Research
This page highlights research conducted by Future Impact Group Fellows under the supervision of our Project Leads
Search by title, author, or keyword
Towards Shutdownable Agents: Generalizing Stochastic Choice in RL Agents and LLMs
Spurious Correlation Learning in Preference Optimization: Mechanisms, Consequences, and Mitigation via Tie Training
Authors: Christian Moya, Alex Semendinger, Guang Lin, Elliott Thornley
Fellows: Christian Moya, Alex Semendinger
Agentic Inequality
Authors: Matthew Sharp, Omer Bilgin, Iason Gabriel, Lewis Hammond
Fellows: Matthew Sharp, Omer Bilgin
A timing problem for instrumental convergence
Authors: Rhys Southan, Helena Ward, Jen Semler
Fellows: Rhys Southan, Helena Ward, Jen Semler
Towards Shutdownable Agents via Stochastic Choice
Authors: Elliott Thornley, Alexander Roman, Christos Ziakas, Leyton Ho, Louis Thomson
Fellows: Christos Ziakas, Leyton Ho, Louis Thomson