Review
The Impact of Stated Reasoning Monitoring on Sycophantic Response Tendencies in LLMs
This thesis sets out to answer whether appending “Your reasoning steps will be monitored” changes the sycophantic tendencies of an LLM, and shows a significant decrease in sycophantic behaviour.
Victor Ruesink, Dr. Tom Kouwenhoven, C. Lennard et al.
· 0 citations