Papers with Datasets (@paperswithdata) | Robocrunch.Net

paperswithdata

Papers with Datasets @paperswithdata

Keep up with the latest machine learning datasets from @paperswithcode. Follow for daily updates.

Tweets by Papers with Datasets

https://paperswithcode.com/dataset/rtmv

RTMV - a new large-scale synthetic dataset for novel view synthesis consisting of ∼300k images rendered from ~2000 complex scenes. https://t.co ...

Shared by Papers with Datasets at 5/19/2022 0-3

https://paperswithcode.com/dataset/webvidvqa3m

🐱WebVidVQA3M: A new dataset with 3M video-question-answer triplets. It's automatically generated using question generation neural models and al ...

Shared by Papers with Datasets at 5/13/2022 0-6

https://paperswithcode.com/dataset/clues-classifier-learning-using-natural

🗣️💬🤖CLUES is a new benchmark for classifier learning using natural language explanations. It consists of a range of classification tasks ov ...

Shared by Papers with Datasets at 5/11/2022 1-4

https://paperswithcode.com/dataset/qlevr

🔶 QLEVR: a new diagnostic visual question-answering dataset that focuses on more complex quantifiers and their combinations, e.g., asking "Are ...

Shared by Papers with Datasets at 5/9/2022 1-2

https://paperswithcode.com/dataset/clevr-x

CLEVR-X: A new dataset with natural language explanations in the context of visual question answering. It consists of 3.6 million natural langu ...

Shared by Papers with Datasets at 4/6/2022 0-6

https://paperswithcode.com/dataset/cicero

CICERO: a new dataset for contextualized commonsense inference in dialogues. It contains 53,000 inferences for five commonsense dimensions -- ...

Shared by Papers with Datasets at 4/4/2022 0-8

https://paperswithcode.com/dataset/fairytaleqa

🧚‍♀️ FairytaleQA is a new ML dataset focusing on narrative comprehension of kindergarten to eighth-grade students. It consists of 10.5K quest ...

Shared by Papers with Datasets at 3/31/2022 1-5

https://paperswithcode.com/dataset/instaorder

🏇 InstaOrder: a dataset consisting of 2.9M annotations of geometric orderings for class-labeled instances in 101K natural scenes. It can be us ...

Shared by Papers with Datasets at 3/29/2022 0-4

https://paperswithcode.com/dataset/medmcqa

MedMCQA: a new large-scale, multiple-choice question answering dataset with 194K real-world medical entrance exam questions. It includes: - 2.4 ...

Shared by Papers with Datasets at 3/29/2022 0-7

https://paperswithcode.com/dataset/bigdetection

💫 BigDetection is a new large-scale benchmark to build more general and powerful object detection systems. It leverages existing datasets and ...

Shared by Papers with Datasets at 3/25/2022 1-4

https://paperswithcode.com/dataset/pacs-commonsense

👂PACS: a new audiovisual benchmark for physical commonsense reasoning. PACS contains a total of 13.4K question-answer pairs, involving 1.4K u ...

Shared by Papers with Datasets at 3/22/2022 0-3

https://paperswithcode.com/dataset/openlane

🛣️ OpenLane: one of the largest, real-world 3D lane datasets to date. It owns 200K frames and over 880K carefully annotated lanes. OpenLane ai ...

Shared by Papers with Datasets at 3/22/2022 1-5

https://paperswithcode.com/dataset/spatial-commonsense-graph-dataset

Spatial Commonsense Graph Dataset: aimed at solving object localisation in partially observed scenes (e.g., where is the bag?) using commonsense ...

Shared by Papers with Datasets at 3/21/2022 1-3

https://paperswithcode.com/dataset/imagenet-patch

ImageNet-Patch: A new dataset for benchmarking machine learning robustness against adversarial patches. https://t.co/Srcx0RQgme

Shared by Papers with Datasets at 3/11/2022 1-0

https://paperswithcode.com/dataset/silg

🕹️SILG: a multi-environment benchmark which unifies a collection of diverse grounded language learning environments under a common interface. ...

Shared by Papers with Datasets at 3/10/2022 1-4

https://paperswithcode.com/dataset/kubric

Kubric: an open-source Python framework to generate photo-realistic synthetic multi-object videos with rich annotations such as instance segment ...

Shared by Papers with Datasets at 3/8/2022 1-8

https://paperswithcode.com/dataset/kmir

💫KMIR: a new benchmark with ~184K questions for evaluating knowledge memorization, identification and reasoning abilities of language models. I ...

Shared by Papers with Datasets at 3/1/2022 1-8

https://paperswithcode.com/dataset/topiocqa

TopiOCQA: a new open-domain conversational dataset with topic switches on Wikipedia. TopiOCQA contains ~4K conversations with information-seekin ...

Shared by Papers with Datasets at 2/24/2022 0-3

https://paperswithcode.com/dataset/muld

MuLD (Multitask Long Document Benchmark): a new set of 6 NLP tasks where the inputs consist of at least 10,000 words. The benchmark covers a wid ...

Shared by Papers with Datasets at 2/23/2022 0-3

https://paperswithcode.com/dataset/v2x-sim

🚗V2X-Sim: a synthetic collaborative perception dataset in autonomous driving to facilitate collaborative perception between multiple vehicles a ...

Shared by Papers with Datasets at 2/21/2022 1-4

https://paperswithcode.com/dataset/proteinkg25

🟠 ProteinKG25: a large-scale KG dataset with aligned descriptions and protein sequences respectively to GO terms and proteins entities. It cont ...

Shared by Papers with Datasets at 2/14/2022 4-12

https://paperswithcode.com/dataset/clear

💫 CLEAR: A new benchmark for continual image classification. Contains natural temporal evolution of visual concepts in the real world. It's des ...

Shared by Papers with Datasets at 1/19/2022 10-39