Mingquan Feng

ssToken: Self-modulated and Semantic-aware Token Selection for LLM Fine-tuning

KO: Kinetics-inspired Neural Optimizer with PDE Simulation Approaches

Rethinking LLM Evaluation: Can We Evaluate LLMs with 200× Less Data?

KITINet: Kinetics Theory Inspired Network Architectures with PDE Simulation Approaches

PhysPDE: Rethinking PDE Discovery and a Physical Hypothesis Selection Benchmark

SINGER: Stochastic Network Graph Evolving Operator for High Dimensional PDEs

HEAP: Hyper Extended A-PDHG Operator for Constrained High-dim PDEs