Homepage

About Me

Hi! I'm Yaxuan Kong, a final-year DPhil (PhD) candidate in Engineering Science at the University of Oxford, supervised by Prof. Stefan Zohren and Prof. Qingsong Wen. Before that, I obtained my MSc in Health Data Analytics and Machine Learning at Imperial College London (Distinction) and BSc in Statistics at the University of St Andrews (First Class).

My research focuses on multimodal temporal intelligence: AI systems that learn from evolving evidence across time series, language and other modalities, and that reason, forecast and make reliable decisions under uncertainty in changing environments. My work can be divided as follows:

  • Time Series Intelligence (e.g., long-term forecasting, question answering, and foundation/reasoning models)
  • Foundation Models and Agentic AI in Temporal Settings (e.g., foundation models for temporal data, LLM agents for time series analysis, tool use and retrieval, multi-turn temporal reasoning, and live benchmarks)
  • Continual World Models for Evolving Environments (e.g., state tracking, future simulation, and adaptation)
  • Trustworthy AI for Dynamic Decision-Making (e.g., bias-aware evaluation, look-ahead bias, and robustness)
  • Applications (e.g., finance and investment management, healthcare, bioinformatics, and scientific discovery)

I am the first or co-first author of papers at ICML, ACL, AAAI (Oral, Top 5%), EMNLP and ICAIF. I also founded the FinHorizons Foundation, a research non-profit advancing responsible AI in finance.

I am always open to collaborations! If you are interested in working together on time series, LLM agents or trustworthy AI for finance, please feel free to reach out.

Representative Works

* Equal contribution. † Project lead.

  1. Taxonomy of biases in financial LLM evaluation: look-ahead, survivorship, narrative, objective and cost bias
    [ICML] Position: Evaluating LLMs in Finance Requires Explicit Bias Consideration
    Yaxuan Kong*†, Hoyoung Lee*, Yoontae Hwang*, Alejandro Lopez-Lira, Bradford Levy, Dhagash Mehta, Qingsong Wen, Chanyeol Choi, Yongjae Lee, and Stefan Zohren
    International Conference on Machine Learning (ICML), 2026
    [paper] [resources]
  2. Example of question answering over a healthcare time series in Time-MQA
    [ACL] Time-MQA: Time Series Multi-Task Question Answering with Context Enhancement
    Yaxuan Kong*†, Yiyuan Yang*, Yoontae Hwang, Wenjie Du, Stefan Zohren, Zhangyang Wang, Ming Jin, and Qingsong Wen
    Annual Meeting of the Association for Computational Linguistics (ACL), 2025. Main Conference
    [paper] [data]
  3. Architecture of P-sLSTM with patching and channel independence
    [AAAI] Unlocking the Power of LSTM for Long Term Time Series Forecasting
    Yaxuan Kong*†, Zepu Wang*, Yuqi Nie, Tian Zhou, Stefan Zohren, Yuxuan Liang, Peng Sun, and Qingsong Wen
    AAAI Conference on Artificial Intelligence (AAAI), 2025. Oral, Top 5%
    [paper] [code]
  4. Overview figure of the TimeSage-EV benchmark
    [EMNLP] TimeSage-EV: A Live Benchmark for Agentic Time Series Analysis in Evolving Environments
    Qingren Yao*, Yaxuan Kong*†, Yuqi Nie, Yichen Li, Stefan Zohren, Anna Vettoruzzo, Qingsong Wen, Ming Jin, and Joaquin Vanschoren
    Findings of the Association for Computational Linguistics: EMNLP 2026
    [paper]
  5. M2VN fuses textual and time series information while avoiding look-ahead bias
    [ICAIF] Fusing Narrative Semantics for Financial Volatility Forecasting
    Yaxuan Kong*†, Yoontae Hwang*, Marcus Kaiser, Chris Vryonides, Roel Oomen, and Stefan Zohren
    ACM International Conference on AI in Finance (ICAIF), 2025
    [paper] [code]
  6. Large language and foundation models trained on or repurposed for time series and spatio-temporal data
    [ACM CSUR] Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook
    Ming Jin, Yaxuan Kong, Qingsong Wen, Yuxuan Liang, Chaoli Zhang, Siqiao Xue, Xue Wang, James Zhang, Yi Wang, Haifeng Chen, Xiaoli Li, Shirui Pan, Vincent S. Tseng, Yu Zheng, Lei Chen, and Hui Xiong
    ACM Computing Surveys (CSUR), 2026
    [paper]
  7. The four task levels and capability matrix of the TimeSage-MT benchmark
    [arXiv] TimeSage-MT: A Multi-Turn Benchmark for Evaluating Agentic Time Series Reasoning
    Yaxuan Kong*†, Qingren Yao*, Yuqi Nie, Yichen Li, Yilei Shao, Stefan Zohren, Anna Vettoruzzo, Joaquin Vanschoren, Ming Jin, and Qingsong Wen
    arXiv:2606.01498
    [preprint] [code]
  8. Three gaps in time series reasoning: model design, task formulation and evaluation
    [arXiv] Achieving Time Series Reasoning Requires Rethinking Model Design, Task Formulation, and Evaluation
    Yaxuan Kong*†, Yiyuan Yang*, Shiyu Wang, Chenghao Liu, Yuxuan Liang, Ming Jin, Stefan Zohren, Dan Pei, Yan Liu, and Qingsong Wen
    arXiv:2502.01477
    [preprint]

