arXiv cs.AI· Qi Feng, Jicong Fan·· 3 天前AI 评分36
MemPursuit：从长期智能体记忆中组合证据以补全缺失信息
Recall What Is Missing: Composing Evidence from Long-Term Agent Memory
AI 导读
MemPursuit 是一种免训练机制，将长期记忆的读取视为查询条件下的证据组合，用稀疏凸查询重建的残差作为表示空间中"当前记忆集缺失什么"的代理，无需语言模型或 cross-encoder 逐对打分。在 LoCoMo 固定预算实验中，Top-10 端到端准确率提升 7.2 个百分点、Top-20 提升 5.5 个百分点，重排序延迟中位数较全批量 BGE-v2-m3 提速 32.9 倍。
正文 · 原文
Abstract:Under a limited context budget, a long-term agent must recall not merely memories that are individually relevant, but evidence that is collectively sufficient for the current query. Conventional memory access retrieves, scores, and truncates memories, often spending multiple context slots on redundant facts while leaving complementary evidence behind. We cast the read phase of long-term memory as query-conditioned evidence composition: each memory is valued by what it adds beyond the evidence already selected. We instantiate this operation with MemPursuit, a training-free mechanism that uses the residual of a sparse convex query reconstruction as a representation-space proxy for what the current memory set is missing. By reusing the query and candidate embeddings already available from retrieval, MemPursuit progressively composes nested evidence contexts without invoking a language model or cross-encoder to score each query--memory pair. The effectiveness of our method is verified on 3 popular benchmarks in comparison to strong baselines. In fixed-budget LoCoMo experiments, MemPursuit improves end-to-end accuracy by 7.2 percentage points at Top-10 and 5.5 percentage points at Top-20, with consistent gains across multiple memory frameworks, while achieving a 32.9$\times$ speedup in median reranking latency over fully batched BGE-v2-m3. Our implementation is available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.15642 [cs.AI] |
| (or arXiv:2610.15642v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.15642 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qi Feng [view email]
[v1]
Sun, 11 Oct 2026 14:18:11 UTC (4,566 KB)
来源：arXiv cs.AI · arxiv.org