Applications and Enhancements of LLM Agents Across Diverse Environments

Bang Liu

University of Montreal

The NLP Reading Group is excited to host Bang Liu, an assistant professor at the University of Montreal, who will be speaking remotely on Zoom and also be projected in room A14 at 11:45 AM on Friday November 1st about Applications and Enhancements of LLM Agents Across Diverse Environments.

Talk Description

Agent systems based on Large Language Models (LLMs) are gaining significant attention and proving essential in various domains. This presentation will delve into the unique demands on LLM agents’ perception, action capabilities, and cognitive reasoning across different environments and tasks, exploring how technical innovations address these challenges. Specifically, we will first introduce the relationship between large models and intelligent agents. Following this, we will examine multimodal embodied agents, inference-intensive agents, and agents in scientific domains, providing an overview of the challenges these agents face in different fields and strategies for enhancing their capabilities.

Speaker Bio

Bang Liu is an Assistant Professor at the University of Montreal and the Mila Institute, holding a Canada CIFAR AI Chair. His research focuses on natural language processing, multimodal and embodied learning, and AI for Science. He has received the University of Montreal’s Research Excellence Award, a Best Paper Nomination at WWW 2023, and the WAIC Yunfan Award 2024. With over 80 publications in top conferences and journals, Liu pioneered the first large model for materials science and advanced graph learning approaches in NLP research, with applications widely deployed in industry systems.

Logistics

Date: November 1st
Time: 11:45AM
Location: A14 or via Zoom (See email)