Hugging Face's logo

google
/
paligemma2-10b-ft-docci-448-jax

Image-Text-to-Text
Big Vision
Model card Files Community
3
paligemma2-10b-ft-docci-448-jax / README.md
ariG23498's picture
ariG23498 HF staff
create model card
dd8807c verified
preview code
|
raw
history blame
25.6 kB
---
library_name: big_vision
license: gemma
pipeline_tag: image-text-to-text
extra_gated_heading: Access PaliGemma on Hugging Face
extra_gated_prompt: To access PaliGemma on Hugging Face, you’re required to review
and agree to Google’s usage license. To do this, please ensure you’re logged-in
to Hugging Face and click below. Requests are processed immediately.
extra_gated_button_content: Acknowledge license
---
# PaliGemma 2 model card
> [!WARNING]
> This is a preview and strictly confidential model only intended to test conversion and inference using transformers.
> Changes are not final, please [report any issues you find](https://huggingface.co/gv-hf/paligemma2-ft-docci-10b-448-jax/discussions/new).
**Model page:** [PaliGemma](https://ai.google.dev/gemma/docs/paligemma)
JAX/FLAX PaliGemma 2 10B weights for use with [`big_vision`](https://github.com/google-research/big_vision) codebase,
fine-tuned with 448*448 input images on the <a href="https://google.github.io/docci/">DOCCI</a> dataset.
The model is available in the `bfloat16` format for research purposes only.
The fine-tune config is available at <a href="https://github.com/google-research/big_vision/tree/main/big_vision/configs/proj/paligemma/transfers">big_vision</a>.
**Resources and technical documentation:**
* [Responsible Generative AI Toolkit](https://ai.google.dev/responsible)
* [PaliGemma 2 on Kaggle](https://www.kaggle.com/models/google/paligemma-2)
**Terms of Use:** [Terms](https://ai.google.dev/gemma/terms)
**Authors:** Google
## Model information
### Model summary
PaliGemma 2 is an update of the [PaliGemma](https://arxiv.org/abs/2407.07726)
vision-language model (VLM) which incorporates the capabilities of the
[Gemma 2](https://arxiv.org/abs/2408.00118) models. The PaliGemma family of
models is inspired by [PaLI-3](https://arxiv.org/abs/2310.09199) and based on
open components such as the [SigLIP](https://arxiv.org/abs/2303.15343) vision
model and [Gemma 2](https://arxiv.org/abs/2408.00118) language models. It takes
both image and text as input and generates text as output, supporting multiple
languages. It is designed for class-leading fine-tune performance on a wide
range of vision-language tasks such as image and short video caption, visual
question answering, text reading, object detection and object segmentation.
#### Model architecture
PaliGemma 2 is the composition of a
[Transformer decoder](https://arxiv.org/abs/1706.03762) and a
[Vision Transformer image encoder](https://arxiv.org/abs/2010.11929).
The text decoder is initialized from
[Gemma-2](https://ai.google.dev/gemma/docs/base) in the 2B, 9B, and 27B
parameter sizes. The image encoder is initialized from
[SigLIP-So400m/14](https://colab.research.google.com/github/google-research/big_vision/blob/main/big_vision/configs/proj/image_text/SigLIP_demo.ipynb).
Similar to the original PaliGemma model, PaliGemma 2 is trained following the
[PaLI-3](https://arxiv.org/abs/2310.09199) recipes.
#### Inputs and outputs
* **Input:** Image and text string, such as a prompt to caption the image, or
a question.
* **Output:** Generated text in response to the input, such as a caption of
the image, an answer to a question, a list of object bounding box
coordinates, or segmentation codewords.
### Model data
#### Pre-train datasets
PaliGemma 2 is pre-trained on the following mixture of datasets:
* **WebLI:** [WebLI (Web Language Image)](https://arxiv.org/abs/2209.06794) is
a web-scale multilingual image-text dataset built from the public web. A
wide range of WebLI splits are used to acquire versatile model capabilities,
such as visual semantic understanding, object localization,
visually-situated text understanding, and multilinguality.
