{"url":"/dataset/indl","name":"InDL","full_name":"In-Diagram Logic","description_markdown":"**Dataset Introduction**\r\n\r\nIn this work, we introduce the In-Diagram Logic (InDL) dataset, an innovative resource crafted to rigorously evaluate the logic interpretation abilities of deep learning models. This dataset leverages the complex domain of visual illusions, providing a unique challenge for these models.\r\n\r\nThe InDL dataset is characterized by its intricate assembly of optical illusions, wherein each instance poses a specific logic interpretation challenge. These illusions are constructed based on six classic geometric optical illusions, known for their intriguing interplay between perception and logic.\r\n\r\n**Motivations and Content**\r\n\r\nThe motivation behind the creation of the InDL dataset arises from a recognized gap in current deep learning research. While models have exhibited remarkable proficiency in various domains such as image recognition and natural language processing, their performance in tasks requiring logical reasoning remains less understood and often opaque due to their inherent 'black box' characteristics. By using the medium of visual illusions, the InDL dataset aims to probe these models in a unique and challenging way, helping to illuminate their logic interpretation capabilities.\r\n\r\nThe InDL dataset is a comprehensive collection of instances where each visual illusion varies in illusion strength. The strength signifies the degree of distortion introduced to challenge the models' logic interpretation. Hence, the dataset not only offers a complexity gradient for model evaluation but also allows the analysis of model performance against varying degrees of challenge intensity.\r\n\r\n**Potential Use Cases**\r\n\r\nThe potential use cases of the InDL dataset are extensive. Beyond the primary goal of evaluating deep learning models' logic interpretation abilities, it also presents a robust tool for researchers to investigate how models react to visual perception challenges. This opens avenues to understand how these models can be improved and how their decision-making processes can be better interpreted.\r\n\r\nAdditionally, the InDL dataset could provide a rich testing ground for model developers. Its diverse and challenging instances could allow them to rigorously benchmark their models and detect potential weaknesses that might be overlooked in more conventional datasets.\r\n\r\nFurthermore, the InDL dataset could serve as a valuable resource for teaching and learning purposes. It provides a visually engaging and intellectually stimulating way to explore the capabilities and limitations of deep learning models, particularly in the realm of logic interpretation.","description_withheld":null,"homepage":"https://github.com/rabbit-magic-wh/InDL","introduced_date":"2023-05-28","introduced_date_note":null,"introduced_by":{"paper":"/paper/indl-a-new-datasets-and-benchmark-for-in","title":"InDL: A New Dataset and Benchmark for In-Diagram Logic Interpretation based on Visual Illusion","first_author":"Haobo Yang","url":null},"license":{"name":"CC BY-NC-SA","url":"https://creativecommons.org/licenses/by-nc-sa/4.0/"},"modalities":[{"name":"Images","url":"/datasets/modality/images"}],"tasks":[{"name":"Classification","url":"/task/classification-1","datasets_with_task":"/datasets/task/classification-1"}],"languages":[{"name":"English","url":"/datasets/language/english"},{"name":"Chinese","url":"/datasets/language/chinese"}],"variants":["InDL"],"data_loaders":[{"repo":"https://github.com/rabbit-magic-wh/NeuralPS-2023-CV-Illusion","url":"https://github.com/rabbit-magic-wh/NeuralPS-2023-CV-Illusion","frameworks":["pytorch"]}],"num_papers_in_archive":10,"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28"},"benchmarks":[{"leaderboard":"/sota/classification-on-indl","task":"Classification","dataset_variant":"InDL","rows":9,"metrics":["Average Recall"],"first_row_in_archive_order":{"model":"ConvNext","paper":"/paper/a-convnet-for-the-2020s","metrics":{"Average