Search papers, labs, and topics across Lattice.
University of Notre Dame
5
0
9
Achieving a 30% boost in conversion rates, MORE redefines how dialogue systems balance reasoning and naturalness without compromising performance.
Forget expensive per-task search: agentic workflows can be synthesized in a single LLM pass by transferring learned structural priors, slashing optimization costs by 3 orders of magnitude.
RL fine-tuning can *hurt* reasoning performance when your base LLM is already too good, unless you force it to explore more diverse solutions.
LLMs still struggle to apply public policy knowledge in real-world scenarios, even when they can memorize facts and understand concepts.
Soft-gating with an "advisor" model can steer LLMs to be safer and more useful, reducing over-refusal without sacrificing detection accuracy.