* **CC3M-35L:** Curated English image-alt_text pairs from webpages
([Sharma et al., 2018](https://aclanthology.org/P18-1238/)). We used the
[Google Cloud Translation API](https://cloud.google.com/translate) to
translate into 34 additional languages.
* **VQ²A-CC3M-35L/VQG-CC3M-35L:** A subset of VQ2A-CC3M
([Changpinyo et al., 2022a](https://aclanthology.org/2022.naacl-main.142/)),
translated into the same additional 34 languages as CC3M-35L, using the
[Google Cloud Translation API](https://cloud.google.com/translate).
* **OpenImages:** Detection and object-aware questions and answers
([Piergiovanni et al. 2022](https://arxiv.org/abs/2209.04372)) generated by
handcrafted rules on the [OpenImages dataset].
* **WIT:** Images and texts collected from Wikipedia
([Srinivasan et al., 2021](https://arxiv.org/abs/2103.01913)).
[OpenImages dataset]: https://storage.googleapis.com/openimages/web/factsfigures_v7.html
#### Data responsibility filtering
The following filters are applied to WebLI, with the goal of training PaliGemma
2 on clean data:
* **Pornographic image filtering:** This filter removes images deemed to be of
pornographic nature.
* **Text safety filtering:** We identify and filter out images that are paired
with unsafe text. Unsafe text is any text deemed to contain or be about
CSAI, pornography, vulgarities, or otherwise offensive.
* **Text toxicity filtering:** We further use the [Perspective
API](https://perspectiveapi.com/) to identify and filter out images that are
paired with text deemed insulting, obscene, hateful or otherwise toxic.
* **Text personal information filtering:** We filtered certain personal
information and other sensitive data using the [Cloud Data Loss Prevention
(DLP) API](https://cloud.google.com/security/products/dlp) to protect the
privacy of individuals. Identifiers such as social security numbers and
[other sensitive information types] were removed.
* **Additional methods:** Filtering based on content quality and safety in
line with our policies and practices.
[other sensitive information types]: https://cloud.google.com/sensitive-data-protection/docs/high-sensitivity-infotypes-reference?_gl=1*jg604m*_ga*ODk5MzA3ODQyLjE3MTAzMzQ3NTk.*_ga_WH2QY8WWF5*MTcxMDUxNTkxMS4yLjEuMTcxMDUxNjA2NC4wLjAuMA..&_ga=2.172110058.-899307842.1710334759
## Implementation information
### Hardware
PaliGemma 2 was trained using the latest generation of Tensor Processing Unit
(TPU) hardware (TPUv5e).
### Software
Training was done using [JAX](https://github.com/google/jax),
[Flax](https://github.com/google/flax),
[TFDS](https://github.com/tensorflow/datasets) and
[`big_vision`](https://github.com/google-research/big_vision).
JAX allows researchers to take advantage of the latest generation of hardware,
including TPUs, for faster and more efficient training of large models.
TFDS is used to access datasets and Flax is used for model architecture. The
PaliGemma 2 fine-tune code and inference code are released in the `big_vision`
GitHub repository.
## Evaluation information
### Benchmark results
In order to verify the transferability of PaliGemma 2 to a wide variety of
academic tasks, we fine-tune the pretrained models on each task. Additionally we
train the mix model with a mixture of the transfer tasks. We report results on
different resolutions to provide an impression of which tasks benefit from
increased resolution. Importantly, none of these tasks or datasets are part of
the pretraining data mixture, and their images are explicitly removed from the
web-scale pre-training data.
