Jiaqing Xie
Spotlight:
NMRTrans: Structure Elucidation from Experimental NMR Spectra (KDD-26)
GSF-χ: Chiral Graph Transformers
R²-Seg: Training-Free OOD Tumor Segmentation (CVPR-26 Oral)
AW-RoPE: Analytic-Walk Rotary Positional Encodings for Graphs
GSF-χ: Chiral Graph Transformers
R²-Seg: Training-Free OOD Tumor Segmentation (CVPR-26 Oral)
AW-RoPE: Analytic-Walk Rotary Positional Encodings for Graphs
PhD Student, School of Computer Science, Fudan University
Master's, ETH Zurich
Undergraduate, University of Edinburgh
Email:
Research interests:
– AI-Accelerated Advanced Graph Optimization
– Spectral Graph Theory
– Graph Algorithms for Science with AI
I am a first-year PhD student in the School of Computer Science at Fudan University. My research develops learning methods grounded in graph structure and spectral theory, with the goal of accelerating graph optimization and scientific discovery.
| Professional | ||
| About me | – | Research – Publications – Contact |
| Publications | – | Selected – Google Scholar |
| Code | – | GitHub – GSF-χ – Benchmarking-PEs |
| Education | – | Fudan (PhD) – ETH Zurich (Master's) – Edinburgh (Undergraduate) |
Graph Learning – Selected Publications
| Graph Optimization | ||
| MCES | – | Equivariant neural primal–dual assignment for maximum common edge subgraphs (arXiv-26) |
| Positional Encodings | ||
| AW-RoPE | – | Analytic-walk rotary positional encodings for graphs (arXiv-26) |
| Directed PEs | – | Gauge-invariant learnable spectral PEs for directed graphs via Hermitian block Krylov subspaces |
| Benchmark | – | Benchmarking positional encodings for GNNs and graph transformers (KDD-26) |
| AI for Science | ||
| NMRTrans | – | Structure elucidation from experimental NMR spectra via set transformers (KDD-26) |
| GSF-χ | – | Global stereochemical fields for chiral graph transformers |
| R²-Seg | – | Training-free OOD medical tumor segmentation via anatomical reasoning and statistical rejection (CVPR-26 Oral) |
| Other | – | Google Scholar – GitHub |
| Datasets | ||
| PE Benchmark | – | Positional encodings for GNNs and graph transformers (KDD-26) |
| NMRGym | – | NMR-based molecular structure elucidation (KDD-26) |
| PolyReal | – | Real-world polymer science workflows (CVPR-26) |
| OmniMatBench | – | Human-calibrated multimodal reasoning across 19 materials science subfields (arXiv-26) |
| QCBench | – | LLMs on domain-specific quantitative chemistry (JCIM-25) |
| DeepProtein | – | Deep learning library and benchmark for protein sequence learning (Bioinformatics-25) |
| Research Directions | ||
| Optimization | – | AI-accelerated advanced graph optimization |
| Spectral | – | Spectral graph theory |
| Science | – | Graph algorithms for science with AI |