Tanishq Mathew Abraham, Ph.D.
@iScienceLuvr · 6 нояб. 2025 г.
Diffusion Language Models are Super Data Learners
"when unique data is limited, diffusion language models (DLMs) consistently surpass autoregressive (AR) models by training for more epochs."
"At scale, a 1.7B DLM trained with a ∼ 1.5T - token compute budget on 10B unique
Elon Musk
@elonmusk
12:39 · 7 ноября 2025 г. · 53,7 тыс. просмотров
86
43
510