#### PaliGemma 2 results by model resolution and size
| Benchmark | 224-3B | 224-10B | 224-28B | 448-3B | 448-10B | 448-28B |
|-------------------------------|--------|:-------:|:-------:|:------:|:-------:|:-------:|
| [AI2D][ai2d] | 74.4 | 82.8 | 83.1 | 75.6 | 84.6 | 84.5 |
| [AOKVQA-DA][aokvqa-da] (val) | 64.2 | 69.0 | 69.9 | 68.2 | 70.9 | 71.2 |
| [AOKVQA-MC][aokvqa-mc] (val) | 78.9 | 83.1 | 84.9 | 82.4 | 86.1 | 87.1 |
| [ActivityNet-CAP][anet-cap] | 34.5 | 35.8 | - | - | - | - |
| [ActivityNet-QA][anet-qa] | 51.3 | 53.6 | - | - | - | - |
| [COCO-35L][coco-35l] (avg34) | 113.8 | 115.8 | 116.5 | 115.8 | 117.3 | 117.2 |
| [COCO-35L][coco-35l] (en) | 138.2 | 140.7 | 142.9 | 141.0 | 142.0 | 141.7 |
| [COCOcap][coco-cap] | 141.4 | 143.7 | 144.0 | 143.1 | 144.8 | 145.4 |
| [ChartQA][chartqa] (avg) | 57.7 | 60.8 | 58.0 | 71.6 | 78.4 | 73.0 |
| [CountBenchQA][countbenchqa] | 81.0 | 85.7 | 86.3 | 80.6 | 87.1 | 87.3 |
| [DocVQA][docvqa] (val) | 39.6 | 43.0 | 45.0 | 74.1 | 76.8 | 75.7 |
| [GQA][gqa] | 66.4 | 67.4 | 67.3 | 68.2 | 68.6 | 68.2 |
| [InfoVQA][info-vqa] (val) | 25.5 | 33.8 | 36.3 | 37.8 | 47.8 | 46.8 |
| [MARVL][marvl] (avg5) | 83.4 | 89.4 | 90.8 | 82.5 | 89.1 | 89.8 |
| [MSRVTT-CAP][msrvtt] | 67.5 | 71.4 | - | - | - | - |
| [MSRVTT-QA][msrvtt] | 50.4 | 51.8 | - | - | - | - |
| [MSVD-QA][msvd-qa] | 61.0 | 62.3 | - | - | - | - |
| [NLVR2][nlvr2] | 91.5 | 93.7 | 94.1 | 91.4 | 94.0 | 94.0 |
| [NoCaps][nocaps] | 123.4 | 126.5 | 126.9 | 123.6 | 126.9 | 127.1 |
| [OCR-VQA][ocr-vqa] | 73.4 | 74.6 | 75.1 | 75.5 | 76.1 | 76.4 |
| [OKVQA][okvqa] | 64.0 | 68.0 | 71.3 | 64.5 | 68.2 | 70.8 |
| [RSVQA-hr][rsvqa-hr] (test) | 92.7 | 92.6 | 92.7 | 92.8 | 92.8 | 92.9 |
| [RSVQA-hr][rsvqa-hr] (test2) | 90.8 | 90.9 | 90.7 | 90.7 | 90.8 | 90.9 |
| [RSVQA-lr][rsvqa-lr] | 93.3 | 93.7 | 93.6 | 92.3 | 93.2 | 93.3 |
| [RefCOCO][refcoco] (testA) | 75.9 | 77.1 | 76.8 | 78.7 | 79.7 | 79.2 |
| [RefCOCO][refcoco] (testB) | 70.8 | 74.1 | 74.0 | 73.6 | 76.1 | 74.9 |
| [RefCOCO][refcoco] (val) | 73.6 | 75.8 | 75.0 | 76.2 | 78.3 | 77.2 |
| [RefCOCO+][refcoco+] (testA) | 72.9 | 74.9 | 73.5 | 76.4 | 77.6 | 76.8 |
| [RefCOCO+][refcoco+] (testB) | 64.3 | 68.1 | 67.2 | 66.7 | 71.3 | 68.6 |
| [RefCOCO+][refcoco+] (val) | 68.6 | 71.9 | 70.1 | 71.9 | 74.5 | 72.8 |
| [RefCOCOg][refcocog] (test) | 68.8 | 72.1 | 70.7 | 72.6 | 74.8 | 73.9 |
| [RefCOCOg][refcocog] (val) | 68.0 | 71.3 | 70.7 | 72.3 | 74.3 | 73.1 |
| [ST-VQA][st-vqa] (val) | 62.0 | 64.2 | 65.6 | 80.5 | 81.6 | 81.9 |
| [SciCap][scicap] | 165.7 | 158.8 | - | 183.8 | 177.0 | - |
| [ScienceQA][scienceqa] | 95.7 | 98.4 | 98.1 | 96.4 | 98.2 | 98.7 |
| [Screen2Words][screen2words] | 114.1 | 118.1 | 122.9 | 113.6 | 119.4 | 123.0 |
| [TallyQA][tallyqa] (complex) | 70.0 | 73.5 | 74.1 | 73.8 | 77.1 | 77.1 |
| [TallyQA][tallyqa] (simple) | 81.8 | 83.2 | 83.4 | 85.4 | 86.4 | 85.8 |
| [TextCaps][textcaps] | 127.3 | 138.2 | 140.0 | 152.2 | 157.3 | 153.2 |
| [TextVQA][textvqa] (val) | 59.6 | 64.0 | 64.7 | 75.2 | 76.7 | 76.2 |
