arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

Annual Meeting of the Association for Computational Linguistics · 会议 · Natural Language Processing

共收录 10305
2502.12611 2025-12-30 cs.CL

Who Writes What: Unveiling the Impact of Author Roles on AI-generated Text Detection

谁写什么:揭示作者角色对AI生成文本检测的影响

Jiatao Li, Xiaojun Wan

机构 * Wangxuan Institute of Computer Technology, Peking University(北京大学计算机技术研究院)

AI总结 研究揭示作者角色(如性别、CEFR熟练度等)对AI生成文本检测的影响,提出多因素分析方法,为更公平的检测系统提供实证支持和框架。

Comments ACL 2025 Main Conference

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.21933 2025-12-29 cs.CL

Broken Words, Broken Performance: Effect of Tokenization on Performance of LLMs

破碎的词,破碎的性能:分词对大语言模型性能的影响

Sachin Pawar, Manoj Apte, Kshitij Jadhav, Girish Keshav Palshikar, Nitin Ramrakhiyani

机构 * TCS Research(塔塔咨询服务研究)

AI总结 本文研究了分词对大语言模型性能的影响,提出惩罚函数量化分词质量,并验证其在多个NLP任务上的统计显著性。

Comments International Joint Conference on Natural Language Processing & Asia-Pacific Chapter of the Association for Computational Linguistics (IJCNLP-AACL 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.05622 2025-12-29 cs.RO cs.AI

CityNavAgent: Aerial Vision-and-Language Navigation with Hierarchical Semantic Planning and Global Memory

CityNavAgent:基于层次语义规划和全局记忆的空中视觉-语言导航

Weichen Zhang, Chen Gao, Shiquan Yu, Ruiying Peng, Baining Zhao, Qian Zhang, Jinqiang Cui, Xinlei Chen, Yong Li

机构 * Tsinghua University(清华大学)

AI总结 CityNavAgent通过层次语义规划和全局记忆模块,提升空中视觉-语言导航在复杂城市环境中的导航性能。

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.06157 2025-12-29 cs.CV cs.AI

UrbanVideo-Bench: Benchmarking Vision-Language Models on Embodied Intelligence with Video Data in Urban Spaces

UrbanVideo-Bench: 基于城市空间视频数据评估视觉-语言模型的具身智能

Baining Zhao, Jianjie Fang, Zichao Dai, Ziyou Wang, Jirong Zha, Weichen Zhang, Chen Gao, Yue Wang, Jinqiang Cui, Xinlei Chen, Yong Li

机构 * Tsinghua University(清华大学)

AI总结 UrbanVideo-Bench通过城市空间视频数据评估视频大语言模型的具身智能,揭示其在城市环境中感知、推理和导航能力的局限性,并验证了Sim-to-Real迁移的潜力。

Comments 22 pages

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pp. 32400-32423, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.04661 2025-12-24 cs.RO

DRAE: Dynamic Retrieval-Augmented Expert Networks for Lifelong Learning and Task Adaptation in Robotics

DRAE:动态检索增强专家网络用于机器人终身学习与任务适应

Yayu Long, Kewei Chen, Long Jin, Mingsheng Shang

机构 * Chongqing Institute of Green and Intelligent Technology, Chinese Academy of Sciences(重庆绿色智能技术研究所,中国科学院)

AI总结 DRAE通过动态检索增强专家网络实现机器人终身学习和任务适应,显著提升长期任务保留和知识重用能力。

Comments Accepted to the main conference of the Annual Meeting of the Association for Computational Linguistics (ACL 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.03476 2025-12-23 cs.CL

Delta-KNN: Improving Demonstration Selection in In-Context Learning for Alzheimer's Disease Detection

Delta-KNN: 提高阿尔茨海默病检测中基于上下文学习的演示选择

Chuyuan Li, Raymond Li, Thalia S. Field, Giuseppe Carenini

机构 * Department of Computer Science(计算机科学系) Vancouver Stroke Program and Division of Neurology, Faculty of Medicine(温哥华卒中计划和神经病学系,医学院) The University of British Columbia(不列颠哥伦比亚大学)

AI总结 Delta-KNN通过Delta评分和KNN检索器提升基于上下文学习的阿尔茨海默病检测性能,实现新状态的最先进结果。

Journal ref In Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). ACL 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2503.02324 2025-12-23 cs.CL cs.AI cs.LG

