ICCV 2025: Paper '2.5 Years in Class: A Multimodal Textbook for Vision-Language Pretraining' accepted; ranked #2 on Hugging Face Trending with over 24k downloads
ACL 2025: Two papers accepted on SQL generation (STaR-SQL) and social reasoning LLMs
NeurIPS 2024: Paper 'TaskBench' accepted, focusing on benchmarking LLM task-planning capabilities
EMNLP 2024: Two papers accepted on multimodal data synthesis (Multimodal Self-instruct) and O1-style reasoning
ACL 2024: Four papers accepted on LLM reflection (Self-contrast), self-evolving agents (Agent-Pro), time perception (Time-ToM), and PEFT
IEEE-ACM Transactions on Audio, Speech and Language Processing 2024: One paper on mathematical reasoning accepted
ICLR 2024 LLM Agent Workshop: 'Data-Copilot' awarded Outstanding Paper
Received Distinguished Reviewer Award at CIKM 2024
Released Embodied-Reasoner: an O1-style embodied reasoning model with open-sourced dataset of 9.3K observation-reasoning-action trajectories and 64K images