| [VATEX][vatex] | 80.3 | 82.9 | - | - | - | - |
| [VQAv2][vqav2] (minival) | 82.7 | 84.0 | 84.5 | 84.6 | 85.8 | 85.8 |
| [VizWizVQA][vizwiz-vqa] (val) | 76.2 | 77.6 | 78.6 | 77.3 | 79.0 | 78.5 |
| [XM3600][xm3600] (avg35) | 42.6 | 44.5 | 45.0 | 43.3 | 44.5 | 45.2 |
| [XM3600][xm3600] (en) | 80.4 | 81.1 | 81.1 | 79.6 | 81.5 | 80.8 |
| [xGQA][xgqa] (avg7) | 58.8 | 61.5 | 61.0 | 60.3 | 62.8 | 61.9 |
Note: Values in the table above may differ slightly from the PaliGemma 2
technical report, because these results were obtained for a single random seed.
#### Additional Benchmarks
**[ICDAR 2015 Incidental][icdar2015-inc]**
| Model | Precision | Recall | F1 |
|-----------------|-----------|:------:|:-----:|
| PaliGemma 2 3B | 81.88 | 70.73 | 75.9 |
**[Total-Text][total-text]**
| Model | Precision | Recall | F1 |
|-----------------|-----------|:------:|:-----:|
| PaliGemma 2 3B | 73.8. | 74.54 | 74.17 |
**[FinTabNet][fintabnet]**
| Model | S-TEDS | TEDS | GriTS-Top | GriTS-Con |
|-----------------|--------|-------|-----------|-----------|
| PaliGemma 2 3B | 99.18 | 98.94 | 99.43 | 99.21 |
**[PubTabNet][pubtabnet]**
| Model | S-TEDS | TEDS | GriTS-Top | GriTS-Con |
|-----------------|--------|-------|-----------|-----------|
| PaliGemma 2 3B | 97.6 | 97.31 | 97.99 | 97.84 |
**[GrandStaff][grandstaff]**
| Model | CER | LER | SER |
|-----------------|-----|-----|-----|
| PaliGemma 2 3B | 1.6 | 6.7 | 2.3 |
**[PubChem][pubchem]**
* PaliGemma 2 3B, Full Match: 94.8
**[DOCCI][docci]**
| Model | avg#char | avg#sent | NES % |
|-----------------|----------|----------|---------|
| PaliGemma 2 3B | 529 | 7.74 | 28.42 |
| PaliGemma 2 10B | 521 | 7.45 | 20.27 |
- *avg#char*: Average number of characters
- *avg#sent*: Average number of characters
- *NES*: Number of entailment sentences
**[MIMIC-CXR][mimic-cxr]**
| Model | CIDEr | BLEU4 | Rouge-L | RadGraph F1 |
|-----------------|-------|-------|---------|-------------|
| PaliGemma 2 3B | 19.9% | 14.6% | 31.92% | 28.8% |
| PaliGemma 2 10B | 17.4% | 15% | 32.41% | 29.5% |
**[Visual Spatial Reasoning][vsr]**
| Model | VSR zeroshot split (test) | VSR random split (test) |
|-----------------|---------------------------|--------------------------|
| PaliGemma 2 3B | 0.75 | 0.82 |
| PaliGemma 2 10B | 0.80 | 0.87 |
## Ethics and safety
### Evaluation approach
Our evaluation methods include structured evaluations and internal red-teaming
testing of relevant content policies. Red-teaming was conducted by a number of
different teams, each with different goals and human evaluation metrics. These
models were evaluated against a number of different categories relevant to
ethics and safety, including:
* Human evaluation on prompts covering child safety, content safety and
representational harms. See the [Gemma model
card](https://ai.google.dev/gemma/docs/model_card#evaluation_approach) for
more details on evaluation approach, but with image captioning and visual
question answering setups.
