arXiv cs.AI· Quanjiang Guo, Sijie Wang, Qianyang Zhu, Ben Zhang, Jinchuan Zhang, Ke Yan, Zhao Kang·· 1 小时前AI 评分45
MemCredit：长程 Web 智能体决策时记忆从语义相关性转向决策效用
MemCredit: From Relevance to Utility for Decision-Time Memory in Long-Horizon Web Agents
AI 导读
研究者提出决策中心记忆框架 MemCredit，通过反事实干预估计单条记忆的边际贡献并蒸馏为轻量效用估计器，将记忆选择从语义相关性转向决策效用。在 ALFWorld 上，MemCredit 在 500-token 记忆预算下达到 86.57 SR@50，高于 BM25 检索的 65.67%。反事实分析显示记忆效用稀疏且强依赖状态，决策效用比相关性更适合长程智能体记忆管理。
正文 · 原文
Abstract:Large language model (LLM) agents increasingly operate over long Web interaction histories, making memory management critical for reliable and efficient decision making. Existing methods primarily optimize storage, retrieval, or compression, but the memories most semantically relevant to the current state are not necessarily those that most improve the agent's action. We introduce MemCredit, a decision-centric memory framework that shifts memory selection from semantic relevance to decision utility. MemCredit estimates the state-, reference-context-, and policy-conditional marginal contribution of individual memories through controlled counterfactual interventions, and distills these intervention-derived credits into a lightweight utility estimator. At inference time, semantic relevance is used only for high-recall candidate generation, while predicted utility arbitrates which memories are retained under a fixed token budget, without modifying the underlying agent policy or requiring additional environment interaction. We evaluate MemCredit on ALFWorld and WebShop against representative memory baselines. On ALFWorld, MemCredit achieves 86.57 SR@50 under a 500-token historical-memory budget, compared with 65.67% for BM25 retrieval. Counterfactual analysis further shows that memory utility is sparse, strongly state dependent, and poorly captured by semantic relevance. These results suggest that decision utility provides a more effective principle than relevance alone for long-horizon agent memory management.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.15600 [cs.AI] |
| (or arXiv:2611.15600v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.15600 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Quanjiang Guo [view email]
[v1]
Wed, 11 Nov 2026 15:17:38 UTC (2,320 KB)
来源：arXiv cs.AI · arxiv.org