Retrieval-augmented generation promises to ground language model outputs in external evidence, yet the field has no reliable way to verify whether retrieved context actually governs generation -- a prerequisite for any high-stakes deployment. The standard assumption, that context-consistent output implies context-governed output, breaks when the retrieved document overlaps with the model's pretraining data: the model can produce faithful-looking text entirely from parametric memory, and both pathways yield indistinguishable output. We name this failure the attribution blind spot and introduce Computational Reality Monitoring (CRM) to address it. CRM operationalizes a principle adapted from cognitive science's reality monitoring framework: comparing internal representations with and without context reveals membership-conditioned representational divergence that output-level monitors systematically miss. CRM does not certify which source an individual generation used; it detects whether pretraining exposure leaves a measurable internal trajectory signature, establishing a necessary substrate for source attribution. Across nine model variants spanning three families, this divergence concentrates in architecture-specific layer patterns, receives converging support from block-level noise intervention, and generalizes across tasks and datasets while collapsing on domain-confounded benchmarks. The attribution blind spot is measurable and partially addressable: internal representations carry a diagnostic signal invisible at the output level, establishing a foundation for systems whose internal awareness of evidence provenance governs their external behavior.
The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language Models
Zhejiang University(浙江大学);Binjiang Institute of Zhejiang University(浙江大学滨江研究院);East China Normal University(华东师范大学);National FinTech Evaluation Center (Bank Card Testing Center)(国家金融科技测评中心(银行卡检测中心));Hangzhou Dianzi University(杭州电子科技大学);Shanghai Jiaotong University(上海交通大学)
Context Attribution with Multi-Armed Bandit Optimization
Lucy Family Institute for Data & Society, University of Notre Dame(数据与社会学院卢西家庭研究所,圣约翰大学);Department of Computer Science & Engineering, University of Notre Dame(计算机科学与工程系,圣约翰大学);IBM Research(IBM研究院)
Chain of Evidence: Pixel-Level Visual Attribution for Iterative Retrieval-Augmented Generation
National Engineering Research Center for Software Engineering, Peking University(软件工程国家级工程研究中心,北京大学);City University of Hong Kong(香港城市大学);Peking University(北京大学);Tencent Technology(腾讯科技)
提出Chain of Evidence (CoE)框架,利用视觉语言模型直接对检索到的文档截图进行推理,输出精确边界框以可视化完整推理链,解决迭代检索增强生成中的粗粒度归因和视觉语义丢失问题。
Attention with Dependency Parsing Augmentation for Fine-Grained Attribution
Key Lab of Intelligent Information Processing of Chinese Academy of Sciences (CAS)(中国科学院智能信息处理重点实验室);Institute of Computing Technology, CAS(中国科学院计算技术研究所);University of Chinese Academy of Sciences, CAS(中国科学院大学);Peng Cheng Laboratory(鹏城实验室)
Model Internals-based Answer Attribution for Trustworthy Retrieval-Augmented Generation
Center for Language and Cognition (CLCG), University of Groningen(格罗宁根大学语言与认知中心(CLCG));Institute for Logic, Language and Computation (ILLC), University of Amsterdam(阿姆斯特丹大学逻辑、语言与计算研究所(ILLC))
MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection
Institute of Information Engineering, Chinese Academy of Sciences(中国科学院信息工程研究所);Qiyuan Tech(齐元科技);Laboratory of Multimedia Information Processing, School of Computer Science, Peking University(北京大学计算机科学学院多媒体信息处理实验室);School of Cyber Security, University of Chinese Academy of Sciences(中国科学院大学网络安全学院)
Disentangling Hallucinations: Orthogonal Semantic Projection for Robust Interpretability
U2IS, ENSTA, Institut Polytechnique de Paris(巴黎综合理工学院ENSTA学院U2IS实验室);ISIR, Université Sorbonne, Pierre et Marie Curie(索邦大学皮埃尔和玛丽·居里分校ISIR实验室);AMIAD, Pôle Recherche(AMIAD研究部)
On the Consistency of Multilingual Context Utilization in Retrieval-Augmented Generation
Center for Language and Cognition (CLCG), University of Groningen(格罗宁根大学语言与认知中心);Institute for Logic, Language and Computation (ILLC), University of Amsterdam(阿姆斯特丹大学逻辑、语言与计算研究所)
Enhancing Large Language Models with Domain-specific Retrieval Augment Generation: A Case Study on Long-form Consumer Health Question Answering in Ophthalmology
Yale University(耶鲁大学);National Eye Institute, National Institutes of Health(美国国立卫生研究院国家眼科研究所);Singapore Eye Research Institute(新加坡眼科研究所);Singapore National Eye Centre(新加坡国家眼科中心);National Center for Biotechnology Information, National Library of Medicine, National Institutes of Health(美国国立卫生研究院国家医学图书馆国家生物技术信息中心);Rice University(莱斯大学);National University of Singapore(新加坡国立大学)