SocialLLM 2025
The 1st International Workshop on Large Language Models for Social Media
In conjunction with ACM TheWebConf 2025 @ April 28, 2025
The 1st International Workshop on Large Language Models for Social Media
In conjunction with ACM TheWebConf 2025 @ April 28, 2025
The program of SocialLLM@WWW2025 is announced. Please visit program.
The date/time of SocialLLM@WWW2025 is at 13:30 (local time) on April 28, 2025. (0325)
The deadline of SocialLLM@WWW2025 is extended to January 14th, 2025.
The submission website (via Easychair) can be accessed: https://easychair.org/conferences?conf=socialllm2025
SocialLLM 2025 webpage is online. (2024.12.20)
Since the rapid growth of large language models (LLMs) in 2022, many LLM services are now provided by leading companies such as OpenAI, Google, and Meta. Additionally, numerous LLMs that are free for commercial use are available. Companies can now utilize LLMs for a wide range of applications, and the general public can access them as well. Generative AI extends beyond text generation to include images, videos, speech, and music, enabling broader applications of this technology on social media platforms. With the rise of LLMs, the social media ecosystem is evolving rapidly. Content such as posts, tweets, and ads propagate at high speeds, and similarly, unverified information can be generated and spread just as quickly, including disinformation, propaganda, and even cyberbullying. In response, the natural language processing and machine learning community can leverage technology to help maintain a healthy environment for users.
While Natural Language Processing (NLP) techniques have been instrumental in analyzing social media through our previous SocialNLP workshops, they are often limited in handling the nuanced, multimodal, and real-time nature of modern social interactions. This is where LLMs bring transformative potential. LLMs not only excel in processing vast amounts of data but also demonstrate a deeper understanding of context, user intent, and conversational subtleties that are crucial in social media environments. Moving from SocialNLP to SocialLLM allows us to harness these advancements, enabling more accurate sentiment analysis, detection of misinformation, personalized content generation, and better user behavior prediction. By focusing on LLMs, we aim to push the boundaries of what's possible in social media analysis, ensuring that we are equipped to tackle emerging challenges with cutting-edge technology. This transition is not just an evolution but a necessary leap forward to stay relevant and effective in a rapidly changing digital landscape.
The SocialLLM workshop is a natural fit for The ACM Web Conference (WWW) 2025, building on our mission to explore the impact of Large Language Models (LLMs) on the understanding and evolution of social media. We propose hosting SocialLLM at WWW for several compelling reasons.
LLMs are at the forefront of NLP and have revolutionized how we interact with and interpret the vast, ever-growing landscape of web and social media content. Social media is a dynamic platform that produces immense quantities of user-generated data, characterized by its diversity, speed, and evolving nature. The WWW community has a strong interest in analyzing and understanding this data, and SocialLLM will foster a focused discussion on the unique challenges and opportunities LLMs present in this space. By hosting SocialLLM, WWW offers its participants a vital platform to delve into how LLMs are reshaping social media analysis, encompassing tasks such as sentiment analysis, misinformation detection, trend forecasting, and personalized user engagement.
Recent advancements in LLMs have opened new possibilities for social media research by enabling nuanced semantic understanding, multilingual analysis, and efficient data processing at unprecedented scales. This workshop aligns with the WWW's dedication to advancing web technologies by addressing crucial issues in analyzing and managing online social data. LLMs not only enhance our ability to comprehend textual data on social platforms but also uncover deeper social, psychological, and cultural insights. SocialLLM will encourage researchers from NLP, web science, and social computing to collaborate, explore novel applications, and address the ethical implications and technical challenges of leveraging LLMs in social media.
SocialLLM will bridge academic research and industry applications, encouraging discussions on practical implementations of LLM-driven social media analysis, from social recommendation systems to crisis management, user behavior prediction, and digital wellness. By gathering insights from top researchers and industry professionals, the workshop will foster innovation and the application of LLMs to real-world social media problems. We are confident that SocialLLM will create a high-impact forum for exchanging ideas, fostering interdisciplinary collaborations, and propelling future research on web-based NLP and social media.
