2025
Fast and Low-Cost Genomic Foundation Models via Outlier Removal.
CoRR, May, 2025

Computational Limits of Low-Rank Adaptation (LoRA) Fine-Tuning for Transformer Models.
Proceedings of the Thirteenth International Conference on Learning Representations, 2025

2024
Transformers are Deep Optimizers: Provable In-Context Learning for Deep Model Training.
CoRR, 2024

Computational Limits of Low-Rank Adaptation (LoRA) for Transformer-Based Models.
CoRR, 2024