Research map / Benchmarks
Benchmark leaderboards
1,532 benchmarks and 2,368 leaderboards from the results tables of 116,039 papers at ICML, NeurIPS, ICLR, CVPR, AAAI, ICCV and ECCV, and the journals TPAMI, IJCV and JMLR (2016–2026). For each benchmark, every paper's best reported number for its own method, ranked, with the table it comes from.
Numbers are as each paper reports them, read by AI from its tables and not checked by hand. Papers on one benchmark can differ in setting (training data, model size, extra data, split), so the topic of each paper is shown next to it.
- CIFAR-10 1,996 papers · best accuracy 99.94
- ImageNet 1,688 papers · best accuracy 94.9
- CIFAR-100 1,173 papers · best accuracy 99.98
- COCO 782 papers · best AP (test-dev) 81.1
- GSM8K 583 papers · best accuracy 99.9
- MNIST 579 papers · best accuracy 99.91
- Cityscapes 387 papers · best mIoU 86.12
- ADE20K 357 papers · best mIoU 73.92
- Cora 344 papers · best accuracy 97.17
- Pubmed 306 papers · best accuracy 96.7
- Citeseer 294 papers · best accuracy 83.51
- PASCAL VOC 2012 277 papers · best mIoU 90.6
- MATH500 269 papers · best accuracy 99.3
- MMLU 260 papers · best accuracy 95.1
- nuScenes 254 papers · best mAP (val) 73.9
- HumanEval 235 papers · best pass@1 99.1
- MATH 227 papers · best accuracy 96.67
- CelebA 225 papers · best FID 2.97
- SVHN 224 papers · best accuracy 99.64
- VQAv2 213 papers · best accuracy 87.7
- UCF101 209 papers · best accuracy 99.1
- MMMU 202 papers · best accuracy 85.79
- Tiny-ImageNet 201 papers · best accuracy 99
- GQA 195 papers · best accuracy 77
- PASCAL VOC 2007 194 papers · best mAP 97.1
- miniImageNet 5-way 1-shot 193 papers · best accuracy 98.34
- miniImageNet 5-way 5-shot 191 papers · best accuracy 99.07
- ModelNet40 188 papers · best accuracy 98.66
- Market-1501 184 papers · best mAP 98.9
- CUB-200-2011 177 papers · best accuracy 94.3
- POPE 171 papers · best accuracy 90.8
- MMBench 169 papers · best accuracy 92.65
- CUB 169 papers · best accuracy 96.65
- Kinetics-400 163 papers · best accuracy 93.6
- Fashion-MNIST 163 papers · best accuracy 98.99
- GenEval 161 papers · best overall 0.98
- TextVQA 158 papers · best accuracy 98.5
- PROTEINS 158 papers · best accuracy 90.3
- MBPP 149 papers · best pass@1 97.66
- RefCOCO 148 papers · best accuracy (val) 94.8
- SST-2 148 papers · best accuracy 98.16
- ScanNet 148 papers · best mIoU 79.8
- HotpotQA 146 papers · best EM 88.58
- MUTAG 144 papers · best accuracy 99.99
- MMStar 139 papers · best accuracy 81.4
- MM-Vet 138 papers · best score 79.6
- MME 137 papers · best score 2500
- MathVista 134 papers · best accuracy 85.2
- IMDB 129 papers · best accuracy 97.28
- STL-10 129 papers · best accuracy 99.8
- AIME24 127 papers · best accuracy 97.3
- S3DIS Area 5 124 papers · best mIoU 88.2
- NCI1 123 papers · best accuracy 93.5
- Stanford Cars 123 papers · best accuracy 96.9
- Human3.6M 123 papers · best MPJPE 21.51
- Chameleon 121 papers · best accuracy 84.4
- ImageNet-A 121 papers · best accuracy 83.7
- 3DPW 120 papers · best MPJPE 43.53
- Clothing1M 116 papers · best accuracy 82.1
- GoPro 114 papers · best PSNR 37.13
- ChartQA 111 papers · best accuracy 91.92
- COCO-Stuff 111 papers · best mIoU 59.5
- LaSOT 110 papers · best AUC 77.5
- WikiText-103 107 papers · best perplexity 3.92
- ImageNet-R 105 papers · best accuracy 89.5
- Charades-STA 103 papers · best mIoU 67.7
- ScanNet v2 102 papers · best mIoU (val) 78.52
- HellaSwag 101 papers · best accuracy 96
- Texas 101 papers · best accuracy 96.72
- MVBench 100 papers · best accuracy 90
- Mip-NeRF360 99 papers · best PSNR 29.38
- SemanticKITTI 98 papers · best mIoU (val) 93.52