We consider three plausible directions for SocialLLM: (1) addressing social computing issues using NLP techniques with LLMs; (2) mitigating social media challenges through LLMs; and (3) tackling new problems at the intersection of social computing and natural language processing in the LLM era. We focus on the strength of natural language processing techniques in social media content analysis, especially the use of LLMs in the Web environment.
We focus on the strength of natural language processing techniques in social media content analysis, especially in the Web environment. The topics of SocialLLM technical papers include (but not limited to):
LLMs on Social Media
Training and fine-tuning LLMs for personalized social interactions
LLM training with group and individual human characteristics
Evaluating personalized and socially-aware language understanding
Bias detection and mitigation in human language modeling
Data acquisition for human-contextualized LLMs
Ethical considerations and challenges in Social LLMs
Empirical findings and failure cases in building LLMs from Social Media
Real-world applications of LLMs for social media
LLM-based Social Content analysis
Concept-level sentiment analysis
Summarization of posts/replies on social media
Name entity Recognition on Social media
Relationship extraction on social media
Entity resolution for social media
Search, Indexing, and Evaluation on Social Web
Multilingual Information Retrieval on Social Web
NLP and LLMs on the Web
Folksonomy and Social Tagging
Trend analysis on Wikipedia
Trustworthiness analysis on Wikipedia
Human computing for social-media corpus generation
Social structure and position analysis using Microblog
Trust and Privacy analysis in social contexts
Community detection using blog or Microblog content
Sentiment and Opinion Analysis on Social Media
Big social data analysis
Lexical semantic resources, corpora and annotations of social media for sentiment analysis
Opinion classification, tracking and summarization
Domain specific sentiment analysis
Sarcasm, convincing language, and deception detection
Sentiment analysis for automatic public opinion poll and surveys of user satisfaction
Improvement of NLP tasks using subjectivity and/or sentiment analysis on social platform
Sentiment analysis and human computer interface on social platform
Real-world sentiment applications and systems
Disaster Management Using Social Media
Modeling global events or human activities based on social media texts
Identification and geo-location of social media content
Social-based web platform for disaster management
Disaster or disease prediction and forecasting
Resource allocation using social media
Monitoring emergency responses among social crowds
Analyzing the diffusion of emergent information
Exploiting social media for crisis response and search and rescue activities
Covid-19 on social media
Information Disorder on Social Media
Fake news detection on short texts
Cyberbullying detection
Hate speech detection
Clickbait detection
Malicious account detection
Robust misinformation detection
Flight again machine-generated fake messages
Multi-modal fake message detection
Spread prediction of disinformation
Explainable AI for information disorder
Date: April 28, 2025
Time Zone: Sydney Local Time (GMT+10)
Location: Room C4.5 @ ICC Sydney
** Each paper is with 13-minute presentation plus 2-minute QA. **
13:30 - 13:35 Opening
13:35 - 14:30 Keynote Speech
TBA
14:30 - 14:45 Few Labels with Active Learning: From Weak to Strong Labels for Misinformation Detection
Abdulrahman Alalawi, Abdullah Alsuhaibani, Usman Naseem, Basem Suleiman, Shoaib Jameel and Imran Razzak
14:45 - 15:00 Addressing the Challenges of Mental Health Conversations with Large Language Models
Usman Naseem, Siddhant Bikram Shah, Shuvam Shiwakoti, Surendrabikram Thapa and Imran Razzak
15:00 - 15:15 "Only ChatGPT gets me": an Empirical Anlaysis of GPT versus other Large Language Models for Emotion Detection in Text
Florian Lecourt, Madalina Croitoru and Konstantin Todorov
15:15 - 15:20 Closing Remark
SocialLLM @ TheWebConf 2025
Submission Site (via Easychair): https://easychair.org/conferences?conf=socialllm2025
Page Length: 4 (four) to 8 (eight) pages, with up to 2 additional pages for references and an optional Appendix (that can contain details on reproducibility, proofs, pseudo-code, etc).
Template: ACM template published in the ACM guidelines, selecting the generic “sigconf” sample. Submissions must adhere to the ACM template and format . Please remember to add Concepts and Keywords. Please use the template in traditional double-column format to prepare your submissions. For example, word users may use Word Interim Template, and LaTeX users may use the “sample-sigconf” template. Overleaf users may want to use the ACM proceedings template available in Overleaf.
