File size: 2,786 Bytes
cf4686d 8a0e7c6 cf4686d 7e1a83a cf4686d 7e1a83a cf4686d 7e1a83a df2e0ac 1d5ceb9 7e1a83a cf4686d 7e1a83a 9bc05c0 df2e0ac 9bc05c0 |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 |
---
library_name: transformers
license: apache-2.0
datasets:
- Jialuo21/Science-T2I-Trainset
base_model:
- laion/CLIP-ViT-H-14-laion2B-s32B-b79K
---
<img src="teaser.png" align="center">
# SciScore
SciScore is finetuned on the base model [CLIP-H](https://huggingface.co/laion/CLIP-ViT-H-14-laion2B-s32B-b79K) using [Science-T2I](https://huggingface.co/datasets/Jialuo21/Science-T2I-Trainset) dataset. It takes an implicit prompt and a generated image as input and outputs a score that represents the scientific alignment between them.
## Resources
- [Website](https://jialuo-li.github.io/Science-T2I-Web/)
- [arXiv: Paper](https://arxiv.org/abs/2504.13129)
- [GitHub: Code](https://github.com/Jialuo-Li/Science-T2I)
- [Huggingface: Science-T2I-S&C Benchmark](https://huggingface.co/collections/Jialuo21/science-t2i-67d3bfe43253da2bc7cfaf06)
- [Huggingface: Science-T2I Trainset](https://huggingface.co/datasets/Jialuo21/Science-T2I-Trainset)
## Feature
<img src="exp.png" align="center">
## Qick Start
```
from transformers import AutoProcessor, AutoModel
from PIL import Image
import torch
device = "cuda"
processor_name_or_path = "Jialuo21/SciScore"
model_pretrained_name_or_path = "Jialuo21/SciScore"
processor = AutoProcessor.from_pretrained(processor_name_or_path)
model = AutoModel.from_pretrained(model_pretrained_name_or_path).eval().to(device)
def calc_probs(prompt, images):
image_inputs = processor(
images=images,
padding=True,
truncation=True,
max_length=77,
return_tensors="pt",
).to(device)
text_inputs = processor(
text=prompt,
padding=True,
truncation=True,
max_length=77,
return_tensors="pt",
).to(device)
with torch.no_grad():
image_embs = model.get_image_features(**image_inputs)
image_embs = image_embs / torch.norm(image_embs, dim=-1, keepdim=True)
text_embs = model.get_text_features(**text_inputs)
text_embs = text_embs / torch.norm(text_embs, dim=-1, keepdim=True)
scores = model.logit_scale.exp() * (text_embs @ image_embs.T)[0]
probs = torch.softmax(scores, dim=-1)
return probs.cpu().tolist()
pil_images = [Image.open("./examples/camera_1.png"), Image.open("./examples/camera_2.png")]
prompt = "A camera screen without electricity sits beside the window, realistic."
print(calc_probs(prompt, pil_images))
```
## Citation
```
@misc{li2025sciencet2iaddressingscientificillusions,
title={Science-T2I: Addressing Scientific Illusions in Image Synthesis},
author={Jialuo Li and Wenhao Chai and Xingyu Fu and Haiyang Xu and Saining Xie},
year={2025},
eprint={2504.13129},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2504.13129},
}
``` |