4 papers
Do LLMs Build Spatial World Models? Evidence from Grid-World Maze Tasks
Weijiang Li, Yilin Zhu, Rajarshi Das +1
Foundation models have shown remarkable performance across diverse tasks, yet their ability to construct internal spatial world models for reasoning and planning remains unclear. W…
Optimization over Trained Neural Networks: Going Large with Gradient-Based Algorithms
Jiatai Tong, Yilin Zhu, Thiago Serra +1
When optimizing a nonlinear objective, one can employ a neural network as a surrogate for the nonlinear function. However, the resulting optimization model can be time-consuming to…
Charting Empirical Laws for LLM Fine-Tuning in Scientific Multi-Discipline Learning
Lintao Wang, Zhuqiang Lu, Yilin Zhu +6
While large language models (LLMs) have achieved strong performance through fine-tuning within individual scientific domains, their learning dynamics in multi-disciplinary contexts…
An Extended Validity Domain for Constraint Learning
Yilin Zhu, Samuel Burer
We consider embedding a predictive machine-learning model within a prescriptive optimization problem. In this setting, called constraint learning, we study the concept of a validit…