Proceedings: TheWebConf 2025 proceedings (companion volume)
Workshop papers that have been previously published or are under review for another journal, conference or workshop should not be considered for publication. Papers must be submitted in PDF format. The PDF files must have all non-standard fonts embedded. Workshop papers must be self-contained and in English. Submissions that do not follow these guidelines, or do not view or print properly, will be rejected without review. Each submission will be evaluated by at least 2 program committee members.
SocialLLM review is double-blind. Therefore, please anonymize your submission: do not put the author(s) names or affiliation(s) at the start of the paper, and do not include funding or other acknowledgments in papers submitted for review. In addition to Regular Paper submission, we call for DATA PAPER this year. A data paper should include the details of the created dataset and an experiment illustrating how to use it. Authors should note it as a data paper using the author field and submit at least partial data as accompanied materials. The created dataset should be able to be downloaded or acquired through an application process freely. If the data paper is accepted, we will list the link for accessing the dataset in the SocialLLM website. Note that the review for data papers is also double-blind and it is authors' responsibility to avoid revealing their identities.
To pursue high quality submission, we will have a best paper award of SocialLLM 2025. The selection process will depend on not only the review comments/ratings, but also the quality of paper that is rated by paper authors. Selected, expanded versions of papers presented at the workshop will be published in two follow-on Special Issues of Springer Journal of Information Science and Engineering (JISE) and the International Journal of Computational Linguistics and Chinese Language Processing (IJCLCLP).
Submission Deadline: January 1, 2025 January 14, 2025
Author Notification: January 27, 2025 February 3, 2025
Camera-Ready Submission: February 7, 2025
Workshop Date: April 28, 2025
* All submission deadlines are end-of-day in the Anywhere on Earth (AoE) time zone.
Myrthe Reuver, Free University of Amsterdam
Jenq-Haur Wang, National Taipei University of Technology
Sundong Kim, Institute for Basic Science
Sungkyu Park, Institute for Basic Science (IBS)
Kunwoo Park, Soongsil University
Min-Yuh Day, National Taipei University
Shih-Hung Wu, Dept. of CSIE, Chaoyang University of Technology
Wen-Lian Hsu, Academia Sinica
Kuan-Yu Chen, National Taiwan University of Science and Technology
Muheng Yan, University of Pittsburgh
Ingmar Weber, Qatar Computing Research Institute
Tsung-Ting Kuo, University of California San Diego
Liang-Chih Yu, Yuan Ze University
Zhiqiang Zhong, University of Luxembourg
Lung-Hao Lee, National Central University
Yung-Chun Chang, Taipei Medical University
Bruno Martins, IST and INESC-ID - Instituto Superior Tecnico, University of Lisbon
Danilo Croce, Dept. of Enterprise Engineering - Univ. of Roma Tor Vergata
Hen-Hsen Huang, Institute of Information Science, Academia Sinica
Kai Wei, Amazon
Lun-Wei Ku, Academia Sinica, Taiwan
Cheng-Te Li, National Cheng Kung University, Taiwan
SocialNLP 2024 in TheWebConf 2024
SocialNLP 2023 in TheWebConf 2023 and AACL-IJCNLP 2023
SocialNLP 2022 in TheWebConf 2022 and NAACL 2022
SocialNLP 2021 in TheWebConf 2021 and NAACL 2021
SocialNLP 2020 in TheWebConf 2020 and ACL 2020
SocialNLP 2019 in IJCAI 2019
SocialNLP 2018 in WWW 2018 and ACL 2018
SocialNLP 2017 in EACL 2017 and IEEE BigData 2017
SocialNLP 2016 in IJCAI 2016 and EMNLP 2016
SocialNLP 2015 in WWW 2015 and NAACL 2015
SocialNLP 2014 in COLING 2014
SocialNLP 2013 in IJCNLP 2013
If you are considering submitting to the workshop and have questions regarding the workshop scope or need further information, please do not hesitate to send e-mail to lwku [AT] iis.sinica.edu.tw, chengte [AT] mail.ncku.edu.tw . Thanks!