不用结构化和学习策略，纯确定性方法也能高效检索对话记忆。
SmartSearch: How Ranking Beats Structure for Conversational Memory Retrieval
- 用命名实体加权的子串匹配召回，规则驱动多跳扩展。
- 在两个基准上分别达到93.5%和88.4%准确率，优于所有已知系统。
- 全程无需调参，仅用CPU在650ms内完成，节省8.5倍上下文tokens。
近期对话记忆系统在录入时依赖大模型结构化，在查询时采用学习型检索策略。本文表明二者皆非必要。SmartSearch 从原始未结构化对话历史中检索，采用全确定性流程：基于命名实体加权的子串匹配实现召回，规则驱动的实体发现支持多跳扩展，以及仅有的学习组件——跨编码器与ColBERT的排序融合阶段，运行于CPU上耗时约650ms。在两个基准上的最优分析显示，检索召回率达98.6%，但若无智能排序，仅有22.5%的正确证据能保留至令牌预算限制。通过得分自适应截断且无需针对数据集调优，SmartSearch 在LoCoMo上达93.5%，在LongMemEval-S上达88.4%，均超越同评测协议下所有已知记忆系统，同时使用量仅为全上下文基线的8.5倍。
原文摘要 · Abstract (English)
Recent conversational memory systems invest heavily in LLM-based structuring at ingestion time and learned retrieval policies at query time. We show that neither is necessary. SmartSearch retrieves from raw, unstructured conversation history using a fully deterministic pipeline: NER-weighted substring matching for recall, rule-based entity discovery for multi-hop expansion, and a CrossEncoder+ColBERT rank fusion stage -- the only learned component -- running on CPU in ~650ms. Oracle analysis on two benchmarks identifies a compilation bottleneck: retrieval recall reaches 98.6%, but without intelligent ranking only 22.5% of gold evidence survives truncation to the token budget. With score-adaptive truncation and no per-dataset tuning, SmartSearch achieves 93.5% on LoCoMo and 88.4% on LongMemEval-S, exceeding all known memory systems under the same evaluation protocol on both benchmarks while using 8.5x fewer tokens than full-context baselines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。