arXiv cs.CL· Runze Jiang, Yan Wang, Boran Jin, Bingyu Zhu, Longtao Huang·· 11 小时前AI 评分26
仅定位弱点不够：面向大语言模型的状态感知递归知识改进方法 RKI
Weakness Localization Is Not Enough: State-Aware Recursive Knowledge Improvement for Large Language Models
AI 导读
研究者提出递归知识改进方法 RKI，将大语言模型的知识改进建模为状态感知的序贯干预过程，在 HotpotEval 和 SafeInternalEval 上平均超越现有知识注入方法 18.41 分。
正文 · 原文
Abstract:Knowledge improvement in large language models is shifting from predefined knowledge incorporation pipelines toward evaluation-guided updating. Existing methods use evaluation to localize model-specific knowledge weaknesses and guide targeted post-training. However, identifying what to improve does not necessarily provide sufficient information for deciding how to improve it. The same localized weakness can be compatible with distinct operational knowledge states: the relevant knowledge may be unavailable, available but difficult to access reliably, or accessible but ineffective for downstream use. We call this ambiguity knowledge-state aliasing. When compatible states favor different interventions, greater diagnostic resolution alone is insufficient for intervention selection. Moreover, model updates can change the underlying knowledge state before the observed weakness disappears, such that an intervention appropriate in one round may no longer address the bottleneck in the next. To address these challenges, we propose Recursive Knowledge Improvement (RKI), which formulates knowledge improvement as a sequential, state-aware intervention process. RKI organizes knowledge into graph-structured units that provide a shared basis for diagnostic evaluation and targeted intervention across rounds. At each round, RKI localizes knowledge weaknesses, estimates the model's operational knowledge state from diagnostic evidence and intervention history, selects a corresponding intervention, and re-evaluates the updated model before determining the next intervention. On HotpotEval and SafeInternalEval, RKI outperforms existing knowledge incorporation methods by 18.41 points on average. Further analyses show that knowledge states provide additional guidance for intervention selection beyond weakness localization and can change across model updates even when the observed weakness persists.
| Comments: | 18 pages, 8 figures |
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.36811 [cs.CL] |
| (or arXiv:2610.36811v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36811 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yan Wang [view email]
[v1]
Tue, 27 Oct 2026 08:12:37 UTC (3,426 KB)
来源：arXiv cs.CL · arxiv.org