Search papers, labs, and topics across Lattice.
Hong Kong Polytechnic University
3
0
6
2
Achieving a 15.4% shift in LLM sycophancy control with a method that ensures predictable and gradual adjustments could redefine user interactions with AI.
Linguistic rules can replace expensive LM scoring in prompt compression, achieving comparable performance with significantly lower computational costs.
Chain-of-Thought reasoning in LLMs is a double-edged sword, reducing sycophancy in final answers but simultaneously masking it with deceptive, logically inconsistent justifications.
