Workshop Leadership: Organizing the AAAI 2025 Workshop on “Generative AI and Large Language Models for Health” in Philadelphia.
- Social AI Research: Developed cutting-edge AI models to address societal challenges like hate speech, sexism, homophobia, depression, suicidal tendencies, and misinformation using Twitter, Reddit, and YouTube data.
- Hate Speech Counteraction: Submission of innovative frameworks, such as LLM-aligned counter-speech generation and multilingual counter speech. Also working in African speech.
Deeper Insights
The projects under Artificial Intelligence and Generative Models for Social Good focus on leveraging AI for addressing critical societal issues. The AAAI 2025 Workshop on “Generative AI and Large Language Models for Health” aims to drive innovation and collaboration in the field. Research efforts include analyzing data from social platforms like Twitter, Reddit, and YouTube to combat challenges such as hate speech, sexism, and misinformation. Groundbreaking frameworks like multilingual counterspeech and LLM-aligned counter-speech generation address these issues and have been presented at prestigious forums.
These projects exemplify the use of AI to promote inclusivity and understanding. By developing generative AI solutions tailored for social impact, these initiatives aim to create a safer and more equitable digital space. Additionally, the integration of advanced machine learning and natural language processing techniques underscores the potential of AI to contribute to societal betterment. These efforts reflect a broader commitment to using technology as a tool for positive change.
Counter Speech Generation
Multilingual Counterspeech and LLM-Aligned Counter-Speech Generation are innovative approaches addressing the pervasive problem of online hate speech. Multilingual counterspeech focuses on generating contextually appropriate and culturally sensitive responses in multiple languages, targeting hate speech in a way that fosters understanding and reduces hostility. This involves leveraging advanced machine learning techniques to train models capable of analyzing diverse linguistic nuances and generating tailored responses that resonate with varied audiences, thereby promoting inclusivity in global digital communication.
AI (LLMs, Gen AI) in Health
AI and Large Language Models (LLMs) are revolutionizing healthcare by advancing diagnostic and therapeutic applications. These models analyze extensive datasets, aiding in the early detection of conditions like Alzheimer’s, melanoma, and mental health issues. Tools such as HSBuddy integrate geolocation, NLP, and machine learning to provide personalized support for chronic conditions, improving patients’ quality of life. AI models also enhance public health monitoring by processing unstructured data from social media and electronic health records. These innovations are transforming healthcare delivery, fostering equitable access, and addressing complex health challenges.
