From 6793e2ae1a0725beb1d2668a3efad3d21d18efb3 Mon Sep 17 00:00:00 2001 From: Yuxin-CV <2yuxinfang@gmail.com> Date: Tue, 31 Jan 2023 14:32:45 +0800 Subject: [PATCH] [News] EVA-CLIP enables BLIP-2 --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index d6b5e3b..122bd04 100644 --- a/README.md +++ b/README.md @@ -38,7 +38,7 @@ We launch **EVA**, a vision-centric foundation model to **E**xplore the limits o -- `Jan 31, 2023`: By leveraging [EVA-CLIP](clip/README.md), BLIP-2 ([paper](https://arxiv.org/abs/2301.12597), [code](https://github.com/salesforce/LAVIS/tree/main/projects/blip2)) achieves SoTA performance on various VL tasks! +- `Jan 31, 2023`: Strong visual representations also enable powerful VL foundation models. By leveraging [EVA-CLIP](clip/README.md), BLIP-2 ([paper](https://arxiv.org/abs/2301.12597), [code](https://github.com/salesforce/LAVIS/tree/main/projects/blip2)) achieves SoTA performance on various VL tasks! - `Dec 12, 2022`: [EVA](https://github.com/rwightman/pytorch-image-models#dec-6-2022) and [EVA-L](https://github.com/rwightman/pytorch-image-models#dec-8-2022) model weights are added to the awesome [`timm`](https://github.com/rwightman/pytorch-image-models) library, thanks @[rwightman](https://github.com/rwightman)! - `Dec 07, 2022`: launch [**EVA-L**](https://github.com/baaivision/EVA/tree/master/eva#eva-l-learning-better-mim-representations-from-eva-clip), the **best** ViT-L (304M) to date that can reach up to **89.2** top-1 acc on IN-1K ([weights & logs](https://github.com/baaivision/EVA/tree/master/eva#eva-l-learning-better-mim-representations-from-eva-clip)) by leveraging vision features from [EVA-CLIP](clip/README.md). - `Nov 25, 2022`: release EVA-CLIP zero-shot [evaluation results](https://github.com/baaivision/EVA/tree/master/clip#eva-clip-zero-shot-evaluation-results) on 35 benchmarks.