Hosted By
61 Went
Private Event
MATS Winter Research Talks
Hosted by MATS London
Registration
Registration Closed
This event is not currently taking registrations. You may contact the host or subscribe to receive updates.
About Event
Join us for an evening of research presentations, Q&A and conversation with MATS fellows past and present.
Exploration Hacking: LLMs Can Learn to Resist RL Training
Misalignment Faking at Jailbreaking Time
Measuring LLM Reasoning Inconsistency via Causal Models
Model Organisms of Training Hacking
Talks start at 7:30pm, please arrive by then!
MATS Researchers will present talks across a range of AI Safety / Alignment topics.
Hosted By
61 Went