Yi-Chun Chen

Motivated and self-directed professional with a demonstrated record in conducting and disseminating AI and ML research findings. Skilled in designing and optimizing models to align with research objectives, yielding impactful outcomes. Collaborated closely with faculty and students to consistently excel in research-related tasks. Proficient in Python, PyTorch, Transformer, OpenCV, and ML frameworks. Passionate about Multimodal Understanding, Machine Learning, Computer Vision, and Generative AI.

• Program Committee, CREAI2024
• Reviewer, CogSci2024
• Conference Volunteer, AAMAS2011

Overall Winner, Student Cluster Competition in Supercomputing, 2010.

• Optimized hyperparameters for scientific applications on a Linux-based high-performance computing cluster, enhancing overall computational efficiency.

• Y.-C. Chen and A. Jhala, Collaborative comic generation: Integrating visual narrative theories with ai models for enhanced creativity, Proceedings of International Workshop on Artificial Intelligence and Creativity (CREAI 2024), Accepted, 2024.
• F. Freitas, T. Berreth, Y.-C. Chen, and A. Jhala, Characterizing the perception of urban spaces from visual analytics of street-level imagery, Ai & Society, 38, 4, 1361-1371, 2023.
• Y.-C. Chen and A. Jhala, A customizable generator for comic-style visual narrative, arXiv preprint arXiv:2401.02863, 2023.
• Y.-C. Chen and A. Jhala, Cpst: Comprehension-preserving style transfer for multi-modal narratives, arXiv preprint arXiv:2312.08695, 2023.
• Y.-C. Chen and A. Jhala, Panel transitions for genre analysis in visual narratives, arXiv preprint arXiv:2312.08720, 2023.
• Y.-C. Chen and A. Jhala, A computational model of comprehension in manga style visual narratives, Proceedings of the Annual Meeting of the Cognitive Science Society, 43, 2021.
• Y.-C. Chen, J. Robertson, and A. Jhala, Abstractions for narrative comprehension tasks., INT/WICED@ AIIDE, 2018.
• Y.-W. Huang, Y.-C. Chen, W.-Y. Yu, and V.-W. Soo, Stochastic negotiation with market utility for automated power restoration on a smart grid, Third International Workshop in Agent Technologies for Energy Systems, 2012

Collaborative Comic Generation, https://github.com/RimiChen/2024-Comic-Assistant

• Explored how human creativity collaborates with generative AI.
• Designed and implemented a narrative theory-driven authoring tool for generating image sequences with content planning.

GameTileNet: Game Material Dataset with Semantic, https://github.com/RimiChen/2024-GameTileNet

• Researched how semantic mapping between visual and narrative benefits narrative-rich content generation.
• Provided a dataset of game materials with semantic and affordance annotations by employing vision and language models.
• Leveraged prompt engineering to interact with an LLM, demonstrating proficiency in achieving specific research objectives.

Situation Model for Events in Image Sequences

• Developed a situation model with a knowledge graph of an image sequence to capture the narrative of image sequences.
• Applied diversity of vision and language models encoding features of comics to represent the context of image sequences.
• Designed computational models for narrative understanding.

• Dr. Arnav Jhala, North Carolina State University, Computer Science, ahjhala@ncsu.edu
• Dr. Julio Bahamon, University of North Carolina at Charlotte, College of Computing and Informatics, julio.bahamon@charlotte.edu
• Dr. Ignacio X. Domínguez, North Carolina State University, Computer Science, ignacioxd@ncsu.edu
• Dr. Alexander Card, North Carolina State University, Computer Science, acard@ncsu.edu

• North Carolina State University