In the evolving world of AI, conversational agents are becoming more sophisticated, moving beyond simple rule-based responses to adapt and learn continuously.
At the 1st Workshop of our research lab, AIMultimediaLab, I presented my article, Exploring Conversational Agents and Continual Learning in Artificial Intelligence, where I shared insights from my research into this transformative shift.
Through the integration of continual learning techniques and generative AI, my work aims to create agents capable of understanding context, retaining knowledge from past interactions, and responding more naturally to complex queries. Fundamental components include a survey of advancements in neural network architectures, innovative uses of Retrieval-Augmented Generation (RAG), and multimodal interactions, enabling agents to engage with both text and visual data, such as video and images.
A central focus of this research is the potential for AI agents to approach a form of self-awareness, addressing philosophical questions around learning, knowledge, and even subjectivity. By exploring areas such as digital avatars, knowledge retrieval, and adaptive frameworks, this work aims to push the boundaries of what conversational AI can achieve.
The article also highlights potential applications for AI across various industries, from customer service and education to advanced video processing.
This journey into the intersection of AI, continual learning, and real-world applications reflects the exciting possibilities ahead for intelligent systems that are increasingly contextual, personalized, and aware.
Photos
-
- pres-1
-
- pres-2
-
- AIM group
