4 ms·
Next up is training multimodal models on audio and video. Humans may see less text, but they still train on more data in total.
by fiso64 4y ago
Next up is training multimodal models on audio and video. Humans may see less text, but they still train on more data in total.