Search papers, labs, and topics across Lattice.
4
0
6
12
Sentiment in social-media conversations is significantly influenced by discourse moves, with specific interventions capable of mitigating negativity or amplifying hostility.
Causal graphs reveal how LLMs perceive concepts, transforming our understanding of model reasoning and interpretability.
LLMs aren't just mimicking text, they're exhibiting internal "motivation" that predicts their choices and performance, just like humans.
Reasoning unlocks factual knowledge in LLMs, but beware: hallucinated reasoning steps can poison the well.
