语言大模型 / LLM

An evaluation of GPT models for phenotype concept recognition

Tudor Groza, Harry Caufield, Dylan Gration, Gareth Baynam, Melissa A Haendel, Peter N Robinson, Christopher J Mungall, Justin T Reese

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Incorporating Probing Signals into Multimodal Machine Translation via Visual Question-Answering Pairs

Yuxin Zuo, Bei Li, Chuanhao Lv, Tong Zheng, Tong Xiao, Jingbo Zhu

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Findings of EMNLP2023

We are what we repeatedly do: Inducing and deploying habitual schemas in persona-based responses

Benjamin Kane, Lenhart Schubert

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Code Representation Pre-training with Complements from Program Executions

Jiabo Huang, Jianyu Zhao, Yuyang Rong, Yiwen Guo, Yifeng He, Hao Chen

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Towards Vision-Language Mechanistic Interpretability: A Causal Tracing Tool for BLIP

Vedant Palit, Rohan Pandey, Aryaman Arora, Paul Pu Liang

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Final version for 5th Workshop on Closing the Loop Between Vision and Language (CLVL) @ ICCV 2023. 4 pages, 5 figures

Artificial-Spiking Hierarchical Networks for Vision-Language Representation Learning

Yeming Chen, Siyu Zhang, Yaoru Sun, Weijian Liang, Haoran Wang

专题命中 知识编辑与模型理解 ：foundation model(abstract)；pretraining(abstract)；分类 cs.CL、cs.AI

Rethinking the Role of Scale for In-Context Learning: An Interpretability-based Case Study at 66 Billion Scale

Hritik Bansal, Karthik Gopalakrishnan, Saket Dingliwal, Sravan Bodapati, Katrin Kirchhoff, Dan Roth

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Accepted at Annual Meeting of the Association for Computational Linguistics (ACL) 2023, Main Proceedings

Unveiling the Potential of Knowledge-Prompted ChatGPT for Enhancing Drug Trafficking Detection on Social Media

Chuanbo Hu, Bin Liu, Xin Li, Yanfang Ye

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

ANTM: An Aligned Neural Topic Model for Exploring Evolving Topics

Hamed Rahimi, Hubert Naacke, Camelia Constantin, Bernd Amann

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Finding Neurons in a Haystack: Case Studies with Sparse Probing

Wes Gurnee, Neel Nanda, Matthew Pauly, Katherine Harvey, Dmitrii Troitskii, Dimitris Bertsimas

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Emergent Linguistic Structures in Neural Networks are Fragile

Emanuele La Malfa, Matthew Wicker, Marta Kwiatkowska

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Controlled Text Generation with Hidden Representation Transformations

Vaibhav Kumar, Hana Koorehdavoudi, Masud Moshtaghi, Amita Misra, Ankit Chadha, Emilio Ferrara

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

Comments Accepted at ACL 2023 as a long paper (Findings)

Multivariate Representation Learning for Information Retrieval

Hamed Zamani, Michael Bendersky

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments Accepted for publication at SIGIR 2023

The Closeness of In-Context Learning and Weight Shifting for Softmax Regression

Shuai Li, Zhao Song, Yu Xia, Tong Yu, Tianyi Zhou

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Compressing Sentence Representation with maximum Coding Rate Reduction

Domagoj Ševerdija, Tomislav Prusina, Antonio Jovanović, Luka Borozan, Jurica Maltar, Domagoj Matijević

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Comments 14 pages, 3 figures, accepted on ICT and Electronics Convention (MIPRO), Croatia

Exploring the Representation Manifolds of Stable Diffusion Through the Lens of Intrinsic Dimension

Henry Kvinge, Davis Brown, Charles Godfrey

专题命中 知识编辑与模型理解 ：foundation model(abstract)；prompting(abstract)；分类 cs.CL、cs.LG

Comments 11 pages

A Study of Slang Representation Methods

Aravinda Kolla, Filip Ilievski, Hông-Ân Sandlin, Alain Mermoud

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.AI

DynaMaR: Dynamic Prompt with Mask Token Representation

Xiaodi Sun, Sunny Rajagopalan, Priyanka Nigam, Weiyi Lu, Yi Xu, Belinda Zeng, Trishul Chilimbi

