Zhichun Guo
Senior Research Scientist at TikTok
[Email] [Google Scholar] [GitHub] [LinkedIn] [Selected Publications]
I am a Senior Research Scientist at TikTok, working on LLM for recommendation (LLM4Rec) — including generative recommendation and LLM-driven interest discovery for retrieval. Before TikTok, I worked on AI for science, where I had the fortune to work with Nobel laureate David Baker as a postdoc at the University of Washington's Institute for Protein Design. Earlier, I received my Ph.D. in Computer Science and Engineering at the University of Notre Dame, advised by Prof. Nitesh V. Chawla and supported by the Snap Research Fellowship, after completing my undergraduate studies in Computer Science at Fudan University.
Full list on Google Scholar.
LeSTD: LLM Compression via Learning-based Sparse Tensor Decomposition
International Conference on Learning Representations, 2026
You Only Spectralize Once: Taking a Spectral Detour to Accelerate Graph Neural Network
Conference on Neural Information Processing Systems, 2025
Node Duplication Improves Cold-start Link Prediction
Transactions on Machine Learning Research, 2025
Pure Message Passing Can Estimate Common Neighbor for Link Prediction
Conference on Neural Information Processing Systems, 2024
Can LLMs Solve Molecule Puzzles? A Multimodal Benchmark for Molecular Structure Elucidation
NeurIPS Datasets and Benchmarks Track, 2024
What Indeed Can GPT Models Do in Chemistry? A Comprehensive Benchmark on Eight Tasks
NeurIPS Datasets and Benchmarks Track, 2023
Linkless Link Prediction via Relational Distillation
International Conference on Machine Learning, 2023
Boosting Graph Neural Networks via Adaptive Knowledge Distillation
AAAI Conference on Artificial Intelligence, 2023
SD²: Slicing and Dicing Scholarly Data for Interactive Evaluation of Academic Performance
IEEE Transactions on Visualization and Computer Graphics, 2022
Few-Shot Graph Learning for Molecular Property Prediction
The Web Conference, 2021
GraSeq: Graph and Sequence Fusion Learning for Molecular Property Prediction
ACM International Conference on Information and Knowledge Management, 2020