PromptCoT: Synthesizing Olympiad-level Problems for Mathematical Reasoning in Large Language Models

PromptCoT: 为大型语言模型中的数学推理合成竞赛级问题

Xueliang Zhao, Wei Wu, Jian Guan, Lingpeng Kong

机构 * The University of Hong Kong(香港大学) Ant Group(蚂蚁集团)

AI总结 PromptCoT通过模拟竞赛级问题设计者的思维过程,自动生成高质量数学问题,提升大型语言模型的数学推理能力。

Comments Preprint

Journal ref Findings of the Association for Computational Linguistics: ACL 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.07424 2025-12-23 cs.CL cs.AI

RomanLens: The Role Of Latent Romanization In Multilinguality In LLMs

RomanLens: 潜在罗曼化在LLMs多语言性中的作用

Alan Saji, Jaavid Aktar Husain, Thanmay Jayakumar, Raj Dabre, Anoop Kunchukuttan, Ratish Puduppully

机构 * Nilekani Centre at AI4Bharat(AI4Bharat 奈尔肯中心) Singapore University of Technology and Design(新加坡科技设计大学) Indian Institute of Technology Madras(印度理工学院马德拉斯分校) National Institute of Information and Communications Technology(信息与通信技术国家研究所) Indian Institute of Technology Bombay(印度理工学院孟买分校) Microsoft(微软) IT University of Copenhagen(哥本哈根IT大学)

AI总结 RomanLens研究了罗曼化在LLMs多语言处理中的潜在作用,发现中间层常以罗曼化形式表示目标词,并通过实验表明罗曼化有助于语言转换。

Comments 19 pages, 19 figures

Journal ref Findings of the Association for Computational Linguistics: ACL 2025, pages 26410-26429, Vienna, July 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2507.23358 2025-12-22 cs.CL cs.AI cs.DB cs.IR

Text-to-SQL Task-oriented Dialogue Ontology Construction

面向任务的对话本体构建

Renato Vukovic, Carel van Niekerk, Michael Heck, Benjamin Ruppik, Hsien-Chin Lin, Shutong Feng, Nurul Lubis, Milica Gasic

机构 * Heinrich Heine University Düsseldorf(海因里希-海涅大学杜塞尔多夫)

AI总结 TeQoDO通过LLM自主构建面向任务的对话本体,结合模块化TOD系统概念,提升对话系统的可解释性和可控性。

Comments Accepted to Transactions of the Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.09587 2025-12-19 cs.CL

OpenNER 1.0: Standardized Open-Access Named Entity Recognition Datasets in 50+ Languages

OpenNER 1.0: 50余种语言标准化开放访问命名实体识别数据集

Chester Palen-Michel, Maxwell Pickering, Maya Kruse, Jonne Sälevä, Constantine Lignos

机构 * Michtom School of Computer Science(米切姆计算机科学学院) Brandeis University(布兰迪大学)

AI总结 OpenNER 1.0提供50余种语言的标准化NER数据集,通过修正标注格式和统一表示形式,支持多语言和多本体NER研究,并展示了不同预训练模型在NER任务中的性能差异。

Comments Published in the proceedings of EMNLP 2025

Journal ref Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 33637-33662, Suzhou, China. Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.15793 2025-12-19 cs.CY cs.AI cs.CL

Explainable Ethical Assessment on Human Behaviors by Generating Conflicting Social Norms

通过生成冲突的社会规范进行可解释的伦理评估

Yuxi Sun, Wei Gao, Hongzhan Lin, Jing Ma, Wenxuan Zhang

机构 * Department of Computer Science(计算机科学系) School of Computing and Information Systems(计算与信息学系) Information Systems Technology and Design(信息系统技术与设计)

AI总结 本文提出ClarityEthic方法,通过生成冲突的社会规范提升AI对人类行为效用的预测和解释能力。

Comments Acceppt by Asia-Pacific Chapter of the Association for Computational Linguistics (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.20764 2025-12-18 cs.CL

Feel the Difference? A Comparative Analysis of Emotional Arcs in Real and LLM-Generated CBT Sessions

感知差异?真实与LLM生成CBT对话中情感弧的比较分析

Xiaoyi Wang, Jiwei Zhang, Guangtao Zhang, Honglei Guo

机构 * Department of Computer Science, Shantou University(汕头大学计算机科学系) Department of Automation, Tsinghua University(清华大学自动化系) BNRIST, Tsinghua University(清华大学脑科学与类脑智能研究院)

