Search papers, labs, and topics across Lattice.
University of Cambridge
3
0
6
Transforming uninformative reward signals into actionable insights, Reasoning Arena boosts reasoning performance while slashing training costs.
ASR for Puno Quechua gets a major boost with the first large-scale speech corpus and benchmark, paving the way for digital inclusion of Quechua speakers.
LLMs can learn to mimic the nuanced reasoning patterns of individual annotators by contrasting their explanations with those of others, opening the door to more personalized and scalable annotation strategies.
