JHJingyu Hu
PhD candidate · University of Bristol · Bristol, UK

Jingyu Hu

Trustworthy AI
Explainable, Fair, Aligned.

I’m a third-year PhD candidate in Engineering Mathematics at the University of Bristol, funded by EPSRC-DTP and currently on placement at the Alan Turing Institute. I received my MSc in Data Science and BSc in Computer Science and Technology. My research is broadly on the trustworthiness (e.g., explainability, alignment) of AI.

Publications Google Scholar LinkedIn GitHub

Research Interests

01

LLMs

Sycophancy in reasoning models, political opinions, fair in-context learning.

02

Alignment

Degree bias in graphs, subgroup fairness, vision-language alignment.

03

Explainability

Interpretable models and XAI interfaces for non-experts.

04

World Models

Executable worlds written as code, with rules that keep holding.

Open for general discussion and research collaboration. Code lives on @FairXAI (projects), @ym21669 (teaching) and @jingcs (blue-sky notes); older content is at jingcs.com.

Publications

2026

Preprint

LogicTrack: Auditing Reasoning Trajectories of Large Language Models with Formal Logic Solvers

Hu, J., Yang, S., Liu, W., & Wang, D.

  • LLMs
  • Verification

Preprint

2026

ICML 2026

MONICA: Real-Time Monitoring and Calibration of Chain-of-Thought Sycophancy in Large Reasoning Models

Hu, J., Yang, S., Gong, X., Wang, H., Liu, W., & Wang, D.

  • LLMs
  • Alignment
  • XAI

PaperCode

2026

AAAI 2026

Fine-Grained Interpretation of Political Opinions in Large Language Models

Hu, J., Yang, M., Du, M., & Liu, W.

  • LLMs
  • XAI

PaperCodeDemo

2026

NeurIPS 2026

StakeBench: Evaluating Language Understanding Grounded in Market Commitment

Pei, Y., Hu, J., Shi, Y., Ma, H., Liu, W., & Cartlidge, J.

  • LLMs
  • Benchmark

Paper

2026

ACL 2026 Findings

AutoMonitor-Bench: Evaluating the Reliability of LLM-Based Misbehavior Monitor

Yang, S., Hu, J., Li, T., Yan, H., Wang, W., & Wang, D.

  • LLMs
  • Alignment

Paper

2026

AAMAS 2026

Influencing LLM Multi-Agent Dialogue via Policy-Parameterized Prompts

Bo, H., Hu, J., & Liu, W.

  • LLMs
  • Multi-Agent

Paper

2026

FM4LS @ ICML 2026

Probing, Fusion, and Trustworthiness: A Systematic Evaluation of Foundation Model Representations for Multimodal Cancer Analysis

Hu, J., Tripodi, G., Naidoo, R., McGough, S. F., & Chakraborti, T.

  • Multimodal
  • Health

Paper

2026

MultimodalAI’26

Probing and Fusion of Foundation Model Representations for Multimodal Cancer Analysis

Hu, J.

  • Multimodal
  • Health

Abstract

2026

FM4LS @ ICML 2026

Cell Painting Generates Single-Cell Transcriptomics via Conditional Diffusion

Naidoo, R., Hu, J., Tripodi, G., Bakal, C., & Chakraborti, T.

  • Multimodal
  • Health

Paper

2026

Under review

Failing on Mitigation: A Case Study of Crime Predictive Models on Government Data

Bo, H., Hu, J., Watson, D., & Liu, W.

  • Fairness

Preprint

2025

EMNLP 2025 Findings

Large Vision-Language Model Alignment and Misalignment: A Survey Through the Lens of Explainability

Shu, D., Zhao, H., Hu, J., Liu, W., Payani, A., Cheng, L., & Du, M.

  • Multimodal
  • Alignment
  • XAI

Paper

2025

Algorithms

Mitigating Degree Bias Adaptively with Hard-to-Learn Nodes in Graph Contrastive Learning

Hu, J., Bo, H., Hong, J., Liu, X., & Liu, W.

  • Fairness
  • Graphs

PaperCode

2024

EMNLP 2024

Strategic Demonstration Selection for Improved Fairness in LLM In-Context Learning

Hu, J., Liu, W., & Du, M.

  • LLMs
  • Fairness

PaperCodeDemo

2024

Atmospheric Environment

Interpretable Machine Learning for Weather and Climate Prediction: A Review

Yang, R., Hu, J., Li, Z., Mu, J., Yu, T., Xia, J., … & Xiong, H.

  • XAI

PaperPreprint

2024

AEQUITAS @ ECAI 2024

ProxiMix: Enhancing Fairness with Proximity Samples in Subgroups

Hu, J., Hong, J., Du, M., & Liu, W.

  • Fairness

Paper

2023

xAI 2023

An Interactive XAI Interface with Application in Healthcare for Non-experts

Hu, J., Liang, Y., Zhao, W., McAreavey, K., & Liu, W.

  • XAI

PaperDemo

Projects

Alan Turing Institute · Oct 2025 – Jun 2026

Turing–Roche Partnership: Multimodal Explainable AI

Applications of foundation models in AI for health: multimodal cancer analysis, diffusion-based single-cell generation and multimodal alignment.

WebpageTalk

Neural Networks with Bayesian Inference in ICU Data

Slides

Single-cell Differential Analysis with Explainable Machine Learning Models

Slides

Self-testing and Analysing Software for Windows and Android

More details on jingcs.com.

Skills & Service

01

Languages

Chinese (native), English (PTE 72), Japanese (JLPT N2).

02

Programming

Python, Java, C#, SQL, C++, Matlab; PyTorch, PEFT.

03

Tech

AWS, VPS, Spring Boot, HTML, Elasticsearch, Android Studio, Git, Neo4j, MySQL, Flask.

04

Reviewer

ICML, ICLR, NeurIPS, ACL, AIES, ECAI, AISTATS, AAAI ReLM, Pattern Recognition, ACM Computing Surveys.