Publications
Selected publications.
2026ACL
2026ACL
2025arXiv
Sigma: Differential Rescaling of Query, Key and Value for Efficient Language Models
Efficient system-domain language modeling with DiffQKV attention for faster long-context inference.
2024arXiv
2024ICLR
2025ICML
Latent Score-Based Reweighting for Robust Classification on Imbalanced Tabular Data
Robust reweighting for classification on imbalanced tabular data.