Published 'Think Only When You Need with Large Hybrid-Reasoning Models' at NeurIPS 2025
Published 'Preference Optimization for Reasoning with Pseudo Feedback' and 'Self-Boosting Large Language Models with Synthetic Preference Data' at ICLR 2025
Published 'MathScale: Scaling Instruction Tuning for Mathematical Reasoning' at ICML 2024
Published 'Tuna: Instruction Tuning using Feedback from Large Language Models' at EMNLP 2023
Published 'xRAG: Extreme Context Compression for Retrieval-augmented Generation with One Token' at NeurIPS 2024
Co-authored multiple ArXiv preprints including 'QueST', 'LongNet', 'RedStone', 'WildLong', and 'BitNet b1.58 2B4T Technical Report'