arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~

高校专区

Johns Hopkins University(约翰斯·霍普金斯大学)

2026-04-10 至 2026-04-10 共收录 8
2604.08527 2026-04-10 cs.CL cs.LG

Demystifying OPD: Length Inflation and Stabilization Strategies for Large Language Models

解开OPD之谜:大型语言模型中的长度膨胀与稳定策略

Feng Luo, Yu-Neng Chuang, Guanchu Wang, Zicheng Xu, Xiaotian Han, Tianyi Zhang, Vladimir Braverman

机构 * Department of Computer Science, Rice University, Houston, USA(莱斯大学计算机科学系,美国休斯顿) Department of Computer Science, University of North Carolina at Charlotte, Charlotte, USA(北卡罗来纳大学夏洛特分校计算机科学系,美国夏洛特) Department of Computer Science, Johns Hopkins University, Baltimore, USA(约翰霍普金斯大学计算机科学系,美国巴尔的摩) Department of Computer and Data Sciences, Case Western Reserve University(凯斯西储大学计算机与数据科学系)

AI总结 本文研究了OPD训练中长度膨胀问题,提出StableOPD框架结合参考 divergence 约束和rollout混合蒸馏,有效稳定训练并提升性能。

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.08510 2026-04-10 cs.CL

What do Language Models Learn and When? The Implicit Curriculum Hypothesis

语言模型学习了什么以及何时?隐式课程假说

Emmy Liu, Kaiser Sun, Millicent Li, Isabelle Lee, Lindia Tjuatja, Jen-tse Huang, Graham Neubig

机构 * Language Technologies Institute, Carnegie Mellon University(卡内基梅隆大学语言技术研究所) Department of Computer Science, Data Science and AI Institute, Johns Hopkins University(约翰霍普金斯大学计算机科学系、数据科学与人工智能研究所) Khoury College of Computer Science, Northeastern University(东北大学Khoury计算机科学学院) Department of Computer Science, University of Southern California(南加州大学计算机科学系)

AI总结 研究通过设计简单可组合任务,发现模型技能以可组合顺序出现,且在不同模型间一致,表明预训练过程具有结构性。

详情

展开后加载摘要…

URL PDF HTML 收藏
2504.10284 2026-04-10 cs.CL

arXiv2Table: Toward Realistic Benchmarking and Evaluation for LLM-Based Literature-Review Table Generation

arXiv2Table:面向基于大语言模型的文献综述表生成的现实基准测试与评估

Weiqi Wang, Jiefu Ou, Yangqiu Song, Benjamin Van Durme, Daniel Khashabi

机构 * Center for Speech and Language Processing, Johns Hopkins University(约翰霍普金斯大学语音与语言处理中心) Department of Computer Science and Engineering, HKUST(香港科技大学计算机科学与工程系)

AI总结 本文提出arXiv2Table基准,通过模拟真实用户需求和噪声,评估大语言模型生成文献综述表的能力,实验表明方法优于基线,但任务难度仍高。

Comments ACL 2026 Main Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2409.02136 2026-04-10 cs.LG cs.AI cs.CL

Large Language Models versus Classical Machine Learning: Performance in COVID-19 Mortality Prediction Using High-Dimensional Tabular Data

大语言模型与经典机器学习:在使用高维表格数据预测新冠死亡率中的表现

Mohammadreza Ghaffarzadeh-Esfahani, Mahdi Ghaffarzadeh-Esfahani, Arian Salahi-Niri, Hossein Toreyhi, Zahra Atf, Amirali Mohsenzadeh-Kermani, Mahshad Sarikhani, Zohreh Tajabadi, Fatemeh Shojaeian, Mohammad Hassan Bagheri, Aydin Feyzi, Mohammadamin Tarighatpayma, Narges Gazmeh, Fateme Heydari, Hossein Afshar, Amirreza Allahgholipour, Farid Alimardani, Ameneh Salehi, Naghmeh Asadimanesh, Mohammad Amin Khalafi, Hadis Shabanipour, Ali Moradi, Sajjad Hossein Zadeh, Omid Yazdani, Romina Esbati, Moozhan Maleki, Danial Samiei Nasr, Amirali Soheili, Hossein Majlesi, Saba Shahsavan, Alireza Soheilipour, Nooshin Goudarzi, Erfan Taherifard, Hamidreza Hatamabadi, Jamil S Samaan, Thomas Savage, Ankit Sakhuja, Ali Soroush, Girish Nadkarni, Ilad Alavi Darazam, Mohamad Amin Pourhoseingholi, Seyed Amir Ahmad Safavi-Naini

