arXivDaily arXiv每日学术速递 周一至周五更新

高校专区

University of Oxford(牛津大学)

2025-12-03 至 2025-12-03 共收录 5
2512.02898 2025-12-03 cs.SE cs.AI cs.LO cs.SC

Model-Based Diagnosis with Multiple Observations: A Unified Approach for C Software and Boolean Circuits

基于多观测的模型驱动诊断:面向C语言软件和布尔电路的统一方法

Pedro Orvalho, Marta Kwiatkowska, Mikoláš Janota, Vasco Manquinho

机构 * Department of Computer Science, University of Oxford(计算机科学系,牛津大学) Czech Technical University in Prague(布拉格捷克技术大学)

AI总结 CFaults通过结合多观测的模型驱动诊断方法,为C语言软件和布尔电路提供更高效的故障定位,确保诊断的一致性和最小性。

Comments 50 pages, 9 figures, 6 tables, 5 listings

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.02633 2025-12-03 cs.AI cs.LG

Zero-Shot Instruction Following in RL via Structured LTL Representations

通过结构化LTL表示实现强化学习中的零样本指令跟随

Mattia Giuri, Mathias Jackermeier, Alessandro Abate

机构 * University of Oxford(牛津大学)

AI总结 本文提出通过结构化LTL表示学习多任务策略,以解决强化学习中多事件交互复杂性问题。

Comments ICML 2025 Workshop on Programmatic Representations for Agent Learning

详情

展开后加载摘要…

URL PDF HTML 收藏
2509.13813 2025-12-03 cs.CL

Geometric Uncertainty for Detecting and Correcting Hallucinations in LLMs

几何不确定性用于检测和纠正大语言模型中的幻觉

Edward Phillips, Sean Wu, Soheila Molaei, Danielle Belgrave, Anshul Thakur, David Clifton

机构 * Department of Engineering Science, University of Oxford(牛津大学工程科学系) GlaxoSmithKline(葛兰素史克) Oxford Suzhou Centre for Advanced Research(牛津苏黎世高级研究中心)

AI总结 本文提出几何框架用于检测和纠正大语言模型中的幻觉,通过几何体积和几何怀疑方法提升响应可靠性。

Comments Revision. Clarified positioning as a unified geometric framework for global and local uncertainty in LLMs. Added baselines (Degree, Eccentricity) and expanded comparison to related methods. Included ablations (PCA dimension, number of archetypes, number of samples) and complexity analysis. Extended discussion of medical QA results and model-specific behaviour

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.03768 2025-12-03 cs.CL cs.CR cs.LG

Hidden in Plain Text: Emergence & Mitigation of Steganographic Collusion in LLMs

隐于 plain 文本:LLMs 中隐写术合谋的出现与缓解

Yohan Mathew, Ollie Matthews, Robert McCarthy, Joan Velja, Christian Schroeder de Witt, Dylan Cope, Nandi Schoots

机构 * LASR Labs(LASR实验室) University College London(伦敦大学学院) University of Amsterdam(阿姆斯特丹大学) University of Oxford(牛津大学)

AI总结 本文首次发现LLMs在训练期间因奖励激励设置不当而产生隐写术合谋,并指出现有缓解措施不足,需创新技术以防止此类合谋。

Comments Camera-ready version. Oral presentation at IJCNLP-AACL 2025 (14th International Joint Conference on Natural Language Processing and 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics), Mumbai, India, December 20-24, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.02188 2025-12-03 cs.CV

RobustSurg: Tackling domain generalisation for out-of-distribution surgical scene segmentation

RobustSurg: 解决领域泛化问题以实现分布外手术场景分割

Mansoor Ali, Maksim Richards, Gilberto Ochoa-Ruiz, Sharib Ali

机构 * St John’s College, University of Oxford(牛津大学圣约翰学院) School of Computer Science, University of Leeds(利兹大学计算机学院)

AI总结 RobustSurg通过实例归一化和特征协方差映射技术提升手术场景分割的领域泛化能力,并引入恢复模块保留关键特征,同时提供新数据集以解决多类多中心数据不足问题。

Comments Submitted to Medical Image Analysis

详情

展开后加载摘要…

URL PDF HTML 收藏