AI总结 本研究通过比较真实与LLM生成的CBT对话,揭示了合成对话在情感动态上的不足,强调了情感真实性的关键作用。

Comments Accepted at 2025 EMNLP findings,19 page,2 figures

Journal ref In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 19999-20017

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.12839 2025-12-16 cs.CL

What Matters in Evaluating Book-Length Stories? A Systematic Study of Long Story Evaluation

在评估长篇故事时什么重要?对长篇故事评估的系统研究

Dingyi Yang, Qin Jin

机构 * Renmin University of China(中国人民大学)

AI总结 本文提出LongStoryEval基准和NovelCritique模型,通过系统研究揭示长篇故事评估的关键因素和有效方法。

Comments 24 pages, 7 figures, Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.19764 2025-12-16 cs.CL

Principled Personas: Defining and Measuring the Intended Effects of Persona Prompting on Task Performance

原则性人格:定义并衡量人格提示对任务表现的预期效果

Pedro Henrique Luz de Araujo, Paul Röttger, Dirk Hovy, Benjamin Roth

AI总结 本文研究了专家人格提示对任务表现的影响,发现其效果不一致,模型对无关属性敏感,提出缓解策略但仅对大模型有效,强调需更严谨的人格设计和评估方案。

Comments 30 pages, 29 figures, accepted to EMNLP 2025

Journal ref In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 26845-26874, Suzhou, China. Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.13405 2025-12-16 cs.CL

RealHiTBench: A Comprehensive Realistic Hierarchical Table Benchmark for Evaluating LLM-Based Table Analysis

RealHiTBench: 一个全面的现实感分层表格基准,用于评估基于LLM的表格分析

Pengzuo Wu, Yuhang Yang, Guangcheng Zhu, Chao Ye, Hong Gu, Xu Lu, Ruixuan Xiao, Bowen Bao, Yijing He, Liangyu Zha, Wentao Ye, Junbo Zhao, Haobo Wang

机构 * Zhejiang University(浙江大学) vivo Mobile Communication Co., Ltd(vivo移动通信有限公司) Institute of Computing Innovation, Zhejiang University(浙江大学计算机创新研究院)

AI总结 RealHiTBench是一个用于评估基于LLM的表格分析能力的综合现实感分层表格基准,通过多种输入格式和复杂结构的表格测试,验证了改进LLMs对表格层次结构感知的重要性。

Comments ACL 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.16763 2025-12-16 cs.CV

Self-Rewarding Large Vision-Language Models for Optimizing Prompts in Text-to-Image Generation

自奖励的大型视觉-语言模型用于优化文本到图像生成中的提示

Hongji Yang, Yucheng Zhou, Wencheng Han, Jianbing Shen

机构 * University of Macau(澳门大学)

AI总结 本文提出了一种自奖励的大型视觉-语言模型框架,用于优化文本到图像生成中的提示,通过统一求解器和奖励模型实现自我改进,实验表明其优于其他方法。

Comments Accepted by ACL2025 Findings

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.11998 2025-12-16 cs.CL

Direct Confidence Alignment: Aligning Verbalized Confidence with Internal Confidence In Large Language Models

直接置信对齐:将 verbalized 置信度与内部置信度对齐于大语言模型

Glenn Zhang, Treasure Mayowa, Jason Fan, Yicheng Fu, Aaron Sandoval, Sean O'Brien, Kevin Zhu

机构 * Algoverse AI Research(Algoverse AI研究院)

AI总结 本文提出直接置信对齐方法,通过优化使LLM的 verbalized 置信度与内部置信度对齐,提升模型透明度和可靠性。

Comments Accepted at ACL 2025 SRW, 5 pages body, 14 pages total

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.23315 2025-12-15 cs.CL cs.AI cs.LG

Enhancing Marker Scoring Accuracy through Ordinal Confidence Modelling in Educational Assessments

通过顺序置信建模提升教育评估中标记评分的准确性

Abhirup Chakravarty, Mark Brenchley, Trevor Breakspear, Ian Lewin, Yan Huang

机构 * Applied AI Cambridge University Press & Assessment(应用人工智能剑桥大学出版社与评估)

