*in-person* Harish Tayyar Madabushi | Emergent Abilities in LLMs

*RSVPs are closed due to high demand. If you can no longer attend, please change your RSVP*
Title: Emergent Abilities in Large Language Models: Do they pose an existential threat?
Speaker: Harish Tayyar Madabushi (University of Bath)
Abstract: Large language models, comprising billions of parameters and pre-trained on extensive web-scale corpora, have been claimed to acquire certain capabilities without having been specifically trained on them. These capabilities, referred to as "emergent abilities," have been a driving force in discussions regarding the potentials and risks of language models. A key challenge in evaluating emergent abilities is that they are confounded by model competencies that arise through alternative prompting techniques, including in-context learning, which is the ability of models to complete a task based on a few examples. We present a novel theory that explains emergent abilities, taking into account their potential confounding factors, and rigorously substantiate this theory through over 1000 experiments. Our findings suggest that purported emergent abilities are not truly emergent, but result from a combination of in-context learning, model memory, and linguistic knowledge. Our work is a foundational step in explaining language model performance, providing a template for their efficient use and clarifying the paradox of their ability to excel in some instances while faltering in others. Thus, we demonstrate that their capabilities should not be overestimated.
Speaker Bio: Dr. Tayyar Madabushi's research focuses on understanding the fundamental mechanisms that underpin the performance and functioning of Large Language Models such as ChatGPT. His work was included in the discussion paper on the Capabilities and Risks of Frontier AI, which was used as one of the foundational research works for discussions at the UK AI Safety Summit held at Bletchley Park. His research on the constructional information encoded in language models has been influential in bringing together the fields of construction grammar and pre-trained language models. In addition, his work on language models includes collaborative industrial research aimed at rectifying biases in speech-to-text systems widely utilised across the UK. Before starting his PhD in automated question answering at the University of Birmingham, Dr. Tayyar Madabushi founded and headed a social media data analytics company based in Singapore.

Agenda:
6pm - Talk
6:40pm - Q&A session
7pm - Networking
8pm - Close

When RSVPing, please include your forename and surname for the venue's security.
Please note that the live talk will be recorded and uploaded to the Meetup YouTube channel.

Many thanks to our sponsors, Man Group and ArcticDB, for hosting this Meetup.