TasksSotADatasetsPapersMethodsSubmitAbout
Papers With Code 2

A community resource for machine learning research: papers, code, benchmarks, and state-of-the-art results.

Explore

Notable BenchmarksAll SotADatasetsPapersMethods

Community

Submit ResultsAbout

Data sourced from the PWC Archive (CC-BY-SA 4.0). Built by the community, for the community.

Papers With Code 2

The latest in machine learning research with linked code, benchmarks, and state-of-the-art results.

40,176 benchmarks2,101 tasks589,061 papers
Browse State-of-the-ArtDatasetsSubmit Results

State-of-the-Art Spotlight

Browse all SOTA
SOTAVisual Reasoning/VASR· 1:1 Accuracy: 52.9
VASR: Visual Analogies of Situation Recognition
Swin·Yonatan Bitton, Ron Yosef, Eli Strugo, Dafna Shahaf et al.·2022-12-08
PaperCode
SOTAScene Parsing/3R-Scan· Top-5 Accuracy: 0.87
SceneGraphFusion: Incremental 3D Scene Graph Prediction from RGB-D Sequences
SceneGraphFusion·Shun-Cheng Wu, Johanna Wald, Keisuke Tateno, Nassir Navab et al.·2021-03-27·CVPR 2021 1
PaperCode
SOTASpeech Recognition/swb_hub_500 WER fullSWBCH· Percentage error: 19.1
Building DNN Acoustic Models for Large Vocabulary Speech Recognition
DNN + Dropout·Andrew L. Maas, Peng Qi, Ziang Xie, Awni Y. Hannun et al.·2014-06-30
PaperCode
SOTAScene Text Detection/IC19-ReCTs· F-Measure: 93.36
Omnidirectional Scene Text Detection with Sequential-free Box Discretization
BDN·Yuliang Liu, Sheng Zhang, Lianwen Jin, Lele Xie et al.·2019-06-06
PaperCode
SOTAError Understanding/CUB-200-2011 (ResNet-101)· Average highest confidence: 0.2647
Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks
Grad-CAM++·Aditya Chattopadhyay, Anirban Sarkar, Prantik Howlader, Vineeth N. Balasubramanian et al.·2017-10-30
PaperCode
SOTAText Summarization/Arxiv HEP-TH citation graph· ROUGE-1: 45.01
A Divide-and-Conquer Approach to the Summarization of Long Documents
DANCER PEGASUS·Alexios Gidiotis, Grigorios Tsoumakas·2020-04-13
PaperCode
SOTAFew-Shot Learning/FSS-1000 (5-shot)· FB-IoU: 94.4
Cost Aggregation with 4D Convolutional Swin Transformer for Few-Shot Segmentation
VAT (ResNet-101)·Sunghwan Hong, Seokju Cho, Jisu Nam, Stephen Lin et al.·2022-07-22
PaperCode
SOTAText Generation/WebNLG· Number of parameters (M): 6.2
TrICy: Trigger-guided Data-to-text Generation with Intent aware Attention-Copy
TrICy (trK = trk* = 0.24)·Vibhav Agarwal, Sourav Ghosh, Harichandana BSS, Himanshu Arora et al.·2024-01-25·IEEE/ACM Transactions on Audio, Speech, and Language Processing 2024 1
Paper
SOTAAtari Games/Atari 2600 Skiing· Score: -3660
First return, then explore
Go-Explore·Adrien Ecoffet, Joost Huizinga, Joel Lehman, Kenneth O. Stanley et al.·2020-04-27
PaperCode
SOTAText Generation/DailyDialog· BLEU-1: 14.17
An Auto-Encoder Matching Model for Learning Utterance-Level Semantic Dependency in Dialogue Generation
AEM+Attention·Liangchen Luo, Jingjing Xu, Junyang Lin, Qi Zeng et al.·2018-08-27·EMNLP 2018 10
PaperCode
SOTAAspect Category Detection/Citysearch· F-measure (%): 86.4
Embarrassingly Simple Unsupervised Aspect Extraction
CAt·Stéphan Tulkens, Andreas van Cranenburgh·2020-04-28·ACL 2020 6
PaperCode
SOTAVideo/CRVD· LPIPS (sRGB): 0.0357
Multi-Stage Raw Video Denoising with Adversarial Loss and Gradient Mask
MaskDnGAN·Avinash Paliwal, Libing Zeng, Nima Khademi Kalantari·2021-03-04
PaperCode

Explore by Area

Browse all tasks

Methodology

155 tasks

8038 benchmarks

197,695 papers

Computer Vision

731 tasks

13818 benchmarks

103,299 papers

Natural Language Processing

390 tasks

4864 benchmarks

85,328 papers

Medical

177 tasks

3618 benchmarks

52,177 papers

Miscellaneous

167 tasks

625 benchmarks

25,081 papers

Time Series

74 tasks

1418 benchmarks

24,067 papers

Graphs

75 tasks

1203 benchmarks

21,298 papers

Robots

48 tasks

800 benchmarks

21,044 papers

Knowledge Base

43 tasks

1070 benchmarks

16,652 papers

Adversarial

23 tasks

589 benchmarks

15,385 papers

Audio

65 tasks

2576 benchmarks

13,863 papers

Speech

35 tasks

455 benchmarks

7,057 papers

Playing Games

28 tasks

308 benchmarks

2,005 papers

Reasoning

25 tasks

140 benchmarks

1,912 papers

Computer Code

43 tasks

190 benchmarks

1,618 papers

Music

22 tasks

464 benchmarks

580 papers