AI总结 本研究通过引入核加权有序分类交叉熵损失函数,提升教育评估中自动评分的置信度建模,实现更高的CEFR一致性评分。

Comments This is the preprint version of our paper accepted to ACL 2025 (Industry Track). The DOI will be added once available

Journal ref Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 6: Industry Track), 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.10996 2025-12-15 cs.CL cs.AI

MedBioRAG: Semantic Search and Retrieval-Augmented Generation with Large Language Models for Medical and Biological QA

MedBioRAG: 基于大语言模型的语义搜索与检索增强生成用于医学和生物学问答

Seonok Kim

机构 * Seonok Kim(独立研究者)

AI总结 MedBioRAG通过结合语义搜索、文档检索和监督微调,提升医学和生物学问答任务的性能。

Comments Submitted to ACL 2025. 9 pages, 4 figures, 5 tables (including 2 appendix tables)

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.07288 2025-12-09 cs.CL

Investigating Training and Generalization in Faithful Self-Explanations of Large Language Models

探究大语言模型忠实自解释的训练与泛化

Tomoki Doi, Masaru Isonuma, Hitomi Yanaka

机构 * The University of Tokyo(东京大学) Riken(理化学研究所) Tohoku University(东北大学) NII LLMC(日本信息处理学会大语言模型委员会)

AI总结 本研究通过训练提升大语言模型的自解释忠实性,并验证其在不同任务和风格中的泛化能力。

Comments To appear in the Proceedings of the Asia-Pacific Chapter of the Association for Computational Linguistics: Student Research Workshop (AACL-SRW 2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.14354 2025-12-09 cs.LG cs.CL

Self-Improvement Towards Pareto Optimality: Mitigating Preference Conflicts in Multi-Objective Alignment

迈向帕累托最优的自我改进:缓解多目标对齐中的偏好冲突

Moxin Li, Yuantao Zhang, Wenjie Wang, Wentao Shi, Zhuo Liu, Fuli Feng, Tat-Seng Chua

机构 * National University of Singapore(新加坡国立大学) University of Science and Technology of China(中国科学技术大学)

AI总结 本文提出一种自我改进的DPO框架,通过生成帕累托最优响应缓解多目标对齐中的偏好冲突,提升模型在帕累托前沿的优化效果。

Comments ACL findings (2025)

详情

展开后加载摘要…

URL PDF HTML 收藏
2502.11688 2025-12-09 cs.CL

From Isolates to Families: Using Neural Networks for Automated Language Affiliation

从孤立语到语系:利用神经网络进行自动语言归属

Frederic Blum, Steffen Herbold, Johann-Mattis List

机构 * Department of Linguistic and Cultural Evolution, Max-Planck Institute for Evolutionary Anthropology(语言与文化进化部门,马克斯·普朗克进化人类学研究所) University of Passau, Chair of Multilingual Computational Linguistics(帕绍大学,多语计算语言学教授团) University of Passau, Chair of AI Engineering(帕绍大学,人工智能工程教授团)

AI总结 本文提出利用神经网络模型,通过词汇和语法数据自动分类语言归属,展示模型在识别语言关系和辅助语言归属研究中的应用。

Comments Submitted to the 63rd Annual Meeting of the Association for Computational Linguistics, Vienna, Austria

详情

展开后加载摘要…

URL PDF HTML 收藏
2412.08519 2025-12-09 cs.CL

Bridging Relevance and Reasoning: Rationale Distillation in Retrieval-Augmented Generation

弥合相关性与推理:检索增强生成中的推理蒸馏

Pengyue Jia, Derong Xu, Xiaopeng Li, Zhaocheng Du, Xiangyang Li, Yichao Wang, Yuhao Wang, Qidong Liu, Maolin Wang, Huifeng Guo, Ruiming Tang, Xiangyu Zhao

机构 * City University of Hong Kong(香港城市大学) University of Science and Technology of China(中国科学技术大学) Huawei Noah’s Ark Lab(华为诺亚实验室)

AI总结 RADIO通过推理提取和基于推理的对齐方法,弥合检索增强生成中重排器与生成器之间的相关性差距。

Comments Accepted to ACL 25 Findings

详情

展开后加载摘要…

URL PDF HTML 收藏
2411.06272 2025-12-09 cs.CL cs.CE

Golden Touchstone: A Comprehensive Bilingual Benchmark for Evaluating Financial Large Language Models

