ELOQUENCE Webinar | 28 April, 14:00 (CEST)
As conversational AI continues to evolve, one of the key challenges remains clear: how can machines better understand human intention and respond in a way that feels natural, accurate, and trustworthy?
In our upcoming ELOQUENCE webinar, “Building Trust in Machine Judgement”, we will explore how moving from text-based inputs to more natural forms of communication, such as speech, can help systems better capture user intent. We will also look at how AI-generated responses can be improved, focusing on techniques that highlight the most relevant information and make outputs clearer and more intelligible.
The session will further touch on approaches to increase efficiency and precision, including enabling large language models to reuse relevant excerpts directly from the query context. When combined, these methods can help reduce latency while improving both the accuracy and usability of AI systems.
This is not a one-step solution, but a gradual process. Within the ELOQUENCE project, many of these techniques have been developed and tested individually. As they become increasingly integrated and fine-tuned, they bring us closer to more natural, reliable, and human-centered AI interactions.
The webinar will be led by Oldrich Plchot and Martin Fajčík from Brno University of Technology, who will share insights from their ongoing research.
Date: 28 April 2026
Time: 14:00 CEST
Register here
Following the presentation, participants are invited to join an interactive WebCafé session, where they will have the opportunity to ask questions and engage directly with the speakers.
