Search papers, labs, and topics across Lattice.
Affiliation:
2
0
5
LLMs can learn to generate syntactically valid text without sacrificing semantic quality, even with underspecified prompts, by leveraging self-supervised rewards derived from their own likelihoods.
Fine-grained lecture transcripts, when combined with knowledge graph reasoning, can significantly boost the quality of academic course recommendations compared to relying solely on metadata.