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL、cs.LG

Scaling Laws and Interpretability of Learning from Repeated Data

Danny Hernandez, Tom Brown, Tom Conerly, Nova DasSarma, Dawn Drain, Sheer El-Showk, Nelson Elhage, Zac Hatfield-Dodds, Tom Henighan, Tristan Hume, Scott Johnston, Ben Mann, Chris Olah, Catherine Olsson, Dario Amodei, Nicholas Joseph, Jared Kaplan, Sam McCandlish

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.AI、cs.LG

Comments 23 pages, 22 figures

AdCare-VLM: Towards a Unified and Pre-aligned Latent Representation for Healthcare Video Understanding

Md Asaduzzaman Jabin, Hanqi Jiang, Yiwei Li, Patrick Kaggwa, Eugene Douglass, Juliet N. Sekandi, Tianming Liu

机构 * University of Georgia（佐治亚大学） 纠错

专题命中 知识编辑与模型理解 ：LLM(abstract)；language model(abstract)；分类 cs.AI；foundation model(comments)

Comments 39th Conference on Neural Information Processing Systems (NeurIPS 2025) Workshop: 7th International Workshop on Large Scale Holistic Video Understanding: Toward Video Foundation Models

Journal ref Neural Information Processing Systems (NeurIPS 2025)

Can Knowledge Editing Really Correct Hallucinations?

Baixiang Huang, Canyu Chen, Xiongxiao Xu, Ali Payani, Kai Shu

专题命中 知识编辑与模型理解 ：large language model(abstract)；language model(abstract)；分类 cs.CL；LLM(comments)

Comments ICLR 2025. Main paper: 10 pages; total: 34 pages (including appendix). The first two authors contributed equally to this work. Code, data, results, and additional resources are available on the project website: https://llm-editing.github.io

SARATR-X-v2: Scale-Aware Structural Pre-Training for SAR Foundation Models

SARATR-X-v2：合成孔径雷达基础模型的尺度感知结构预训练

Weijie Li, Yafei Song, Yongxiang Liu, Bowen Peng, Jie Zhou, Jingyuan Xia, Wei Yang, Tianpeng Liu, Zhen Liu, Li Liu

专题命中 知识编辑与模型理解 ：foundation model(title)

AI总结 研究针对SAR预训练中重建目标设计未确定的问题，提出应满足物理稳定性和语义尺度兼容性。通过SARATR-X-v2协调两者，其目标由固定结构提取器构建并融合成监督信号，在多个基准测试中性能最优，确立了预训练目标设计框架。

STeP: Signal Temporal Logic for Precise Specifications for Action Generation with Vision Language Models

STeP：用于视觉语言模型动作生成精确规范的信号时序逻辑

Kasra Torshizi, Anukriti Singh, Sidharth Mathur, Khuzema Habib, Leo Du, Pratap Tokekar

机构 * University of Maryland（马里兰大学） 纠错

专题命中 知识编辑与模型理解 ：language model(title)

AI总结 针对视觉语言动作模型缺乏可解释性及难以遵循精确自然语言指令的问题，提出用信号时序逻辑（STL）连接高级语言理解与低级机器人执行的分层框架，经实验验证该框架能提高语言条件下机器人规划的精度、可靠性和可解释性。

Comments 14 pages, 6 figures

An LLM-Based Automatic Sportscast Solution for Robot Soccer Matches

一种基于大语言模型的机器人足球比赛自动体育解说解决方案

Francesco Petri, Michele Brienza, Daniele Nardi, Domenico Daniele Bloisi, Aldo Gangemi, Vincenzo Suriani

机构 * Sapienza University of Rome（罗马第一大学） ； Institute for Cognitive Sciences and Technologies (ISTC-CNR)（认知科学与技术研究所（ISTC-CNR）） ； International University of Rome（罗马国际大学） ； University of Bologna（博洛尼亚大学） 纠错

专题命中 知识编辑与模型理解 ：LLM(title)

AI总结 针对机器人世界杯比赛，提出基于神经符号架构的自动体育解说方案，弥合运动跟踪与自然语言生成差距，能在直播和赛后生成统计数据与解说，适应不同机器人共享场地的联赛新动态。

Comments Poster presentation at RoboCup Symposium 2026

Emergent Region-Level Facial Correspondence in Frozen Vision Foundation Models

冻结视觉基础模型中新兴的区域级面部对应

Izaldein Al-Zyoud, Abdulmotaleb El Saddik

专题命中 知识编辑与模型理解 ：foundation model(title)