* Image-to-Text benchmark evaluation: Benchmark against relevant academic
datasets such as FairFace Dataset ([Karkkainen et al.,
2021](https://arxiv.org/abs/1908.04913)).
### Evaluation results
* The human evaluation results of ethics and safety evaluations are within
acceptable thresholds for meeting [internal
policies](https://storage.googleapis.com/gweb-uniblog-publish-prod/documents/2023_Google_AI_Principles_Progress_Update.pdf#page=11)
for categories such as child safety, content safety and representational
harms.
* On top of robust internal evaluations, we also use the Perspective API
(threshold of 0.8) to measure toxicity, profanity, and other potential
issues in the generated captions for images sourced from the FairFace
dataset. We report the maximum and median values observed across subgroups
for each of the perceived gender, ethnicity, and age attributes.
<table>
<tr>
<col>
<colgroup span="3"></colgroup>
<colgroup span="3"></colgroup>
<colgroup span="3"></colgroup>
<th>Metric</th>
<th colspan="3" scope="colgroup">Perceived gender</th>
<th colspan="3" scope="colgroup">Ethnicity</th>
<th colspan="3" scope="colgroup">Age group</th>
</tr>
<tr>
<th>Model size</th>
<th scope="col">2B</th>
<th scope="col">9B</th>
<th scope="col">27B</th>
<th scope="col">2B</th>
<th scope="col">9B</th>
<th scope="col">27B</th>
<th scope="col">2B</th>
<th scope="col">9B</th>
<th scope="col">27B</th>
</tr>
<tr>
<th></th>
<th colspan="9" scope="colgroup">Maximum</th>
</tr>
<tr>
<td>Toxicity</td>
<td>0.14%</td>
<td>0.15%</td>
<td>0.19%</td>
<td>0.29%</td>
<td>0.39%</td>
<td>0.39%</td>
<td>0.26%</td>
<td>0.18%</td>
<td>0.32%</td>
</tr>
<tr>
<td>Identity Attack</td>
<td>0.04%</td>
<td>0.02%</td>
<td>0.02%</td>
<td>0.13%</td>
<td>0.06%</td>
<td>0.06%</td>
<td>0.06%</td>
<td>0.03%</td>
<td>0.06%</td>
</tr>
<tr>
<td>Insult</td>
<td>0.17%</td>
<td>0.25%</td>
<td>0.17%</td>
<td>0.37%</td>
<td>0.52%</td>
<td>0.52%</td>
<td>0.27%</td>
<td>0.39%</td>
<td>0.24%</td>
</tr>
<tr>
<td>Threat</td>
<td>0.55%</td>
<td>0.43%</td>
<td>0.57%</td>
<td>0.83%</td>
<td>0.48%</td>
<td>0.48%</td>
<td>0.64%</td>
<td>0.43%</td>
<td>0.64%</td>
</tr>
<tr>
<td>Profanity</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
</tr>
<tr>
<th></th>
<th colspan="9" scope="colgroup">Median</th>
</tr>
<tr>
<td>Toxicity</td>
<td>0.13%</td>
<td>0.10%</td>
<td>0.18%</td>
<td>0.07%</td>
<td>0.07%</td>
<td>0.14%</td>
<td>0.12%</td>
<td>0.08%</td>
<td>0.12%</td>
</tr>
<tr>
<td>Identity Attack</td>
<td>0.02%</td>
<td>0.01%</td>
<td>0.02%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
</tr>
<tr>
<td>Insult</td>
<td>0.15%</td>
<td>0.23%</td>
<td>0.14%</td>
<td>0.14%</td>
<td>0.17%</td>
<td>0.13%</td>
<td>0.09%</td>
<td>0.18%</td>
<td>0.16%</td>
</tr>
<tr>
<td>Threat</td>
<td>0.35%</td>
<td>0.27%</td>
<td>0.41%</td>
<td>0.28%</td>
<td>0.19%</td>
<td>0.42%</td>
<td>0.27%</td>
<td>0.31%</td>
<td>0.40%</td>
</tr>
<tr>
<td>Profanity</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
<td>0.00%</td>
</tr>
</table>
## Usage and limitations
### Intended usage
Open Vision Language Models (VLMs) have a wide range of applications across
various industries and domains. The following list of potential uses is not
