(Registration Closed) HKML S6E5 - Meetup with Maven - Recent advances in LLMs

We will be holding our next meetup with Maven on Oct 4th.

Given the instructions by Maven, the space is limited to a maximum of 20 people.

Please register for getting the final information regarding the location.
https://forms.gle/Kex3dYe3KXP62Vyz9

Speaker 1: Gautier Marti
Founder of HKML and a quantitative researcher

Gautier Marti has over 10 years of experience as a quantitative researcher. With a PhD in computer science, he will share his insights and findings from his research, which was recently presented at the prestigious International Conference on Machine Learning (ICML). Gautier's expertise and passion for pushing the boundaries of language AI make him a valuable resource for those interested in understanding the cutting-edge developments in this rapidly evolving field.

Speaker 2: Boris Kriuk

Co-Founder and CEO at Sparcus Technologies, Deep Learning Researcher, Scientific Member at Computer Vision Foundation and the Association for Computing Machinery.

Semantic Textual Similarity: Extracting Greater Value from Text Data.
Semantic Textual Similarity (STS) is a crucial task in natural language processing that quantifies the semantic relatedness between text fragments. This work presents an analysis of fundamental similarity functions and introduces a novel multiobjective approach for optimal threshold selection in STS tasks. The authors explore the strengths and limitations of commonly used similarity measures and propose a multiobjective optimization framework to determine the most suitable threshold to extract greater value from the text data.