Tasks SotA Datasets Papers Methods Submit About

Papers With Code 2

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

Explore

Notable Benchmarks All SotA Datasets Papers Methods

Community

Submit Results About

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

Models/BAT

BAT

Reported on 30 benchmarks across 9 tasks · 7 papers · 11 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing16 results

Sentiment AnalysisonSemEval-2014 Task-4
Laptop (F1)· 2020-01-30
85.57
best: 92.3 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Sentiment AnalysisonSemEval-2014 Task-4
Mean F1 (Laptop + Restaurant)· 2020-01-30
83.54
best: 92.53 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Laptop (F1)· 2020-01-30
85.57
best: 92.3 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Mean F1 (Laptop + Restaurant)· 2020-01-30
83.54
best: 92.53 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect ExtractiononSemEval-2014 Task-4
Laptop (F1)· 2020-01-30
85.57
best: 92.3 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect ExtractiononSemEval-2014 Task-4
Mean F1 (Laptop + Restaurant)· 2020-01-30
83.54
best: 92.53 (InstructABSA)
SOTA
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Text Clusteringon20 Newsgroups
Accuracy· 2020-04-26
35.66
best: 41.25 (G-BAT)
Neural Topic Modeling with Bidirectional Adversarial Training arXiv:2004.12331
Sentiment AnalysisonSemEval-2014 Task-4
Laptop (Acc)· 2020-01-30
79.35
best: 8276 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Sentiment AnalysisonSemEval-2014 Task-4
Mean Acc (Restaurant + Laptop)· 2020-01-30
82.69
best: 8611 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Sentiment AnalysisonSemEval-2014 Task-4
Restaurant (Acc)· 2020-01-30
86.03
best: 8946 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Sentiment AnalysisonSemEval-2014 Task-4
Restaurant (F1)· 2020-01-30
81.5
best: 92.76 (InstructABSA)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Laptop (Acc)· 2020-01-30
79.35
best: 8276 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Mean Acc (Restaurant + Laptop)· 2020-01-30
82.69
best: 8611 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Restaurant (Acc)· 2020-01-30
86.03
best: 8946 (ABSA-DeBERTa)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect-Based Sentiment Analysis (ABSA)onSemEval-2014 Task-4
Restaurant (F1)· 2020-01-30
81.5
best: 92.76 (InstructABSA)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316
Aspect ExtractiononSemEval-2014 Task-4
Restaurant (F1)· 2020-01-30
81.5
best: 92.76 (InstructABSA)
Adversarial Training for Aspect-Based Sentiment Analysis with BERT arXiv:2001.11316

Computer Vision10 results

Image ClassificationonImageNet-1K (with DeiT-T)
Top 1 Accuracy· 2022-11-21
72.3
best: 72.9 (dTPS)
SOTA
Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers arXiv:2211.11315
Object TrackingonKITTI
mean precision· 2021-08-10
75.2
best: 83.4 (M2-Track)
SOTA
Box-Aware Feature Enhancement for Single Object Tracking on Point Clouds arXiv:2108.04728
Object TrackingonKITTI
mean success· 2021-08-10
55
best: 62.9 (M2-Track)
SOTA
Box-Aware Feature Enhancement for Single Object Tracking on Point Clouds arXiv:2108.04728
Visual TrackingonLasHeR
Precision· 2023-12-17
70.2
best: 77.3 (FlexTrack)
Bi-directional Adapter for Multi-modal Tracking arXiv:2312.10611
Visual TrackingonLasHeR
Success· 2023-12-17
56.3
best: 62 (FlexTrack)
Bi-directional Adapter for Multi-modal Tracking arXiv:2312.10611
Visual TrackingonRGBT234
Precision· 2023-12-17
86.8
best: 94.6 (SUTrack-L224)
Bi-directional Adapter for Multi-modal Tracking arXiv:2312.10611
Visual TrackingonRGBT234
Success· 2023-12-17
64.1
best: 70.8 (SUTrack-L224)
Bi-directional Adapter for Multi-modal Tracking arXiv:2312.10611
Image ClassificationonImageNet-1K (With LV-ViT-S)
GFLOPs· 2022-11-21
4.7
best: 6.6 (Base (LV-ViT-S))
Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers arXiv:2211.11315
Image ClassificationonImageNet-1K (With LV-ViT-S)
Top 1 Accuracy· 2022-11-21
83.1
best: 83.5 (MCTF ($r=8$))
Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers arXiv:2211.11315
Image ClassificationonImageNet-1K (with DeiT-T)
GFLOPs· 2022-11-21
0.8
best: 1.2 (Base (DeiT-T))
Beyond Attentive Tokens: Incorporating Token Importance and Diversity for Efficient Vision Transformers arXiv:2211.11315

Medical2 results

Medical Image SegmentationonISIC 2018
Mean IoU· 2021-10-08
0.843
best: 0.867 (DuAT)
SOTA
Boundary-aware Transformers for Skin Lesion Segmentation arXiv:2110.03864
Medical Image SegmentationonISIC 2018
mean Dice· 2021-10-08
0.912
best: 0.9253 (Polar Res-U-Net++)
SOTA
Boundary-aware Transformers for Skin Lesion Segmentation arXiv:2110.03864

Audio2 results

Speech RecognitiononAISHELL-1
Params(M)· 2023-05-19
90
best: 1100 (FireRedASR-AED)
BAT: Boundary aware transducer for memory-efficient and low-latency ASR arXiv:2305.11571
Speech RecognitiononAISHELL-1
Word Error Rate (WER)· 2023-05-19
4.97
best: 0.55 (FireRedASR-AED)
BAT: Boundary aware transducer for memory-efficient and low-latency ASR arXiv:2305.11571