Paper Accepted to EMNLP 2026 Main Conference
First-authored by our Ph.D. student Hao An, the paper proposes FiSCoRe, jointly fine-tuning models with correctness and semantic-entropy uncertainty.
Read moreFirst-authored by our Ph.D. student Hao An, the paper proposes FiSCoRe, jointly fine-tuning models with correctness and semantic-entropy uncertainty.
Read moreCongratulations to Yongyuan Li on the acceptance of his paper to COLM 2026.
Read moreCongratulations to all authors!
Read moreOur work on evaluating text generation with spectral distances of surprisal was accepted to EMNLP 2025 Findings.
Read moreOur work on distinguishing human language from model language was accepted to EMNLP 2024.
Read more