Possible advisorsa guess from early papers, not confirmed

• Min‐Yen Kan

  Possible advisor · last author on 4 of their early first-author papers, 2026

  Suggested from co-authorship

Is this you? Claim this profile to confirm or dismiss it.

• Translating Natural Language to Planning Goals with Large-Language Models

  Yaqi Xie, Yu Chen, Tongyao Zhu, Jinbin Bai, Gong, Ze, Harold Soh

  arXiv · 2023

  The empirical results on GPT 3.5 variants show that LLMs are much better suited towards translation rather than planning, and these models are promising for translation to structured planning languages, but care should be taken in their use.

  200
• Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas

  Shiqi Chen, Tongyao Zhu, Ruochen Zhou, Jinghan Zhang, Siyang Gao, Juan Carlos Niebles, Mor Geva, J.‐J. He, +2 more

  arXiv · 2025

  This work studies the spatial reasoning challenge from the lens of mechanistic interpretability, diving into the model's internal states to examine the interactions between image and text tokens and proposes ADAPTVIS based on inference-time confidence scores to sharpen the attention on highly relevant regions when confident, while smoothing and broadening the attention window to consider a wider context when confidence is lower.

  119
• Bring Reason to Vision: Understanding Perception and Reasoning through Model Merging

  Chen, Shiqi, Jinghan Zhang, Tongyao Zhu, Wei Liu, Siyang Gao, Xiong, Miao, Manling Li, J.‐J. He

  arXiv · 2025

  It is found that perception capabilities are predominantly encoded in the early layers of the model, whereas reasoning is largely facilitated by the middle-to-late layers, which sheds light on the potential of model merging as a tool for multimodal integration and interpretation.

  50
• SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?

  Shiqi Chen, Jingze Gai, Ruochen Zhou, Jinghan Zhang, Tongyao Zhu, Junlong Li, Kangrui Wang, Zihan Wang, +8 more

  arXiv · 2026

  SkillCraft is introduced, a benchmark explicitly stress-test agent ability to form and reuse higher-level tool compositions, and a lightweight evaluation protocol is proposed that enables agents to auto-compose atomic tools into executable Skills, cache and reuse them inside and across tasks, thereby improving efficiency while accumulating a persistent library of reusable skills.

  31
• Multi-domain Dialogue State Tracking with Recursive Inference

  Lizi Liao, Tongyao Zhu, Le Hong Long, Tat‐Seng Chua

  Proceedings of the Web Conference 2021 · 2021

  The quantitative and qualitative experimental results on the MultiWOZ 2.1 corpus demonstrate that the proposed ReInf not only outperforms the state-of-the-art methods, but also achieves reasonable turn reference and interpretable slot co-reference.

  12
• When Does Mixing Help? Analyzing Query Embedding Interpolation in Multilingual Dense Retrieval

  Tongyao Zhu, Huang Chao Ming, Min‐Yen Kan

  Annual Meeting of the Association for Computational Linguistics (ACL) · 2026

  A ratio-controlled study on mMARCO that systematically evaluates retrieval performance by varying the mixing proportion of parallel query translations via embedding-level mixing, uncovering a distinct asymmetry driven by English dominance.

  –
• How Can Synthetic Data Improve Multilingual Language Model Pretraining? A Data Quality Perspective

  Association for Computational Linguistics 2026, Shiqi Chen, Longxu Dou, Junxian He, Qian Liu, Chang Ma, Jinghan Zhang, Tongyao Zhu

  Underline Science Inc. · 2026

  –
• When Models Edit Too Much: On the Fidelity of Minimal Code Edits

  Tongyao Zhu, Wei Hern Lim, Min‐Yen Kan

  arXiv · 2026

  –
• Why Do LLM Agents Fail in Exploring New Environments? A World-Modeling Perspective

  Chen, Shiqi, Tongyao Zhu, Zian Wang, Jinghan Zhang, Kangrui Wang, Ruochen Zhou, Siyang Gao, Teng Xiao, +3 more

  arXiv · 2025

  –
• Beyond Memorization: The Challenge of Random Memory Access in Language Models

  Tongyao Zhu, Qian Liu, Liang Pang, Zhengbao Jiang, Min‐Yen Kan, Min Lin

  Annual Meeting of the Association for Computational Linguistics (ACL) · 2024

  –