns. We introduce Chain on Demand (CoD), a query-time evidence organization framework over an unchanged turn-level memory. Given a broad candidate pool, CoD constructs a temporary evidence chain by allocating a fixed anchor budget according to query relevance, lexical diversity relative to selected anchors, and newly covered query terms, followed by deterministic restoration of each selected anchor's same-session neighborhood. The resulting chain is used only for the current query, while the underlying conversational memory remains unchanged. In a paired comparison over all 1,986 LoCoMo questions, CoD achieves 82.2% accuracy on 1,540 non-adversarial questions and 76.3% on all 1,986 questions. On LongMemEval-S, one fixed configuration that indexes both user and assistant turns obtains 86.6% across all 500 questions. These results show that long-term memory can preserve the original conversation while deferring evidence organization until the information need is known.
| Comments: | 19 pages, 2 figures, 6 tables |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.02480 [cs.CL] |
| (or arXiv:2611.02480v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.02480 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Di Lu [view email]
[v1]
Mon, 2 Nov 2026 03:54:52 UTC (269 KB)
来源：arXiv cs.CL · arxiv.org