AI总结 研究人脸区域级对应问题，利用冻结的DINOv3特征及FaRL，在真实视频上评估跨身份匹配和时间标签传播，结果表明DINOv3是区域级面部对应强大零样本表示，中间自监督特征对密集面部分析最有用。

CoughPhase-CLR: Designing an acoustics-informed foundation model for coughing sound classification

CoughPhase-CLR：面向咳嗽声音分类的声学信息基础模型设计

Marius Moldovan, Anton Batliner, Thomas M. Berghaus, Björn W. Schuller, Andreas Triantafyllopoulos

机构 * Chair of Health Informatics at the TUM University Hospital（慕尼黑工业大学大学医院健康信息学教席） ； Munich Center for Machine Learning（慕尼黑机器学习中心） ； Munich Data Science Institute（慕尼黑数据科学研究所） ； University Hospital Augsburg at the University of Augsburg（奥格斯堡大学奥格斯堡大学医院） ； Medical Faculty, Ludwig Maximilians University of Munich（慕尼黑路德维希-马克西米利安大学医学院） ； Group on Language, Audio, & Music at Imperial College London（帝国理工学院语言、音频与音乐组） ； Technical University of Munich（慕尼黑工业大学） 纠错

专题命中 知识编辑与模型理解 ：foundation model(title)

AI总结 提出自监督学习框架CoughPhase-CLR，利用咳嗽的生理相位构建正样本对进行表征学习，在COVID-19检测、COPD状态分类等下游任务中优于标准随机裁剪方法。

Multi-Task Crack Foundation Model for Engineering-Reliable Crack Representation and Topology Preservation in Civil Infrastructure

面向工程可靠裂缝表示与拓扑保持的土木基础设施多任务裂缝基础模型

Blessing Agyei Kyem, Joshua Kofi Asamoah, Eugene Denteh, Armstrong Aboah

机构 * NDSU（内达苏大学） 纠错

专题命中 知识编辑与模型理解 ：foundation model(title)

AI总结 提出 CrackGeoFM 多任务框架，结合冻结视觉基础骨干与裂缝专用适配模块，实现掩码预测、骨架重建和不确定性估计，在20个数据集上达到最优分割、拓扑保持和校准不确定性。

Comments 60 pages, 17 figures, 11 tables

Vision-Language Model Based Multi-Expert Fusion for CT Image Classification

基于视觉-语言模型的多专家融合用于CT图像分类

Jianfa Bai, Kejin Lu, Runtian Yuan, Qingqiu Li, Jilan Xu, Junlin Hou, Yuejie Zhang, Rui Feng

机构 * College of Computer Science and Artificial Intelligence, Shanghai Key Laboratory of Intelligent Information Processing, Fudan University（复旦大学计算机科学与人工智能学院，上海智能信息处理重点实验室） ； University of Oxford（牛津大学） ； The Hong Kong University of Science and Technology（香港科技大学） 纠错

专题命中 知识编辑与模型理解 ：language model(title)

AI总结 本文提出一种三阶段源感知多专家框架，通过构建肺部感知3D专家、开发MedSigLIP基专家和训练源分类器，提升多源CT图像中新冠检测的鲁棒性，实验结果显示在不同阶段模型在宏F1、ACC和AUC指标上均取得优异成绩。

Sometimes You Need Facts, and Sometimes a Hug: Understanding Older Adults' Preferences for Explanations in LLM-Based Conversational AI Systems

有时你需要事实，有时你需要一个拥抱：理解老年人在基于LLM的对话AI系统中对解释的偏好

Niharika Mathur, Tamara Zubatiy, Agata Rozga, Jodi Forlizzi, Elizabeth Mynatt

专题命中 知识编辑与模型理解 ：LLM(title)

AI总结 研究探讨老年人在基于LLM的对话AI系统中对解释的偏好，通过实验发现解释的高情境依赖性及互动性，为设计更符合老年人需求的AI系统提供指导。

Comments To be published at ACM CHI 2026

RoomPlanner: Explicit Layout Planner for Easier LLM-Driven 3D Room Generation

RoomPlanner: 一种显式布局规划器，用于更易由LLM驱动的3D房间生成

Wenzhuo Sun, Mingjian Liang, Wenxuan Song, Xuelian Cheng, Zongyuan Ge

专题命中 知识编辑与模型理解 ：LLM(title)

AI总结 RoomPlanner通过显式布局规划和高效优化策略，实现了快速生成高质量3D室内场景。