Scaling and instruction tuning increase sycophancy in LLMs on opinion and fact tasks, but a synthetic data fine-tuning intervention reduces it on held-out prompts.
and Angeli, Gabor and Potts, Christopher, and Manning, Christopher D
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2023 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Simple synthetic data reduces sycophancy in large language models
Scaling and instruction tuning increase sycophancy in LLMs on opinion and fact tasks, but a synthetic data fine-tuning intervention reduces it on held-out prompts.