Latest News

FinHorizons Seminar: online talks on AI in finance by researchers and practitioners, every Saturday on Zoom. See upcoming talks and registration, and recordings on YouTube. Everyone is welcome!

  • Upcoming: Co-organizing two AAAI 2027 workshops, FinFM: Foundation Models and Generative AI for Finance and Making Sense: Foundation Models for Multisensory Intelligence, and a tutorial on Trustworthy Document Intelligence in Finance at ICAIF 2026.
  • 2026.10: Kicked off the FinHorizons Seminar, our weekly online series on AI in finance. Many thanks to our first speakers, Yoontae Hwang and Yuqi Nie!
  • 2026.10: Launched the FinHorizons Foundation, a research non-profit advancing responsible AI in finance.
  • 2026.10: TimeSage-EV appears in Findings of EMNLP 2026!
  • 2026.10: Started as a Demonstrator (Teaching Assistant) in the School of Computing and Mathematical Sciences, Birkbeck, University of London.
  • 2026.07: Our position paper on bias in evaluating LLMs for finance appears at ICML 2026!
  • 2026.06: ChatMDV is published in GigaScience!
  • 2026.06: Taught time series forecasting at the Oxford Summer School in Economic Networks.
  • 2026.03: Invited talk at the Bank of America Quant Speaker Series Seminar.
  • 2025.11: M2VN appears at ICAIF 2025!
  • 2025.11: Co-organized the Workshop on Rethinking Financial Time Series at ICAIF 2025.
  • 2025.07: Time-MQA appears in the ACL 2025 main conference!
  • 2025.03: P-sLSTM is presented as an Oral (Top 5%) at AAAI 2025!

Publications

* Equal contribution. † Project lead. Also on Google Scholar.

Preprint

2026

  1. [EMNLP'26] Qingren Yao*, Yaxuan Kong*†, Yuqi Nie, Yichen Li, Stefan Zohren, Anna Vettoruzzo, Qingsong Wen, Ming Jin, and Joaquin Vanschoren. TimeSage-EV: A Live Benchmark for Agentic Time Series Analysis in Evolving Environments. Findings of the Association for Computational Linguistics: EMNLP 2026. [paper]
  2. [ICML'26] Yaxuan Kong*†, Hoyoung Lee*, Yoontae Hwang*, Alejandro Lopez-Lira, Bradford Levy, Dhagash Mehta, Qingsong Wen, Chanyeol Choi, Yongjae Lee, and Stefan Zohren. Position: Evaluating LLMs in Finance Requires Explicit Bias Consideration. International Conference on Machine Learning (ICML), 2026. [paper] [resources]
  3. [ACM CSUR'26] Ming Jin, Yaxuan Kong, Qingsong Wen, Yuxuan Liang, Chaoli Zhang, Siqiao Xue, Xue Wang, James Zhang, Yi Wang, Haifeng Chen, Xiaoli Li, Shirui Pan, Vincent S. Tseng, Yu Zheng, Lei Chen, and Hui Xiong. Large Models for Time Series and Spatio-Temporal Data: A Survey and Outlook. ACM Computing Surveys (CSUR), 2026. [paper]
  4. [ESWA'26] Yoontae Hwang, Yaxuan Kong, Stefan Zohren, and Yongjae Lee. Decision-informed Neural Networks with Large Language Model Integration for Portfolio Optimization. Expert Systems with Applications (ESWA), 2026. [paper]
  5. [GigaScience'26] Maria Kiourlappou, Peter Todd, Yaxuan Kong, Jayesh Hire, Sibgathullah Furquan Nawab Mohammed, Devika Agarwal, Martin Sergeant, Stefan Zohren, Brian Marsden, Jim Hughes, and Stephen Taylor. ChatMDV: Reducing Technical Barriers in Bioinformatics Analysis using Large Language Models. GigaScience, 2026. [paper] [code]
  6. [Book Chapter'26] Yaxuan Kong. AI for Time Series: Unlocking Patterns with Deep Learning, Vol. 1 (contributed chapter). CRC Press / Taylor & Francis, 2026. Edited by M. Wu, E. Eldele, Z. Chen, S. Pan, X. Li, and Q. Wen. [book]

