Picture for Mingyu Liang

Mingyu Liang

Lumos: Efficient Performance Modeling and Estimation for Large-scale LLM Training

Add code
Apr 12, 2025
Viaarxiv icon

CALF: Benchmarking Evaluation of LFQA Using Chinese Examinations

Add code
Oct 02, 2024
Figure 1 for CALF: Benchmarking Evaluation of LFQA Using Chinese Examinations
Figure 2 for CALF: Benchmarking Evaluation of LFQA Using Chinese Examinations
Figure 3 for CALF: Benchmarking Evaluation of LFQA Using Chinese Examinations
Figure 4 for CALF: Benchmarking Evaluation of LFQA Using Chinese Examinations
Viaarxiv icon

Bounding the Excess Risk for Linear Models Trained on Marginal-Preserving, Differentially-Private, Synthetic Data

Add code
Feb 06, 2024
Viaarxiv icon

Mystique: Accurate and Scalable Production AI Benchmarks Generation

Add code
Dec 16, 2022
Figure 1 for Mystique: Accurate and Scalable Production AI Benchmarks Generation
Figure 2 for Mystique: Accurate and Scalable Production AI Benchmarks Generation
Figure 3 for Mystique: Accurate and Scalable Production AI Benchmarks Generation
Figure 4 for Mystique: Accurate and Scalable Production AI Benchmarks Generation
Viaarxiv icon