4 ms·
Vision Transformers Need Registers
- refulgentis 1y ago[2023] Previously: 2 years ago: https://news.ycombinator.com/item?id=37794996 https://news.ycombinator.com/item?id=37794996. 1 year ago: https://news.ycombinator.com/item?id=40329675 https://news.ycombinator.com/item?id=40329675
- bigdict 1y agoHas this been used widely since?
- heyitsguay 1y agoI ran a comparison of DINOv2 with and without registers on some image embedding tasks for work; DINOv2+registers saw a performance metric bump of 2-3%. Not nothing, not transformative, worth using when the only difference for inference is the model name string you're loading.
- ashvardanian 1y agoI can't speak for the whole industry, but we used it in older UForm <https://github.com/unum-cloud/uform https://github.com/unum-cloud/uform> and saw good adoption, especially among those deploying on the Edge, where every little trick counts. It's hard to pin down exact numbers since most deployments didn't go through Hugging Face, but at the time, these models were likely among the more widely deployed by device count.
- godelski 1y agoyes
- kombine 1y agoFor example, it is used here https://github.com/facebookresearch/vggt/ https://github.com/facebookresearch/vggt/
- Scene_Cast2 1y agoSo basically multiple CLS tokens. Fwiw, I tried multiple global tokens in my chess neural net and didn't see any uplift compared to my baseline of just having one.
- ninja3925 1y agoNote that it's not done for performance reason but rather to generate clear feature maps.
- Ameo 1y agoExtremely cool! It's interesting and honestly encouraging that this kind of thing can be discovered and understood using just "simple linear methods" and high-level analysis of patterns in layer activations.