Two papers accepted by NeurIPS 2025: 'When and How Unlabeled Data Provably Improve In-Context Learning' and 'BREAD: Branched Rollouts from Expert Anchors Bridge SFT & RL for Reasoning'
One paper accepted by COLM 2025: 'Gating is Weighting: Understanding Gated Linear Attention through In-context Learning'
One paper accepted by AISTATS 2025: 'Provable Benefits of Task-Specific Prompts for In-context Learning'
Published 'Fine-grained Analysis of In-context Linear Estimation: Data, Architecture, and Beyond' at NeurIPS 2024
Published 'Can Mamba In-Context Learn Task Mixtures?' at ICML 2024 Workshop on In-Context Learning
Recipient of the CPAL Rising Star Award (2025)
Selected for KAUST Rising Stars in AI Symposium 2025
Awarded the Sea Prize for Graduation Day presentation