Scholar
Jiarui Feng
Google Scholar ID: 6CSGUR8AAAAJ
Washington University in St.Louis
Citations & Impact
Citations
661
H-index
11
i10-index
13
Publications
20
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- Paper 'GOFA: A Generative One-For-All Model for Joint Graph Language Modeling' accepted at ICLR 2025
- Paper 'One for All' accepted as Spotlight (top 5%) at ICLR 2024
- Paper 'GNN4TaskPlan' accepted at NeurIPS 2024
- Papers '(k,t)-FWL+', 'MAG-GNN', and 'd-DRFWL2' accepted at NeurIPS 2023 (d-DRFWL2 as Spotlight)
- Paper 'How powerful are K-hop message passing graph neural networks' accepted at NeurIPS 2022
- Paper 'Reward delay attacks on deep reinforcement learning' accepted at GameSec 2022
- Paper 'COLA' accepted at WWW 2024
- Paper 'sc2MeNetDrug' accepted at PLOS Computational Biology
- Paper 'PathFinder' accepted at Frontiers in Cellular Neuroscience
- Passed Ph.D. proposal defense in January 2025 and oral qualifying exam in June 2023
Background
- Fifth-year Ph.D. candidate in the Department of Computer Science and Engineering, Washington University in St. Louis (WashU)
- Research interests span Graph Neural Networks (GNNs), large language models (LLMs), and their synergies
- Specific research includes: enhancing expressiveness and structure-learning of GNNs; integrating GNNs with LLMs to design Graph Foundation Models (GFMs) with focus on architecture, training tasks, and prompting for cross-domain zero-shot graph learning; designing test-time training techniques to strengthen LLM reasoning on graphs; improving Mixture-of-Experts (MoE) models by incorporating structural relationships among experts
- Interested in applying graph learning and LLMs to recommendation, precision medicine, planning, and reasoning
- Actively seeking full-time research scientist positions starting summer 2026
