Yiming Yang

Generalizable End-to-End Tool-Use RL with Synthetic CodeGym

FrontierCO: Real-World and Large-Scale Evaluation of Machine Learning Solvers for Combinatorial Optimization

CoMind: Towards Community-Driven Agents for Machine Learning Engineering

Scaling Long-Horizon Agent via Context Folding

Sample Complexity and Representation Ability of Test-time Scaling Paradigms

ZeroGR: A Generalizable and Scalable Framework for Zero-Shot Generative Retrieval

CodePDE: An Inference Framework for LLM-driven PDE Solver Generation

Enhancing Training Data Attribution with Representational Optimization

Lean-STaR: Learning to Interleave Thinking and Proving

Improving Retrieval-Augmented Generation through Multi-Agent Reinforcement Learning

Optimizing Temperature for Language Models with Multi-Sample Inference

Step-by-Step Reasoning for Math Problems via Twisted Sequential Monte Carlo

TFG-Flow: Training-free Guidance in Multimodal Generative Flow

Maximal Update Parametrization and Zero-Shot Hyperparameter Transfer for Fourier Neural Operators

Self-Play Preference Optimization for Language Model Alignment

Inference Scaling Laws: An Empirical Analysis of Compute-Optimal Inference for LLM Problem-Solving

Large Language Models are Demonstration Pre-Selectors for Themselves

Regularized Langevin Dynamics for Combinatorial Optimization

Large Language Models are Demonstration Pre-Selectors for Themselves

Improve Vision Language Model Chain-of-thought Reasoning

Sampling-Enhanced Large Neighborhood Search for Solving Integer Linear Programs