Bin Wang

alt text 

AI Research Scientist
MiroMind, Singapore
Email: bwang28c@gmail.com / bin.wang@miromind.ai

[Google Scholar]  [GitHub]  [HuggingFace
[LinkedIn]  [Twitter]  [RedNote


(About Me) I am currently a research scientist at MiroMind, Singapore, where I focus on artificial general intelligence (AGI) and its practical applications. My previous roles include AI scientist at the Institute for Infocomm Research (I2R, A*STAR) and research fellow at NUS. I earned my Ph.D. from the University of Southern California, Los Angeles, under the supervision of C.C. Jay Kuo. I have a strong track record of academic contributions, with over 40 papers published in top-tier conferences and journals including ACL, EMNLP, NAACL, ACM KDD, TNNLS, and TASLP. My research has been recognized with multiple best paper awards. I have served as Publication Chair for EMNLP 2023 and as an editorial board member for APSIPA Transactions.

(Openings@MiroMind) We have openings for Interns (paid) and Full-Time positions at MiroMind. Positions are available in Singapore, and other locations. Interested candidates are welcome to reach out via email.

Highlighted Projects

  • Open Deep Research @ MiroMind: MiroFlow and MiroThinker are leading open-source agentic frameworks with models.

  • Reasoning Models @ MiroMind: MiroMind-M1.

  • Audio Multimodal LLM @ I2R, ASTAR: MERaLiON is a flagship national AI project.

News

  • 2025.05 We published a comprehensive survey paper titled “100 Days After DeepSeek-R1: A Survey on Replication Studies and More Directions for Reasoning Language Models” arXiv.

  • 2025.03 A talk on AudioLLM evaluation will be given at Lorong AI. Check out the slides here.

  • 2025.03 I will be joining MiroMind as an AI Research Scientist working on LLMs in May. I will still be based in Singapore and remain open to collaboration opportunities.

  • 2025.01 Our AudioBench work is officially accepted to NAACL 2025 Main Conference! Leaderboard and Code.

  • 2024.12 We released MERaLiON models, the first audio-based large language model designed specifically for Singlish and other related tasks!

  • 2024-2026 Our team is working on National LLM Project Press Coverage.

  • 2024 AudioBench was released with a leaderboard and evaluation toolkit - AudioLLM Evaluation.

  • 2024 NAACL SeaEval for Multilingual Evaluation.

Full list of publications.