Roberto L. Castro
- Papers
- 4
Cite
Notes
Only stored in your browser.
4papers
Authored papers
4Quartet: Native FP4 Training Can Be Optimal for Large Language Models
arXiv 2025
HALO: Hadamard-Assisted Lower-Precision Optimization for LLMs
arXiv 2025
QuEST: Stable Training of LLMs with 1-Bit Weights and Activations
arXiv 2025
MARLIN: Mixed-Precision Auto-Regressive Parallel Inference on Large Language Models
arXiv 2024
Affiliations
No known affiliations.
Frequent co-authors
9from 4 papers