Scholar
Linwei Tao
Google Scholar ID: GeNHmf4AAAAJ
University of Sydney
Confidence Calibration
Large Language Model
Trustworthy Machine Learning
Follow
Homepage
↗
Google Scholar
↗
Citations & Impact
All-time
Citations
147
H-index
5
i10-index
4
Publications
17
Co-authors
11
list available
Contact
Email
linwei.tao@sydney.edu.au
CV
Open ↗
GitHub
Open ↗
Publications
9 items
Confidence Calibration under Ambiguous Ground Truth
2026
Cited
0
Can Large Language Models Express Uncertainty Like Human?
2025
Cited
0
Calibrating Graph Neural Networks with Wavelet-Aware Temperature Scaling
2025
Cited
0
Sample Margin-Aware Recalibration of Temperature Scaling
2025
Cited
0
Revisiting Uncertainty Estimation and Calibration of Large Language Models
2025
Cited
0
Beyond One-Hot Labels: Semantic Mixing for Model Calibration
2025
Cited
0
Uncertainty Weighted Gradients for Model Calibration
2025
Cited
0
Mitigating Object Hallucinations in Large Vision-Language Models via Attention Calibration
2025
Cited
0
Load more
Resume (English only)
Academic Achievements
Papers accepted to top-tier conferences: ICML 2025, CVPR 2025, ICLR 2025, AAAI 2025
ICLR 2025 paper received Spotlight (top 5.1% of 11,500 submissions)
Elected to the Student Committee of AAAI 2026 (June 2025)
Awarded Google Cloud Research Credits ($2340 AUD) in May 2024
Received International Tuition Fee Scholarship in June 2023
Published multiple papers on model calibration, uncertainty quantification, and trustworthiness in LLMs, many with code, datasets, and project pages
Co-authors
11 total
Chang Xu
University of Sydney
Minjing Dong
Assistant Professor of Computer Science, City University of Hong Kong
Jinxu Lin
The University of Sydney
Daochang Liu
Lecturer, University of Western Australia
Philip Torr
Professor, University of Oxford
Tao Huang
Shanghai Jiao Tong University
Haohui Lu
Charles Darwin University
Yunke Wang
University of Sydney
×
Welcome back
Sign in to Agora
Welcome back! Please sign in to continue.
Email address
Password
Forgot password?
Continue
Do not have an account?
Sign up