From Direction to Magnitude: How Multimodal Instruction-Tuning Reorganizes the Geometric Encoding of Identity-Specifying Prompts in Transformer Hidden States
从方向到大小:多模态指令微调如何在Transformer隐藏状态中重新组织身份指定提示的几何编码
Jorge A. Castillo, Marco Torres Yévenes, Juan Carlos Lanas
Comments18 pages, 7 tables. v2: corrections to the classical sources (Quintilian, Cicero) and to several cited figures, and Appendix B corpus statistics aligned to the delivered dataset; measurements, results and conclusions unchanged. Data, code, and the trained LoRA adapter: https://federicoboggia.binatomy.com/pubblicazioni/
Gauge dependence and structured-output corruption in sign-branched repetition penalties: measurements across models, inference stacks, and alternative repetition controls
From Actions to Understanding: Conformal Interpretability of Temporal Concepts in LLM Agents
从行为到理解:LLM代理中时间概念的符合可解释性
Trilok Padhi, Ramneet Kaur, Krishiv Agarwal, Adam D. Cobb, Daniel Elenius, Manoj Acharya, Colin Samplawski, Alexander M. Berenbeim, Nathaniel D. Bastian, Susmit Jha, Ugur Kursuncu, Anirban Roy
机构
*
Georgia State University(佐治亚州立大学)
;
Computer Science Lab, SRI(SRI计算机科学实验室)
;
Army Cyber Institute(陆军网络学院)
;
United States Military Academy(美国军事学院)
;
University of Florida(佛罗里达大学)
Aligning Language Model Benchmarks with Pairwise Preferences
将语言模型基准与成对偏好对齐
Marco Gutierrez, Xinyi Leng, Hannah Cyberey, Jonathan Richard Schwarz, Ahmed Alaa, Thomas Hartvigsen
机构
*
School of Data Science, University of Virginia(弗吉尼亚大学数据科学学院)
;
Imperial College London(伦敦帝国理工学院)
;
Thomson Reuters Foundational Research(汤姆森路透基础研究)
;
Department of Electrical Engineering and Computer Science, UC Berkeley and UCSF(伯克利大学电气工程与计算机科学系及旧金山大学)
Distribution Preference Optimization: A Fine-grained Perspective for LLM Unlearning
分布偏好优化:大语言模型遗忘的细粒度视角
Kai Qin, Jiaqi Wu, Jianxiang He, Haoyuan Sun, Yifei Zhao, Xu Wang, Bin Liang, Yongzhe Chang, Cheng Li, Tiantian Zhang, Houde Liu
机构
*
Tsinghua University(清华大学)
;
The Hong Kong University of Science and Technology(香港科技大学)
;
Jianghuai Advanced Technology Center(江淮先进技术中心)
;
University of Technology Sydney(悉尼大学)
Apparent Psychological Profiles of Large Language Models are Largely a Measurement Artifact
大语言模型的心理特征很大程度上是测量假象
Jelena Meyer, David Garcia, Dirk U. Wulff
机构
*
Max Planck Institute for Human Development(马克斯·普朗克人类发展研究所)
;
University of Konstanz(康斯坦茨大学)
;
Barcelona Supercomputing Center(巴塞罗那超级计算中心)
;
University of Basel(巴塞尔大学)
CommentsAccepted to Interspeech 2026. This publication is part of the project Responsible AI for Voice Diagnostics (RAIVD) with file number NGF.1607.22.013 of the research programme NGF AiNed Fellowship Grants, which is financed by the Dutch Research Council (NWO)
机构
*
Graduate School of Artificial Intelligence and Science, Rikkyo University(立命馆大学人工智能与科学研究生院)
;
AI Technical Sector, Mamezo Co., Ltd.(Mamezo公司人工智能技术部门)
;
AI Consulting Division, Mamezo Co., Ltd.(Mamezo公司人工智能咨询部门)