Lecture series: Why do LLMs need to worry?

Why do LLMs need to worry?

Lecture in English

The Achilles heel of Natural Language Processing research has for a long time been the noise in its datasets, in particular inaccurate labels. While there has been a paradigm shift from small expert models trained on labelled datasets to large language models (LLMs) trained by largely unlabelled data, which are capable of solving a variety of tasks, the problems of overconfidence and bias persist. In this talk I will present some methods for estimating uncertainty in task-oriented dialogue and utilising it to automatically correct the labels in the underlying datasets. I will hypothesise how this may open the door to solving related issues in LLMs.?

About Prof. Dr. Milica Ga?i?:

Milica Ga?i? is a Professor of the Dialog Systems and Machine Learning Group at the Heinrich Heine University Düsseldorf. Her research focuses on fundamental questions of human-computer dialogue modelling and lie in the intersection of Natural Language Processing and Machine Learning. Prior to her current position she was a Lecturer in Spoken Dialogue Systems at the Department of Engineering, University of Cambridge where she was leading the Dialogue Systems Group. Previously, she was a Research Associate and a Senior Research Associate in the same group and a Research Fellow at Murray Edwards College. She completed her PhD under the supervision of Professor Steve Young and the topic of her thesis was Statistical Dialogue Modelling for which she received an EPSRC PhD Plus Award. She holds an MPhil degree in Computer Speech, Text and Internet Technology from the University of Cambridge and Diploma (BSc. equivalent) in Mathematics and Computer Science from the University of Belgrade. She is a member of ACL, a member of ELLIS and a senior member of IEEE as well as a member of the International Scientific Advisory Board of DFKI.