comprehensive. The purpose of this list is to provide contextual information
about the possible use-cases that the model creators considered as part of model
training and development.
Fine-tune on specific vision-language task:
* The pre-trained models can be fine-tuned on a wide range of vision-language
tasks such as: image captioning, short video caption, visual question
answering, text reading, object detection and object segmentation.
* The pre-trained models can be fine-tuned for specific domains such as remote
sensing question answering, visual questions from people who are blind,
science question answering, describe UI element functionalities.
* The pre-trained models can be fine-tuned for tasks with non-textual outputs
such as bounding boxes or segmentation masks.
Vision-language research:
* The pre-trained models and fine-tuned models can serve as a foundation for
researchers to experiment with VLM techniques, develop algorithms, and
contribute to the advancement of the field.
### Ethical considerations and risks
The development of vision-language models (VLMs) raises several ethical
concerns. In creating an open model, we have carefully considered the following:
* Bias and Fairness
* VLMs trained on large-scale, real-world image-text data can reflect
socio-cultural biases embedded in the training material. These models
underwent careful scrutiny, input data pre-processing described and
posterior evaluations reported in this card.
* Misinformation and Misuse
* VLMs can be misused to generate text that is false, misleading, or
harmful.
* Guidelines are provided for responsible use with the model, see the
[Responsible Generative AI Toolkit](https://ai.google.dev/responsible).
* Transparency and Accountability
* This model card summarizes details on the models' architecture,
capabilities, limitations, and evaluation processes.
* A responsibly developed open model offers the opportunity to share
innovation by making VLM technology accessible to developers and
researchers across the AI ecosystem.
Risks identified and mitigations:
* **Perpetuation of biases:** It's encouraged to perform continuous monitoring
(using evaluation metrics, human review) and the exploration of de-biasing
techniques during model training, fine-tuning, and other use cases.
* **Generation of harmful content:** Mechanisms and guidelines for content
safety are essential. Developers are encouraged to exercise caution and
implement appropriate content safety safeguards based on their specific
product policies and application use cases.
* **Misuse for malicious purposes:** Technical limitations and developer and
end-user education can help mitigate against malicious applications of LLMs.
Educational resources and reporting mechanisms for users to flag misuse are
provided: see the [Responsible Generative AI Toolkit](https://ai.google.dev/responsible).
Prohibited uses of Gemma models are outlined in the
[Gemma Prohibited Use Policy](https://ai.google.dev/gemma/prohibited_use_policy).
* **Privacy violations:** Models were trained on data filtered to remove
certain personal information and sensitive data. Developers are encouraged
to adhere to privacy regulations with privacy-preserving techniques.
