publications

publications by categories in reversed chronological order.

2026

  1. ACL
    Bears, All Bears, and Some Bears: Language Constraints on Language Models’ Inductive Inferences
    Sriram Padmanabhan, Siyuan Song, and Kanishka Misra
    In Findings of the Association for Computational Linguistics: ACL 2026, Jul 2026
  2. ICML
    A Very Big Video Reasoning Suite
    Maijunxian Wang , Ruisi Wang, Juyi Lin, Ran Ji, Thaddäus Wiedemer, Qingying Gao, Dezhi Luo, Yaoyao Qian, Lianyu Huang, Zelong Hong, Jiahui Ge, Qianli Ma, Hang He, Yifan Zhou, Lingzi Guo, Lantao Mei, Jiachen Li, Hanwen Xing, Tianqi Zhao, Fengyuan Yu, Weihang Xiao, Yizheng Jiao, Jianheng Hou, Danyang Zhang, Pengcheng Xu, Boyang Zhong, Zehong Zhao, Gaoyun Fang, John Kitaoka, Yile Xu, Hua Xu, Kenton Blacutt, Tin Nguyen, Siyuan Song, Haoran Sun, Shaoyue Wen, Linyang He , Runming Wang , Yanzhi Wang, Mengyue Yang, Ziqiao Ma, Raphaël Millière, Freda Shi, Nuno Vasconcelos, Daniel Khashabi, Alan Yuille, Yilun Du , Ziming Liu, Bo Li, Dahua Lin , Ziwei Liu, Vikash Kumar, Yijiang Li, Lei Yang, Zhongang Cai, and Hokin Deng
    In Proceedings of the 43rd International Conference on Machine Learning, Jul 2026
  3. EACL
    babybabelmap.jpg
    BabyBabelLM: A Multilingual Benchmark of Developmentally Plausible Training Data
    Jaap Jumelet, Abdellah Fourtassi, Akari Haga, Bastian Bunzeck, Bhargav Shandilya, Diana Galvan-Sosa, Faiz Ghifari Haznitrama, Francesca Padovani, Francois Meyer, Hai Hu, Julen Etxaniz, Laurent Prevot, Linyang He, María Grandury, Mila Marcheva, Negar Foroutan, Nikitas Theodoropoulos, Pouya Sadeghi, Siyuan Song, Suchir Salhan, Susana Zhou, Yurii Paniv, Ziyin Zhang, Arianna Bisazza, Alex Warstadt, and Leshem Choshen
    In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), Mar 2026
  4. TACL
    grammaticality-probability.jpeg
    What Can String Probability Tell Us About Grammaticality?
    Transactions of the Association for Computational Linguistics, Mar 2026
  5. TACL
    zhoblimp.png
    A Systematic Assessment of Language Models with Linguistic Minimal Pairs in Chinese
    Yikang Liu, Yeting Shen, Hongao Zhu, Lilong Xu, Zhiheng Qian, Siyuan Song, Kejia Zhang, Jialong Tang, Pei Zhang, Baosong Yang, Rui Wang, and Hai Hu
    Transactions of the Association for Computational Linguistics, Mar 2026
  6. NLPCC
    The First ChineseBabyLM Challenge: Training Data-Efficient and Cognitively Plausible Language Models for Chinese
    Siyuan Song, Zhiheng Qian, Yunhao Zhang, Linyang He, Xiaozhe Ji, Yingxin Lin, Hongao Zhu, Chongtian Shao, Chuhan Lang, Luan Li, Rui Wang , Renfen Hu , Shaonan Wang, and Hai Hu
    arXiv preprint, Mar 2026
    Summary paper for the ChineseBabyLM Challenge at NLPCC 2026.
  7. Preprint
    Reasoning or Memorization: Can LLMs Understand and Generate Chinese Xiehouyu Riddles?
    Hai* Hu, Siyuan* Song, Chongtian* Shao, Kejia Zhang , Tianjian Zhu, and Xiaojing Zhao
    arXiv preprint, Mar 2026

2025

  1. Workshop
    introspection2.png
    Privileged Self-Access Matters for Introspection in AI
    Siyuan SongHarvey LedermanJennifer Hu, and Kyle Mahowald
    arXiv preprint, Mar 2025
    Poster presented at the Philosophy of Machine Learning Workshop (PhilML @ ICML 2026).
  2. Preprint
    Global PIQA: Evaluating Physical Commonsense Reasoning Across 100+ Languages and Cultures
    Tyler A. Chang, Catherine Arnett, and  others
    arXiv preprint, Mar 2025
  3. COLM
    introspectionfig1.jpeg
    Language Models Fail to Introspect About Their Knowledge of Language
    Siyuan Song , Jennifer* Hu , and Kyle* Mahowald
    Conference on Language Modeling, Mar 2025

2024

  1. wlwz.webp
    Do Large Language Models Understand Conversational Implicature: A Case Study with a Chinese Sitcom
    Shisen YueSiyuan Song, Xinyuan Cheng, and Hai Hu
    In Chinese Computational Linguistics: 23rd China National Conference, CCL 2024, Mar 2024

Presentations & Posters

2026

  1. Poster
    Error Sensitivity in the Next-Word Predictions of Humans and Language Models
    Siyuan Song, and Thomas Hikaru Clark
    2026
    Poster presentation at HSP 2026