Recall":"93.47%"},"code_links":[{"title":"keras-team/keras","url":"https://github.com/keras-team/keras/blob/master/keras/applications/convnext.py"},{"title":"rwightman/pytorch-image-models","url":"https://github.com/rwightman/pytorch-image-models"},{"title":"pytorch/vision","url":"https://github.com/pytorch/vision"},{"title":"lucidrains/denoising-diffusion-pytorch","url":"https://github.com/lucidrains/denoising-diffusion-pytorch"},{"title":"facebookresearch/ConvNeXt","url":"https://github.com/facebookresearch/ConvNeXt"},{"title":"PaddlePaddle/PaddleClas","url":"https://github.com/PaddlePaddle/PaddleClas"},{"title":"open-mmlab/mmclassification","url":"https://github.com/open-mmlab/mmclassification"},{"title":"towhee-io/towhee","url":"https://github.com/towhee-io/towhee"},{"title":"mzeromiko/vmamba","url":"https://github.com/mzeromiko/vmamba"},{"title":"BR-IDL/PaddleViT","url":"https://github.com/BR-IDL/PaddleViT/tree/develop/image_classification/ConvNeXt"},{"title":"sithu31296/semantic-segmentation","url":"https://github.com/sithu31296/semantic-segmentation"},{"title":"murufeng/awesome_lightweight_networks","url":"https://github.com/murufeng/awesome_lightweight_networks"},{"title":"Westlake-AI/openmixup","url":"https://github.com/Westlake-AI/openmixup"},{"title":"leondgarse/keras_cv_attention_models","url":"https://github.com/leondgarse/keras_cv_attention_models/tree/main/keras_cv_attention_models/convnext"},{"title":"facebookresearch/ppuda","url":"https://github.com/facebookresearch/ppuda"},{"title":"mindspore-ai/models","url":"https://github.com/mindspore-ai/models/tree/master/research/cv/convnext"},{"title":"frgfm/Holocron","url":"https://github.com/frgfm/Holocron"},{"title":"mit-han-lab/litepose","url":"https://github.com/mit-han-lab/litepose"},{"title":"martinsbruveris/tensorflow-image-models","url":"https://github.com/martinsbruveris/tensorflow-image-models"},{"title":"PaddlePaddle/PASSL","url":"https://github.com/PaddlePaddle/PASSL"},{"title":"mindspore-ecosystem/mindcv","url":"https://github.com/mindspore-ecosystem/mindcv/blob/main/mindcv/models/convnext.py"},{"title":"duyhominhnguyen/LVM-Med","url":"https://github.com/duyhominhnguyen/LVM-Med"},{"title":"DarshanDeshpande/jax-models","url":"https://github.com/DarshanDeshpande/jax-models"},{"title":"sayakpaul/ConvNeXt-TF","url":"https://github.com/sayakpaul/ConvNeXt-TF"},{"title":"SarthakYadav/audax","url":"https://github.com/SarthakYadav/audax"},{"title":"waterdisappear/nudt4mstar","url":"https://github.com/waterdisappear/nudt4mstar"},{"title":"james77777778/keras-image-models","url":"https://github.com/james77777778/keras-image-models"},{"title":"bamps53/convnext-tf","url":"https://github.com/bamps53/convnext-tf"},{"title":"protonx-tf-04-projects/ConvNext-2020s","url":"https://github.com/protonx-tf-04-projects/ConvNext-2020s"},{"title":"hmichaeli/alias_free_convnets","url":"https://github.com/hmichaeli/alias_free_convnets"},{"title":"birder/birder","url":"https://gitlab.com/birder/birder"},{"title":"zibbini/convnext-v2_tensorflow","url":"https://github.com/zibbini/convnext-v2_tensorflow"},{"title":"dongkyuk/ConvNext-tensorflow","url":"https://github.com/dongkyuk/ConvNext-tensorflow"},{"title":"tuanio/nextformer","url":"https://github.com/tuanio/nextformer"},{"title":"Owais-Ansari/Unet3plus","url":"https://github.com/Owais-Ansari/Unet3plus"},{"title":"IMvision12/keras-vision-models","url":"https://github.com/IMvision12/keras-vision-models"},{"title":"k-h-ismail/convnext-dcls","url":"https://github.com/k-h-ismail/convnext-dcls"},{"title":"hanfried/hanfried-bookmarks","url":"https://github.com/hanfried/hanfried-bookmarks"},{"title":"MindCode-4/code-3","url":"https://github.com/MindCode-4/code-3/tree/main/convnext"},{"title":"leanderme/ConvNeXt-Tensorflow","url":"https://github.com/leanderme/ConvNeXt-Tensorflow"},{"title":"avocardio/resnet_vs_convnext","url":"https://github.com/avocardio/resnet_vs_convnext"},{"title":"2024-MindSpore-1/Code2","url":"https://github.com/2024-MindSpore-1/Code2/tree/main/model-1/convnext"},{"title":"jmnolte/hccnet","url":"https://github.com/jmnolte/hccnet"},{"title":"Raghvender1205/ConvNeXt","url":"https://github.com/Raghvender1205/ConvNeXt"},{"title":"AlassaneSakande/A-ConvNet-of-2020s","url":"https://github.com/AlassaneSakande/A-ConvNet-of-2020s"},{"title":"yaya-yns/tart","url":"https://github.com/yaya-yns/tart"},{"title":"yangyucheng000/convnext","url":"https://github.com