arXivYufei Xue, Yushi Huang, Jiawei Shao, Pingcheng Dong, Yonghao Tan, Shiyao Li, Kwang-Ting Cheng, Jun ZhangThu, Nov 5, 2026, 8:04 AM PST
score 16.9
Training method shrinks multimodal AI models without losing accuracy
Original: ProQuant: Progressive Quantization-aware Training for Edge MLLMs
Source: arxiv.org ↗
Writing ELI5 summary…