Publications

2026

  1. Magenta: Closing the Loop Between Mathematical Reasoning and Lean Verification
    Joshua Ong Jun Leang, Haonan Li, Zheng Zhao, Xinyi Shang, and 5 more authors
    arXiv preprint, 2026
    PDF
  2. AoA: Theorem Proving Agent over Abstract Syntax Tree of Redesigned Language
    Qiyuan Xu, Joshua Ong Jun Leang, Renxi Wang, Wenda Li, and 3 more authors
    Accepted to International Conference on Software Engineering (ICSE), 2026
    PDF
  3. Pythagoras-Prover: Advancing Efficient Formal Proving via Augmented Lean Formalisation
    Joshua Ong Jun Leang, Zheng Zhao, Mihaela Catalina Stoian, Qiyuan Xu, and 4 more authors
    Technical Report, arXiv preprint arXiv:2606.12594, 2026
    Best Paper Award at the 4th Workshop on Mathematical Natural Language Processing (MathNLP 2026), EMNLP 2026; non-archival
    PDF
  4. SCOPE: Self-Play via Co-Evolving Policies for Open-Ended Tasks
    Wai-Chung Kwan, Aryo Pradipta Gema, Joshua Ong Jun Leang, and Pasquale Minervini
    Accepted to Advances in Neural Information Processing Systems (NeurIPS), 2026
    PDF
  5. Can I Have Your Order? Monte-Carlo Tree Search for Slot Filling Ordering in Diffusion Language Models
    Joshua Ong Jun Leang, Yu Zhao, Mihaela Cătălina Stoian, Wenda Li, and 2 more authors
    In Proceedings of the International Conference on Machine Learning (ICML), 2026
    PDF
  6. Neural Theorem Proving For Verification Conditions: A Real-World Benchmark
    Qiyuan Xu, Xiaokun Luan, Renxi Wang, Joshua Ong Jun Leang, and 4 more authors
    In Proceedings of the International Conference on Learning Representations (ICLR), 2026
    PDF
  7. PiCSAR: Probabilistic Confidence Selection And Ranking for Reasoning Chains
    Joshua Ong Jun Leang, Zheng Zhao, Aryo Pradipta Gema, Sohee Yang, and 6 more authors
    In Findings of the Association for Computational Linguistics (ACL), 2026
    PDF

2025

  1. OpenSIR: Open-Ended Self-Improving Reasoner
    Wai-Chung Kwan, Joshua Ong Jun Leang, Pavlos Vougiouklis, Jeff Z Pan, and 2 more authors
    arXiv preprint, 2025
    PDF
  2. CoMAT: Chain of mathematically annotated thought improves mathematical reasoning
    Joshua Ong Jun Leang, Aryo Pradipta Gema, and Shay B Cohen
    In Proceedings of the Empirical Methods in Natural Language Processing (EMNLP), 2025
    PDF
  3. Theorem Prover as a Judge for Synthetic Data Generation
    Joshua Ong Jun Leang, Giwon Hong, Wenda Li, and Shay B Cohen
    In Proceedings of the Association for Computational Linguistics (ACL), 2025
    PDF
  4. Are We Done with MMLU?
    Aryo Pradipta Gema, Joshua Ong Jun Leang, Giwon Hong, Alessio Devoto, and 7 more authors
    In Proceedings of the Nations of the Americas Chapter of the Association for Computational Linguistics (NAACL), 2025
    PDF