Search papers, labs, and topics across Lattice.
2
0
6
8
LLMs' fact-checking abilities in African languages are surprisingly weak, but can be boosted by up to 43% with few-shot prompting and 26% with fine-tuning.
SLMs are shockingly vulnerable: combining adversarial audio and text unlocks 1.5x to 10x higher jailbreak rates than attacking either modality alone.