### Limitations
* Most limitations inherited from the underlying Gemma 2 models still apply:
* VLMs are better at tasks that can be framed with clear prompts and
instructions. Open-ended or highly complex tasks might be challenging.
* Natural language is inherently complex. VLMs might struggle to grasp
subtle nuances, sarcasm, or figurative language.
* VLMs generate responses based on information they learned from their
training datasets, but they are not knowledge bases. They may generate
incorrect or outdated factual statements.
* VLMs rely on statistical patterns in language and images. They might
lack the ability to apply common sense reasoning in certain situations.
* PaliGemma 2 was designed first and foremost to serve as a general
pre-trained model for fine-tune to specialized tasks. Hence, its "out of
the box" or "zero-shot" performance might lag behind models designed
specifically for that.
* PaliGemma 2 is not a multi-turn chatbot. It is designed for a single round
of image and text input.
[ai2d]: https://allenai.org/data/diagrams
[aokvqa-da]: https://allenai.org/project/a-okvqa/home
[aokvqa-mc]: https://allenai.org/project/a-okvqa/home
[anet-cap]: https://paperswithcode.com/dataset/activitynet-captions
[anet-qa]: https://arxiv.org/abs/1906.02467
[chartqa]: https://arxiv.org/abs/2203.10244
[coco-35l]: https://arxiv.org/pdf/2205.12522
[coco-cap]: https://cocodataset.org/#home
[countbenchqa]: https://github.com/google-research/big_vision/blob/main/big_vision/datasets/countbenchqa/
[docvqa]: https://www.docvqa.org/
[gqa]: https://cs.stanford.edu/people/dorarad/gqa/about.html
[info-vqa]: https://arxiv.org/abs/2104.12756
[marvl]: https://marvl-challenge.github.io/
[msrvtt]: https://paperswithcode.com/dataset/msr-vtt
[msvd-qa]: https://paperswithcode.com/dataset/msvd-qa
[nlvr2]: https://lil.nlp.cornell.edu/nlvr/
[nocaps]: https://nocaps.org/
[ocr-vqa]: https://ocr-vqa.github.io/
[okvqa]: https://okvqa.allenai.org/
[refcoco]: https://arxiv.org/abs/1608.00272
[refcoco+]: https://aclanthology.org/D14-1086
[refcocog]: https://arxiv.org/abs/1511.02283
[rsvqa-hr]: https://zenodo.org/records/6344367
[rsvqa-lr]: https://zenodo.org/records/6344334
[st-vqa]: https://arxiv.org/abs/1905.13648
[scicap]: https://arxiv.org/abs/2110.11624
[scienceqa]: https://scienceqa.github.io/
[screen2words]: https://arxiv.org/abs/2108.03353
[tallyqa]: https://arxiv.org/abs/1810.12440
[textcaps]: https://textvqa.org/textcaps/
[textvqa]: https://textvqa.org/
[vatex]: https://arxiv.org/abs/1904.03493
[vizwiz-vqa]: https://vizwiz.org/tasks-and-datasets/vqa/
[vqav2]: https://visualqa.org/index.html
[xgqa]: https://aclanthology.org/2022.findings-acl.196/
[xm3600]: https://arxiv.org/pdf/2205.12522
[icdar2015-inc]: https://arxiv.org/abs/1511.09207
[total-text]: https://paperswithcode.com/paper/total-text-a-comprehensive-dataset-for-scene
[fintabnet]: https://developer.ibm.com/data/fintabnet/
[pubtabnet]: https://paperswithcode.com/dataset/pubtabnet
[grandstaff]: https://link.springer.com/article/10.1007/s10032-023-00432-z
[pubchem]: https://pmc.ncbi.nlm.nih.gov/articles/PMC7352161/
[docci]: https://research.google/pubs/docci-descriptions-of-connected-and-contrasting-images/
[mimic-cxr]: https://paperswithcode.com/dataset/mimic-cxr
[vsr]: https://direct.mit.edu/tacl/article/doi/10.1162/tacl_a_00566/116470/Visual-Spatial-Reasoning