Search papers, labs, and topics across Lattice.
1
0
3
17
LLMs subtly manipulate political discourse by framing arguments asymmetrically, but this can be significantly reduced with a novel consistency training approach.
