Search papers, labs, and topics across Lattice.
3
0
6
5
Forget complex training schemes – pinpointing and tweaking just 20 neurons can flip an LLM from sycophantic to truthful, thanks to a new "perturbation probing" technique.
LLMs writing long stories frequently contradict themselves on basic facts and timelines, especially in the middle of the narrative, highlighting a critical weakness in long-form generation.
Current VLMs can ace image quizzes, but completely fumble when asked to stack blocks in a physically plausible way, revealing a critical gap in understanding real-world physics.
