跳到正文
arXiv cs.AI· Jingyu Liu, Zongze Li, Zach Xu, Zhanhui Zhou, Tahseen Rabbani, Dawn Song, Ce Zhang·· 56 分钟前AI 评分58

BRIDGE 论文:最小化 RAG 记忆在标准基准上打平 SOTA,并提出五类对抗探针检验智能体记忆

RAG in a Trenchcoat: When Minimal Memory Is Enough for Agentic Systems, and When It Isn't

AI 导读

arXiv 论文(arXiv:2611.22179,NeurIPS 2026)提出 BRIDGE,一个仅用链式观察笔记、BM25 混合余弦检索和按时间渲染的最小 RAG 记忆系统。

正文 · 原文

View PDF HTML (experimental)

Abstract:Agentic memory research has accumulated extraction, supersession, reflection, and graph synthesis on top of retrieval-augmented generation. However, we found that almost none of those primitives earn their cost in our tested memory benchmarks: we show this with BRIDGE, a deliberately minimal RAG memory with chained observation notes, hybrid BM25 with cosine retrieval, and chronological rendering, which reaches parity within bootstrap CI with the prior SOTA on LongMemEval-S and BEAM, and beats the best memory baseline on EverMemBench by 5.7-6.3 points at matched reader. It runs at ~5-7K reader-prompt tokens per query, using a single LLM call for each ingestion or query. We argue that this parity is itself the result: standard benchmarks score what memory retrieves, not what memory enables when the agent acts. To test the latter, we introduce five adversarial probes (Rule Application +-, Pattern Induction Style / Procedural, Dynamic Multiturn) judged at generation time by a cross vendor verification chain, where each query can only be solved by applying a remembered rule or pattern it never mentions. Across 5 probes x 20 scenarios, no architecture wins universally: atomic fact extractors (LangMem) win rule application, while BRIDGE's block-summary memory wins pattern induction. This result holds across four domains, three graders, and three reader families. Each design fails where the other wins, for reasons a recall benchmark cannot see: atomic-fact extractors discard the form a demonstrated pattern takes, while block-summary memory retrieves the rule but the reader does not act on it, a gap that presenting the rule more prominently largely closes. What separates memory designs is not what they retrieve but what they enable an agent to do, and that is what agentic memory should be evaluated on.
Comments: 40th Conference on Neural Information Processing Systems (NeurIPS 2026), Main Conference
Subjects: Artificial Intelligence (cs.AI)
Cite as: arXiv:2611.22179 [cs.AI]
  (or arXiv:2611.22179v1 [cs.AI] for this version)
  https://doi.org/10.48550/arXiv.2611.22179

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Jingyu Liu [view email]
[v1] Tue, 17 Nov 2026 04:16:34 UTC (188 KB)

来源:arXiv cs.AI · arxiv.org