/yangyucheng000/convnext/blob/main/convnext.py"},{"title":"kingcong/convnext","url":"https://github.com/kingcong/convnext"},{"title":"flytocc/ConvNeXt-paddle","url":"https://github.com/flytocc/ConvNeXt-paddle"},{"title":"kingcong/convnext-","url":"https://github.com/kingcong/convnext-"},{"title":"lyqcom/convnext","url":"https://github.com/lyqcom/convnext"},{"title":"0jason000/convnext","url":"https://github.com/0jason000/convnext"},{"title":"Asthestarsfalll/ConvNeXt-MegEngine","url":"https://github.com/Asthestarsfalll/ConvNeXt-MegEngine"},{"title":"2023-MindSpore-4/Code8","url":"https://github.com/2023-MindSpore-4/Code8/tree/main/convnext"}]},"note":"rows are the archive's own order at snapshot; nothing here re-ranks them"}],"papers_with_a_benchmark_row":[{"paper":"/paper/a-convnet-for-the-2020s","title":"A ConvNet for the 2020s","date":"2022-01-10","rows_on_this_dataset":1,"code_links":54,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":80,"samples_ran":54,"samples_unverified":26,"pointer_only_for_licence":11,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/resnet-strikes-back-an-improved-training","title":"ResNet strikes back: An improved training procedure in timm","date":"2021-10-01","rows_on_this_dataset":1,"code_links":14,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":3,"samples_ran":0,"samples_unverified":3,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/searching-for-mobilenetv3","title":"Searching for MobileNetV3","date":"2019-05-06","rows_on_this_dataset":1,"code_links":67,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":105,"samples_ran":58,"samples_unverified":47,"pointer_only_for_licence":46,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/yolov3-an-incremental-improvement","title":"YOLOv3: An Incremental Improvement","date":"2018-04-08","rows_on_this_dataset":1,"code_links":311,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":124,"samples_ran":18,"samples_unverified":106,"pointer_only_for_licence":19,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/learning-transferable-architectures-for","title":"Learning Transferable Architectures for Scalable Image Recognition","date":"2017-07-21","rows_on_this_dataset":1,"code_links":17,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":0,"samples_unverified":1,"pointer_only_for_licence":1,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/xception-deep-learning-with-depthwise","title":"Xception: Deep Learning with Depthwise Separable Convolutions","date":"2016-10-07","rows_on_this_dataset":1,"code_links":41,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":15,"samples_ran":1,"samples_unverified":14,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/densely-connected-convolutional-networks","title":"Densely Connected Convolutional Networks","date":"2016-08-25","rows_on_this_dataset":1,"code_links":146,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":71,"samples_ran":18,"samples_unverified":53,"pointer_only_for_licence":7,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/inception-v4-inception-resnet-and-the-impact","title":"Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning","date":"2016-02-23","rows_on_this_dataset":1,"code_links":87,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":1,"samples_ran":1,"samples_unverified":0,"pointer_only_for_licence":0,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}},{"paper":"/paper/very-deep-convolutional-networks-for-large","title":"Very Deep Convolutional Networks for Large-Scale Image Recognition","date":"2014-09-04","rows_on_this_dataset":1,"code_links":305,"syntology":{"read_at":"2026-09-24T18:15:14+00:00","samples_harvested":122,"samples_ran":12,"samples_unverified":110,"pointer_only_for_licence":4,"claim":"Per-sample execution on synthesized fixtures; not a correctness claim."}}],"syntology_totals":{"read_at":"2026-09-24T18:15:14+00:00","papers_with_samples":9,"samples_harvested":522,"samples_ran":162,"samples_unverified":360,"pointer_only_for_licence":88,"papers_with_no_sample_that_ran":2,"note":"the per-paper counts above, summed; not a rate"},"papers_note":"The archive never published its papers-using-dataset list; these are papers with a leaderboard row on this dataset's benchmarks."}