机构 * Research Institute for Gastroenterology and Liver Diseases, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学胃肠病与肝病研究所) Faculty of Medicine, Isfahan University of Medical Sciences(伊斯法罕医科大学医学院) Faculty of Business and Information Technology, Ontario Tech University(安大略理工大学商业与信息技术学院) School of Medicine, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学医学院) Digestive Disease Research Institute, Tehran University of Medical Sciences(德黑兰医科大学消化疾病研究所) Department of Surgery, The Johns Hopkins University(约翰霍普金斯大学外科学系) Student Research Committee, School of Nursing and Midwifery, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学护理与助产学院学生研究委员会) MPH department, Shiraz University of Medical Sciences(设拉子医科大学公共卫生硕士系) Department of Emergency Medicine, School of Medicine, Safety Promotion and Injury Prevention Research Center, Imam Hossein Hospital, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学伊玛目侯赛因医院医学院急诊医学系安全促进与伤害预防研究中心) Karsh Division of Gastroenterology and Hepatology, Cedars-Sinai Medical Center(西达赛奈医疗中心卡什胃肠病与肝病科) Department of Medicine, Stanford University(斯坦福大学医学系) Division of Data Driven and Digital Health (D3M), The Charles Bronfman Institute for Personalized Medicine, Icahn School of Medicine at Mount Sinai(西奈山伊坎医学院查尔斯·布朗夫曼个性化医学研究所数据驱动与数字健康部) Infectious Diseases and Tropical Medicine Research Center, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学传染病与热带医学研究中心) Department of Infectious Diseases, Loghman Hakim Hospital, Shahid Beheshti University of Medical Sciences(沙希德·贝赫什提医科大学洛格曼·哈基姆医院传染病科) National Institute for Health and Care Research (NIHR), Nottingham Biomedical Research Centre, Hearing Sciences, Mental Health and Clinical Neurosciences, School of Medicine, University of Nottingham(诺丁汉大学医学院国家健康与护理研究所诺丁汉生物医学研究中心听力科学、心理健康与临床神经科学)

AI总结 本文比较了经典特征机器学习模型与大语言模型在预测新冠死亡率中的性能,发现经典模型在处理高维表格数据方面仍占优势,但通过微调大语言模型可显著提升其效果。

Comments Code is available at: https://github.com/mohammad-gh009/Large-Language-Models-vs-Classical-Machine-learning and https://github.com/Sdamirsa/Tehran_COVID_Cohort. The datasets are available from the corresponding author on reasonable request (sdamirsa@ymail.com)

Journal ref Scientific Reports 15, 42712 (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2604.07584 2026-04-10 cs.AI

From Papers to Property Tables: A Priority-Based LLM Workflow for Materials Data Extraction

从论文到属性表:一种基于优先级的LLM工作流用于材料数据提取

Koushik Rameshbabu, Jing Luo, Ali Shargh, Khalid A. El-Awady, Jaafar A. El-Awady

机构 * Department of Applied Mathematics and Statistics, Johns Hopkins University(约翰霍普金斯大学应用数学与统计系) Department of Mechanical Engineering, Johns Hopkins University(约翰霍普金斯大学机械工程系)

AI总结 本文提出一种基于优先级的LLM工作流,用于从科研论文中自动提取和重建结构化材料实验数据,通过整合文本、表格、图表和物理推导信息,以合金劈裂强度为案例,实现了高精度的数据提取与验证。

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.18749 2026-04-10 astro-ph.CO astro-ph.IM cs.LG cs.NE

Symbolic Emulators for Cosmology: Accelerating Cosmological Analyses Without Sacrificing Precision

符号模拟在宇宙学中的应用:在不牺牲精度的情况下加速宇宙学分析

Deaglan J. Bartlett, Shivam Pandey

机构 * Astrophysics, University of Oxford(牛津大学天体物理学) IAP(巴黎天体物理研究所) Department of Physics and Astronomy, Johns Hopkins University(约翰霍普金斯大学物理与天文学系)

AI总结 本文提出符号模拟方法,用于加速宇宙学分析,通过提高计算效率和内存使用,实现与传统数值方法一致的精度。

Comments 22 pages, 6 figures. Invited contribution for the Royal Society Philosophical Transactions A special issue "Symbolic regression in the physical sciences"

Journal ref Philos Trans A Math Phys Eng Sci (2026) 384 (2317): 20240585

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.00017 2026-04-10 cs.CL cs.AI cs.DB cs.LG

ReCellTy: Domain-Specific Knowledge Graph Retrieval-Augmented LLMs Reasoning Workflow for Single-Cell Annotation

ReCellTy:领域特定的知识图谱检索增强型大语言模型推理工作流用于单细胞注释

Dezheng Han, Yibin Jia, Ruxiao Chen, Wenjie Han, Shuaishuai Guo, Jianbo Wang

机构 * Shandong University(山东大学) Qilu Hospital of Shandong University(山东大学齐鲁医院) Johns Hopkins University(约翰霍普金斯大学) Cognicore Artificial Intelligence Co., Ltd.(科智人工智能有限公司)

AI总结 本文提出ReCellTy,通过整合生物信息知识图谱和多任务推理工作流,提升单细胞注释的准确性与自动化水平,改进人类评估得分和语义相似度,缩小大中小语言模型性能差距。

详情

展开后加载摘要…

URL PDF HTML 收藏
2411.02622 2026-04-10 cs.LG cs.AI

AdaProb: Efficient Machine Unlearning via Adaptive Probability

AdaProb: 通过自适应概率实现高效的机器遗忘

Zihao Zhao, Yuchen Yang, Anjalie Field, Yinzhi Cao

机构 * Johns Hopkins University(约翰霍普金斯大学) The Pennsylvania State University(宾夕法尼亚州立大学)

AI总结 本文提出AdaProb方法,通过自适应概率实现高效隐私保护的数据遗忘,实验显示其在遗忘误差、对抗攻击防护和计算效率方面均优于现有方法。

详情

展开后加载摘要…

URL PDF HTML 收藏