2025

  1. [ICAIF'25] Yaxuan Kong*†, Yoontae Hwang*, Marcus Kaiser, Chris Vryonides, Roel Oomen, and Stefan Zohren. Fusing Narrative Semantics for Financial Volatility Forecasting. ACM International Conference on AI in Finance (ICAIF), 2025. [paper] [code]
  2. [ACL'25] Yaxuan Kong*†, Yiyuan Yang*, Yoontae Hwang, Wenjie Du, Stefan Zohren, Zhangyang Wang, Ming Jin, and Qingsong Wen. Time-MQA: Time Series Multi-Task Question Answering with Context Enhancement. Annual Meeting of the Association for Computational Linguistics (ACL), 2025. Main Conference [paper] [data]
  3. [AAAI'25] Yaxuan Kong*†, Zepu Wang*, Yuqi Nie, Tian Zhou, Stefan Zohren, Yuxuan Liang, Peng Sun, and Qingsong Wen. Unlocking the Power of LSTM for Long Term Time Series Forecasting. AAAI Conference on Artificial Intelligence (AAAI), 2025. Oral, Top 5% [paper] [code]

2024

  1. [JPM'24] Yaxuan Kong*, Yuqi Nie*, Xiaowen Dong, John M. Mulvey, H. Vincent Poor, Qingsong Wen, and Stefan Zohren. Large Language Models for Financial and Investment Management: Models, Opportunities and Challenges. The Journal of Portfolio Management (JPM), 2024. [paper]
  2. [JPM'24] Yaxuan Kong*, Yuqi Nie*, Xiaowen Dong, John M. Mulvey, H. Vincent Poor, Qingsong Wen, and Stefan Zohren. Large Language Models for Financial and Investment Management: Applications and Benchmarks. The Journal of Portfolio Management (JPM), 2024. [paper]
  3. [ICAIF'24] Dragos Gorduza, Yaxuan Kong, Stefan Zohren, and Xiaowen Dong. Extracting Alpha from Financial Analyst Networks. ACM International Conference on AI in Finance (ICAIF), 2024. [paper]

Experience

  • 2026.10 – Present: Founder, FinHorizons Foundation, a research non-profit advancing responsible AI in finance.
  • 2025.02 – 2026.09: Part-time, Equities (Global Industrials), a multi-billion-dollar London-based hedge fund.
  • 2022.07 – 2023.10: Analyst, Operations & Technology, Citigroup, London.

Talks

  • 2026.03: Invited talk, Bank of America. "Fusing Narrative Semantics for Financial Volatility Forecasting". [paper]
  • 2025.03: Oral talk, AAAI 2025. "Unlocking the Power of LSTM for Long Term Time Series Forecasting". [paper]
  • 2024.07: Poster, ELLIS Robust LLMs. "ChatMDV: Interactive Bioinformatics Visualisation with LLMs". [paper]

Teaching

  • 2026.10 – Present: Demonstrator (TA), School of Computing and Mathematical Sciences, Birkbeck, University of London. Data Analytics using R; Data Structures and Algorithms; Database Management; Programming in Java.
  • 2026.11: Tutorial Presenter, ICAIF 2026, Milan. Trustworthy Document Intelligence in Finance, a hands-on tutorial on question answering over SEC filings and earnings calls.
  • 2026.06: Tutorial Instructor, Oxford Summer School in Economic Networks. Deep learning for time series.

Research mentoring: a high-school researcher in the Byram Hills Science Research Program, New York, on multimodal RAG (2026 – Present).

Teaching interests: machine learning, deep learning, time series analysis, statistics and data analysis (R, Python), foundation models and LLMs, AI for finance, responsible and trustworthy AI, data science for healthcare.

Services

  • Organizer, Workshop on FinFM: Foundation Models and Generative AI for Finance, AAAI 2027.
  • Organizer, Workshop on Making Sense: Foundation Models for Multisensory Intelligence, AAAI 2027.
  • Organizer, Tutorial on Trustworthy Document Intelligence in Finance, ICAIF 2026.
  • Organizer, Workshop on Rethinking Financial Time Series, ICAIF 2025.
  • Conference Reviewer: NeurIPS, ICML, ICLR, AISTATS, ACL, AAAI, CIKM.
  • Journal Reviewer: Pattern Recognition, Neural Networks, Financial Innovation, Journal of the Royal Statistical Society: Series C.

Honors and Awards

  • 2025: AAAI Oral Presentation, Top 5%.
  • 2025: Deutsche Bank Industry Scholarship.
  • 2025: London-Based Hedge Fund Industry Scholarship.
  • 2023 – Present: Linacre College Brewer Street Scholarship, University of Oxford.
  • 2022: Distinction, MSc, Imperial College London.
  • 2021: First Class Honours, BSc in Statistics, University of St Andrews.
  • 2018/19, 2020/21: Dean's List, University of St Andrews.
  • 2018, 2019: Academic Prize (research assistantship), University of St Andrews.