[ECCV 2024] Official PyTorch implementation of ESREAL
6
stars
1
commits
Python
primary language
Oct 9, 2024
updated
Official PyTorch implementation of ESREAL. For details, see the paper ESREAL: Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models.
Authors: Minchan Kim*, Minyeong Kim*, Junik Bae*, Suhwan Choi, Sungkyung Kim, Buru Chang (*: equal contribution)
Oct 2024: Code is available now.Jul 2024: ESREAL got accepted to ECCV 2024.git clone https://github.com/kmy17518/ESREAL.git
git submodule update --init --recursive
poetry install
sh scripts/reward_server.sh
sh scripts/train.sh
sh scripts/infer.sh
For more detail, refer to https://www.conventionalcommits.org/en/v1.0.0/.
We are very grateful for the great previous works including LLaVA, InstructBLIP, mPLUG-Owl2, SDXL, SDXL Turbo, HyperSDXL, and Grounding DINO.
If you find our work helpful for your research, please consider citing the following BibTeX entry.
@article{esreal,
title={Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models},
author={Minchan Kim and Minyeong Kim and Junik Bae and Suhwan Choi and Sungkyung Kim and Buru Chang},
year={2024},
url={https://arxiv.org/abs/2403.16167},
}
1 commits
Python
97.5%
Shell
1.7%
[ECCV 2024] Official PyTorch implementation of ESREAL
6
stars
1
commits
Python
primary language
Oct 9, 2024
updated
Official PyTorch implementation of ESREAL. For details, see the paper ESREAL: Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models.
Authors: Minchan Kim*, Minyeong Kim*, Junik Bae*, Suhwan Choi, Sungkyung Kim, Buru Chang (*: equal contribution)
Oct 2024: Code is available now.Jul 2024: ESREAL got accepted to ECCV 2024.git clone https://github.com/kmy17518/ESREAL.git
git submodule update --init --recursive
poetry install
sh scripts/reward_server.sh
sh scripts/train.sh
sh scripts/infer.sh
For more detail, refer to https://www.conventionalcommits.org/en/v1.0.0/.
We are very grateful for the great previous works including LLaVA, InstructBLIP, mPLUG-Owl2, SDXL, SDXL Turbo, HyperSDXL, and Grounding DINO.
If you find our work helpful for your research, please consider citing the following BibTeX entry.
@article{esreal,
title={Exploiting Semantic Reconstruction to Mitigate Hallucinations in Vision-Language Models},
author={Minchan Kim and Minyeong Kim and Junik Bae and Suhwan Choi and Sungkyung Kim and Buru Chang},
year={2024},
url={https://arxiv.org/abs/2403.16167},
}
1 commits
Python
97.5%
Shell
1.7%