黄金触点:一种全面的双语基准,用于评估金融大语言模型

Xiaojun Wu, Junxi Liu, Huanyi Su, Zhouchi Lin, Yiyan Qi, Chengjin Xu, Jiajun Su, Jiajie Zhong, Fuwei Wang, Saizhuo Wang, Fengrui Hua, Jia Li, Jian Guo

机构 * IDEA Research(IDEA研究机构) The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州)) The Hong Kong University of Science and Technology(香港科学与技术大学) Nanjing University(南京大学) South China Normal University(华南师范大学) DataArcTech Ltd.(DataArcTech有限公司)

AI总结 本研究提出Golden Touchstone双语基准,用于全面评估金融大语言模型的性能,通过对比分析揭示模型在处理复杂金融信息时的优势与局限。

Comments Published in Findings of EMNLP 2025

Journal ref In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 22544-22560, Suzhou, China. Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.06814 2025-12-09 cs.CL cs.AI

CAuSE: Decoding Multimodal Classifiers using Faithful Natural Language Explanation

CAuSE:利用忠实的自然语言解释解码多模态分类器

Dibyanayan Bandyopadhyay, Soham Bhattacharjee, Mohammed Hasanuzzaman, Asif Ekbal

机构 * Indian Institute of Technology Patna(印度理工学院帕纳瓦分校) Queen’s University Belfast(贝尔法斯特女王大学)

AI总结 CAuSE提出了一种生成多模态分类器忠实自然语言解释的新框架,通过因果抽象和交换干预提升解释的因果忠实性,并在多模态设置中验证其有效性。

Comments Accepted at Transactions of the Association for Computational Linguistics (TACL). Pre-MIT Press publication version

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.20833 2025-12-09 cs.CL

LLMs are Biased Evaluators But Not Biased for Retrieval Augmented Generation

LLMs是偏向评估者但不偏向检索增强生成

Yen-Shan Chen, Jing Jin, Peng-Ting Kuo, Chao-Wei Huang, Yun-Nung Chen

机构 * National Taiwan University(国立台湾大学)

AI总结 研究发现LLM在RAG框架中无自我偏好,但事实准确性显著影响输出。

Comments 15 pages, 14 tables, 5 figures Accepted to ACL Findings 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2506.08123 2025-12-05 cs.CL

QA-LIGN: Aligning LLMs through Constitutionally Decomposed QA

QA-LIGN:通过宪法分解的问答对对齐大语言模型

Jacob Dineen, Aswin RRV, Qin Liu, Zhikun Xu, Xiao Ye, Ming Shen, Zhaonan Li, Shijie Lu, Chitta Baral, Muhao Chen, Ben Zhou

AI总结 QA-LIGN通过分解奖励信号提升LLM对齐效果,降低攻击成功率并保持低拒绝率,实现安全与帮助性的帕累托最优。

Comments Findings of the Association for Computational Linguistics: EMNLP 2025, pages 20619-20642, Suzhou, China

Journal ref Findings of the Association for Computational Linguistics: EMNLP 2025, pages 20619-20642, Suzhou, China, 2025

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.14607 2025-12-04 cs.CL cs.CR

sudoLLM: On Multi-role Alignment of Language Models

sudoLLM: 关于语言模型的多角色对齐

Soumadeep Saha, Akshay Chaturvedi, Joy Mahapatra, Utpal Garain

机构 * ISI Kolkata(印度Kolkata ISI研究所) IRIT Toulouse(法国图卢兹 IRIT 研究所)

AI总结 sudoLLM通过注入用户偏见信号,实现多角色对齐的LLM,提升安全性和抗攻击能力。

Comments Accepted to EMNLP 2025 (findings)

Journal ref In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 366-384, Suzhou, China. Association for Computational Linguistics

详情

展开后加载摘要…

URL PDF HTML 收藏
2410.02660 2025-12-04 cs.CL cs.LG

How to Train Long-Context Language Models (Effectively)

如何有效训练长上下文语言模型

Tianyu Gao, Alexander Wettig, Howard Yen, Danqi Chen

机构 * Princeton Language and Intelligence(普林斯顿语言与智能)

AI总结 ProLong-8B 通过有效利用长上下文数据,实现了在长上下文任务上的卓越性能,尽管训练数据量仅为 Llama-3.1-8B-Instruct 的 5%。

Comments Accepted to ACL 2025. Our code, data, and models are available at https://github.com/princeton-nlp/ProLong

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