Dongqi Fu
Scholar

Dongqi Fu

Google Scholar ID: WByXZAcAAAAJ
Research Scientist, Meta AI
Geometric Deep LearningSequence ModelingProbabilistic Graphical Models
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
596
 
H-index
13
 
i10-index
19
 
Publications
20
 
Co-authors
0
 
Contact
CVOpen ↗TwitterOpen ↗GitHubOpen ↗LinkedInOpen ↗
Publications
34 items
Overcoming Prior Barriers: Supervised Fine-Tuning under Long-Tail Distribution
2026
Cited
0
Hierarchical Credit Assignment for RLVR on Fused Gromov-Wasserstein Geometry
2026
Cited
0
ReGDiff: Guided Diffusion in Regulated Latent Space for Exploring Metamaterial Voxel Geometry
2026
Cited
0
Inference-Time Graph Engineering for Multi-Agent LLM Workflows
2026
Cited
0
Predict, Don't Iterate: Efficient Adaptive-Length Infilling for Diffusion Language Models
2026
Cited
0
EvoHarness-RL: Learning Self-Evolving Runtime Harness for Long-Horizon LLM Agents
2026
Cited
0
Structuring and Tokenizing Distributed User Interest Context for Generative Recommendation
2026
Cited
0
KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
2026
Cited
0
Resume
Academic Achievements
  • Rising Star in Data Science 2023 (By UChicago DSI and UCSD HDSI)
  • CS Ph.D. Fellowship 2023 (By UIUC CS)
  • C.W. Gear Outstanding Graduate Student 2023 (By UIUC CS)
  • Rising Star in Data Mining and Management 2022 (By Frontiers Journal)
  • NeurIPS Top 8% Reviewer 2022 with PC Registration Award (By NeurIPS Conference)
Research Experience
  • Research Scientist from Meta MRS Algorithms
Background
  • Research interests: Geometric Deep Learning, Sequence Modeling, and Probabilistic Graphical Models for their broad applications.