arXiv cs.CL· Yixu Huang, Jiaqi W. Ma·· 2 天前AI 评分29
RIKA：面向持续学习智能体的可靠知识获取方法
Reliable Knowledge Acquisition for Continual Learning Agents
AI 导读
针对持续学习智能体从任务流中提取的知识约 12–33% 会损害后续任务表现的问题，研究者提出 RIKA（Reliable Knowledge Acquisition），将事实、概念、程序与元认知知识的分层提取与基于证据的知识准入机制结合，无需额外验证任务即可决定哪些主张进入智能体上下文。
正文 · 原文
Abstract:As agents powered by foundation models move into open-ended environments, continual learning from experience becomes essential. For agents built on models with fixed parameters, this learning requires reusing prior task experience in context to improve performance and efficiency on future tasks. Existing memory and experience-learning systems typically focus on either retained facts or procedures, and lack a mechanism for establishing the reliability of claims extracted from the ongoing task stream. Indeed, we find that approximately 12--33\% of entries produced by representative systems harm future task performance when reused. To address these limitations, we propose \textbf{RIKA} (Reliable Knowledge Acquisition), which combines hierarchical extraction of factual, conceptual, procedural, and metacognitive knowledge with Evidence-based Knowledge Admission, which uses source evidence and subsequent experience to govern which claims enter the task agent's context, without additional validation tasks. Across six task streams from Continual Learning Bench and four model backbones, RIKA achieves the highest aggregate score on every backbone.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG) |
| Cite as: | arXiv:2610.21739 [cs.CL] |
| (or arXiv:2610.21739v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.21739 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yixu Huang [view email]
[v1]
Thu, 15 Oct 2026 11:55:57 UTC (1,938 KB)
来源：arXiv cs.CL · arxiv.org