4 ms·
Off the shelf embedding models definitely underpromise and overdeliver. In ten years I'd be very surprised if companies weren't fine-tuning embedding models for
by nostrebored 2y ago
Off the shelf embedding models definitely underpromise and overdeliver. In ten years I'd be very surprised if companies weren't fine-tuning embedding models for search based on their data in any competitive domains.
- kkielhofner 2y agoMy startup (Atomic Canyon) developed embedding models for the nuclear energy space[0]. Let's just say that if you think off-the-shelf embedding models are going to work well with this kind of highly specialized content you're going to have a rough time. [0] - https://huggingface.co/atomic-canyon/fermi-1024 https://huggingface.co/atomic-canyon/fermi-1024