MML4SG Workshop 2024

Mutimodal Learning for Social Good Workshop 2024

Overview

The first Multimodal Learning for Social Good (MML4SG) workshop will be held as part of the 2024 IEEE International Conference on Multimedia and Expo (ICME). This workshop aims to gather diverse perspectives to explore multimodal learning's potential, especially in foundation models, to tackle complex societal challenges. Key questions include and are not limited to: How will foundation models transform computational social science? How can multimodal learning approaches enhance our understanding and management of public health crises? In what ways can multimodal learning help in detecting and countering misinformation online? Can multimodal learning be used to identify and mitigate the spread of polarizing content across social media platforms? Furthermore, how can multimodal learning assist in combating climate change and environmental degradation? It aims to facilitate exchange of ideas, identify future research directions, and foster collaborations among researchers and practitioners in from various areas such as AI, computational social science, public health, and environmental studies.

The topics of multimodal learning and social benefits are particularly relevant now due to the rapid advancements in AI technologies, specifically Large Language Models (LLMs) and Large Multimodal Models (LMMs), which have significantly broadened the scope of AI applications. These areas are currently at the forefront of global concerns, and AI researchers are increasingly focused on applying advanced AI techniques to provide innovative solutions that are not only technologically advanced but also socially beneficial.

Invited Speaker

Max Lu
Computer Science, Massachusetts Institute of Technology

Max Lu is a final-year PhD candidate in the EECS department at MIT, specializing in AI and ML research. Working with Professor Faisal Mahmood, he focuses on representation learning, foundation models, and multimodal generative AI in computational pathology. Max is also the Founding Chief Scientific Officer of Modella AI. He holds a B.S. in Biomedical Engineering and Applied Mathematics & Statistics from Johns Hopkins University and an M.S. in Computer Science from MIT and has been a recipient of the Tau Beta Pi and Siebel Scholar PhD Fellowships.

Schedule

Date & Time: July, 19 at 13:00 (Canada/Toronto)

Location: Salon A 3F at Niagara Falls Marriott on the Falls

5 min Intro

13:05 Keynote Session

Max Lu. Data-driven Multimodal Foundation Models for Computational Pathology.

14:00 Contributed Session (13 min talk + 2 min Q&A)

Dristi Datta, Manoranjan Paul, Manzur Murshed, Shyh Wei Teng, and Leigh M. Schmidtke. Unveiling Soil-Vegetation Interactions: Reflection Relationships and an Attention-Based Deep Learning Approach for Carbon Estimation. [PDF]

Hao Liu, Lijun He, and Jiaxi Liang. Joint Modal Circular Complementary Attention for Multimodal Aspect-Based Sentiment Analysis. [PDF]

Pantid Chantangphol, Sattaya Singkul, Thanawat Lodkaew, Nattasit Maharattanamalai, Atthakorn Petchsod, Theerat Sakdejayont, and Tawunrat Chalothorn. An Enhanced Multimodal Negative Feedback Detection Framework with Target Retrieval in Thai Spoken Audio. [PDF]

Liman Wang* and Hanyang Zhong*. LLM-SAP: Large Language Models Situational Awareness-Based Planning. [PDF]

Yundi Zhang, Xin Wang, Ziyi Zhang, Xueying Wang, Xiaohan Ma, Yingying Wu, Han-Wu-Shuang Bao, Xiyang Zhang. Using Large Language Models to Understand Leadership Perception and Expectation. [PDF]

Call for Papers

We welcome submissions of technical papers from the field of multimodal learning, social science, and beyond, to explore the integration of advanced multimedia technologies in addressing pressing societal challenges. The topics of interest include (but are not limited to):

Accessibility in Multimedia Content
Benchmark Datasets
Crisis Response and Humanitarian Aid
Cross-Cultural and Multilingual Analysis
Digital Humanities and Social Sciences
Ethical Considerations and Societal Impacts of Deploying Multimedia Technologies
Human-computer Interaction in Mulimodal Interfaces
Impact of Deepfake in Multimedia Journalism
Inclusivity and Diversity in Social Media
Innovative Multimodal Data Analysis Techniques
Integration of AI and IoT for Environmental Monitoring
Multimodal Content Generation for Social Good
Multimodal Health Informatics
Multimodal Sentiment Analysis and Public Opinion Mining
Novel Evaluation Metrics and Methods
Public Policy Deployment
Social Media and Mental Health
Social Movement and Activism Analysis
Use of LLMs/LMMs in Understanding Cultural Societal Trends
User Behavior Analysis on Social Platforms
VR/AR for Education and Training

All submissions should present original, unpublished work that is relevant to the workshop's themes. The review will be double-blind. Authors should prepare their manuscript according to the Guide for Authors of ICME available at Author Information and Submission Instructions: https://2024.ieeeicme.org/author-information-and-submission-instructions/

Submission address: https://cmt3.research.microsoft.com/ICME2024W

Track name: ICME2024-Workshop-MML4SG