提出写入时筛选机制，让AI像人一样有选择地存信息并保留历史版本。
Selective Memory for Artificial Intelligence: Write-Time Gating with Hierarchical Archiving
- 用来源信誉、新颖性等综合评分，在存入时筛选知识，避免无效信息堆积。
- 在8:1干扰数据下准确率仍达100%，读取时过滤方法则降为0%。
- 适合需要长期记忆且对准确性要求高的AI系统，尤其无训练数据场景。
检索增强生成会无差别存储内容，导致噪声累积降低准确率；参数化方法将知识压缩到权重中，无法实现选择性更新。两者均未模拟生物记忆的特性：基于显著性进行编码过滤，并归档而非删除过时信息。本文提出写入时门控机制，利用来源声誉、新颖性、可靠性等复合显著性评分筛选输入知识对象，同时维护版本链以保留历史状态。在无黄金标准标签的LLM真实评估中，写入门控实现100%准确率，而无门控存储仅达13%。关键发现：当干扰比例升至8:1时，读取时过滤（Self-RAG）准确率降至0%，而写入门控维持100%，揭示写入端优于读取端的结构性优势。在维基百科（20实体）、程序生成药理数据及2026年arXiv论文上的验证确认该结果。门控优势随参数化记忆支持减弱而上升：维基百科+25个百分点，截断后arXiv+48个百分点，程序数据零训练知识下+65个百分点。信号消融实验表明方法不依赖与黄金标签相关的元数据。写入门控在仅九分之一查询成本下达到Self-RAG的准确率。
原文摘要 · Abstract (English)
Retrieval-augmented generation stores all content indiscriminately, degrading accuracy as noise accumulates. Parametric approaches compress knowledge into weights, precluding selective updates. Neither mirrors biological memory, which gates encoding based on salience and archives rather than deletes superseded information. We introduce write-time gating that filters incoming knowledge objects using composite salience scores (source reputation, novelty, reliability) while maintaining version chains that preserve prior states. Using real LLM evaluation without oracle access to quality labels, write gating achieves 100 percent accuracy versus 13 percent for ungated stores. The critical finding emerges under distractor scaling: at 8:1 distractor ratios, read-time filtering (Self-RAG) collapses to 0 percent while write gating maintains 100 percent, revealing a structural advantage of write-time over read-time curation. Validation on Wikipedia (20 entities), procedurally generated pharmacology data, and 2026 arXiv papers confirms these findings. The gating advantage scales inversely with parametric memory support: +25pp for Wikipedia, +48pp for post-cutoff arXiv, +65pp for procedural data with zero training knowledge. Signal ablation confirms the method does not depend on oracle-correlated metadata. Write gating matches Self-RAG accuracy at one-ninth the query-time cost.
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。