Humor in Pixels: Benchmarking Large Multimodal Models Understanding of Online Comics
Published in TBD, 2025
Evaluating humor and sequence comprehension in Large Multimodal Models with web comics.
Recommended citation:
Download Paper
Research Interests
I center my work on Multimodal Collaborative AI: intelligent systems that use different types of information to collaborate with humans and non-humans to achieve shared goals. Multimodal because of how we naturally perceive and interact with our world using different modes of information (our five senses: sight, hearing, smell, taste and touch). Collaborative because I aspire to develop (socially) intelligent systems that work together with us, the humans, and also with non-humans like other agents, systems or even animals/pets to achieve shared goals.
Current Work
Within this broad theme of Collaborative AI, I try to work on two fundamental challenges: Representation and Quantification. These two fields are quintessential in multimodal machine learning to enable downstream collaborative tasks.
Intrigued but Unavailable
I’m always intrigued by the potential of AI and the impact it can make in a variety of topics. Below is a list of projects that I wanted to explore, but currently unable to due to existing commitments. These projects are my “hear me out” ideas. If you are interested in collaborating for any of these, please do not hesitate to reach out!
Supervised Yearning: Learning the language of Love. Beyond the “5 love languages” (acts of service, quality time, gifts, touch, and affirmation) that naturally involve multiple modalities, I’m intrigued by the interplay of culture and romance. In particular,
- Can models learn “love languages” and adapt that knowledge to different cultures (East vs West).
- Can we quantify romance? What makes a love letter romantic?
- Can a model, trained on being an expert in romance (lol), counter love scams: a type of scam that is not only common in Singapore, but also painful financially and emotionally?