Bio
Hi, I’m Weichen.
I’m Weichen, a computer science student at Georgia Tech. I study how artificial and biological systems represent the world, with interests in mechanistic interpretability, representation geometry, and computational neuroscience.
I’m particularly interested in what language and vision models learn, how we can understand their internal representations, and where those representations meet the human brain.
Research
Oct 2025–Present
Georgia Tech - Language Intelligence Lab
Undergraduate Researcher · PI: Dr. Anna Ivanova
- Probed hidden activations with linear probes to test which representational subspaces best predict neural responses.
- Explored prompt recovery from hidden activations as a diagnostic for semantic compression.
- Implemented representational similarity analysis between LLM hidden states and voxel-wise fMRI responses.
Jun 2024–Oct 2024
MIT - DiCarlo Lab
RSI Researcher · PI: Dr. James J. DiCarlo
- Implemented CKA and RSA to investigate representation geometry and architectural inductive biases.
- Contributed model analysis, representation similarity experiments, and hypothesis testing to an ICLR 2025 paper.
Feb 2023–Jan 2025
University College Dublin
Research Intern
- Studied spurious correlations in chest X-ray interpretation using reinforcement learning and self-/semi-supervised training.
- Developed MRI–PET fusion with dynamic image compression and evaluated it on ADNI.
- Built multimodal contrastive learning with tabular attention, improving Alzheimer’s prediction accuracy to above 83.8%.
- Proposed adaptive graph construction for learning from varying numbers of modalities.
Publications
Vision Models Trained to Estimate Spatial Latents Learn Ventral-Stream-Aligned Representations ↗
Y. Xie, W. Huang, E. Alter, J. Schwartz, J. Tenenbaum, J. DiCarlo
ICLR 2025 · 2025
Multimodal Representation Learning using Adaptive Graph Construction ↗
Weichen Huang
NeurIPS 2024 (High School Projects, top 4/335) · 2024
Multimodal Contrastive Learning and Tabular Attention for Automated Alzheimer's Disease Prediction ↗
Weichen Huang
ICCV 2023 · 2023
Spurious Correlation Mitigation in CXR Images via Reinforcement Learning and Self-Supervision ↗
Weichen Huang, K. Curran
IMVIP 2023 · 2023
Improved Multimodal Diagnosis of Alzheimer’s Disease through MRI and PET Image Fusion
Z. Chang, W. Huang, K. Curran
RSNA 2023 · 2023
Feature Scaling and Attention Convolutions for Extreme Precipitation Prediction ↗
Weichen Huang
ACM KDD 2022 Workshop · 2022
Education
2025–Present
Georgia Institute of Technology
B.S. in Computer Science (Class of 2029) · Atlanta, GA
GPA: 4.0/4.0
Theoretical Linear Algebra (MATH 3406); Probability & Statistics (MATH 3215); Applied Combinatorics (MATH 3012); General Psychology (PSYC 1101).
2019–2025
St Andrew's College
International Baccalaureate · Dublin, Ireland
IB: 43/45; SAT 1580
HL: Math AA (7), Physics (7), Chemistry (7).
Skills & interests
Mechanistic Interpretability · Representation Learning · Computational Neuroscience · Multimodal Learning
Machine learning
PyTorch · PyTorch Lightning · Hugging Face · Self-supervised learning · Contrastive learning
Interpretability
Linear probes · Sparse autoencoders · RSA · Representation geometry
Neuroscience
fMRI encoding models · Voxel-wise regression · Multimodal alignment
Programming & mathematics
Python · C++ · Linear algebra · Probability · Optimization
Activities
Fall 2026 –Present
Big O Theory Club · Georgia Tech
Officer · Theoretical computer science and student research
Fall 2026 –Present
AI Safety Initiative (AISI) · Georgia Tech
Operations Team · AI safety research, education, and outreach
2024
RSI-MIT 2024
Rickoid · Research Science Institute at MIT (22 June - 3 August)
Presented 'Convergent Neural Representations' research at MIT BCS DiCarlo Lab
Competitive programming
IOI, USACO & ICPC
Ireland’s IOI team in 2025, USACO Gold since 2023, and Georgia Tech’s ICPC Regional in 2025.
Self Help Africa ↗
Presented research on extreme precipitation forecasting with attention-augmented convolutions.
Awards
2024
NeurIPS'24 Paper Award
High School Projects track (top 4/335)
2025
IOI Irish National Team
International Olympiad in Informatics
2023
USACO Gold
2024
RSI 2024 Scholar
Research Science Institute at MIT
2025
ICLR 2025 Reviewer
2021-2024
BT Young Scientist Winner
4x Category Winner (2021-2024); Irish Research Council Award (2024)
Contact
Atlanta, Georgia 30332, USA
wh@gatech.edu
+1-943-235-5093
References available on request.