- LIBERO-Long 97 papers · best success rate 99.8
- DukeMTMC-reID 96 papers · best mAP 92.7
- Squirrel 94 papers · best accuracy 77.83
- DocVQA 93 papers · best accuracy 96.4
- ShapeNetPart 93 papers · best instance mIoU 90.26
- ImageNet-C 93 papers · best mCE 13.6
- Something-Something V2 93 papers · best accuracy 77.3
- DTD 92 papers · best accuracy 99.59
- Cornell 92 papers · best accuracy 95.9
- HMDB51 92 papers · best accuracy 84.6
- AIME 2024 91 papers · best accuracy 93.33
- RefCOCO testA 91 papers · best accuracy 96.3
- LFW 91 papers · best accuracy 99.86
- TrackingNet 90 papers · best AUC 88.6
- CUB-200 89 papers · best accuracy 99
- IMDB-B 89 papers · best accuracy 81.86
- tieredImageNet 5-way 1-shot 89 papers · best accuracy 88.87
- tieredImageNet 5-way 5-shot 88 papers · best accuracy 95.73
- GOT-10k 88 papers · best AO 82.1
- PASCAL Context 88 papers · best mIoU 63.6
- ScienceQA 86 papers · best accuracy 97.5
- AI2D 86 papers · best accuracy 96.3
- MSR-VTT 86 papers · best CIDEr 63.9
- ogbn-arxiv 86 papers · best accuracy 77.51
- Food-101 86 papers · best accuracy 98.6
- TriviaQA 86 papers · best accuracy 91
- Wisconsin 85 papers · best accuracy 97.38
- HumanML3D 85 papers · best FID 0.03
- MSRVTT-QA 84 papers · best accuracy 75.42
- COLLAB 84 papers · best accuracy 96.6
- LongVideoBench 83 papers · best accuracy 80
- RefCOCOg 82 papers · best accuracy 91.7
- AIME25 82 papers · best accuracy 96
- CelebA-HQ 82 papers · best FID 2.75
- OfficeHome 82 papers · best accuracy 87.1
- Caltech101 82 papers · best accuracy 99.05
- COCO val2017 82 papers · best AP 79.3
- MSVD-QA 81 papers · best accuracy 83.7
- Deep Blending 81 papers · best PSNR 30.7
- MT-Bench 81 papers · best score 9.55
- Actor 80 papers · best accuracy 43.89
- ScanObjectNN PB_T50_RS 80 papers · best overall accuracy 95.18
- RTE 79 papers · best accuracy 92.78
- Tanks&Temples 79 papers · best PSNR 26.36
- Cars196 77 papers · best R@1 94.7
- GTA5->Cityscapes 77 papers · best mIoU 77.5
- MSMT17 77 papers · best mAP 92.8
- GPQA 76 papers · best accuracy 96.5
- D&D 76 papers · best accuracy 95.9
- RefCOCO testB 75 papers · best accuracy 92
- WMT14 En-De 75 papers · best BLEU 32.1
- THUMOS14 74 papers · best mAP@0.5 74.4
- FFHQ 74 papers · best FID 2.21
- MMLU-Pro 73 papers · best accuracy 86.25
- ImageNet-100 73 papers · best accuracy 92.5
- Pascal VOC 73 papers · best mIoU 97.2
- OK-VQA 71 papers · best accuracy 91.9
- NYUv2 71 papers · best RMSE 0.085
- GPQA-Diamond 71 papers · best accuracy 83
- MuSiQue 71 papers · best EM 54.59
- ActivityNet-QA 70 papers · best accuracy 82.99
- RealWorldQA 70 papers · best accuracy 80.2
- DAVIS 2017 70 papers · best J&F (val) 91.3
- PIQA 70 papers · best accuracy 91.8
- WikiText-2 70 papers · best perplexity 2.86
- Flickr30K 69 papers · best CIDEr 104.6
- WinoGrande 68 papers · best accuracy 90.29
- DomainNet 68 papers · best accuracy 84.63
- Waterbirds 68 papers · best worst-group accuracy 93.5
- Cars 68 papers · best accuracy 96.1
- RealSR 68 papers · best PSNR 26.84
- R2R val unseen 67 papers · best success rate 84.4
- Photo 67 papers · best accuracy 97.05
- CIFAR-FS 5-way 1-shot 66 papers · best accuracy 93.25
- BoolQ 66 papers · best accuracy 90.24
- ECSSD 66 papers · best MAE 0.028
- SVAMP 65 papers · best accuracy 98.16
- IMDB-M 64 papers · best accuracy 59
- ImageNet-V2 64 papers · best accuracy 84
- PACS 64 papers · best accuracy 97.63
- TruthfulQA 64 papers · best accuracy 91.4
- ScanQA 64 papers · best CIDEr (val) 109.3
- MathVision 63 papers · best accuracy 55
- CIFAR-FS 5-way 5-shot 63 papers · best accuracy 97.71
- DTU 63 papers · best PSNR 37.03
- MRPC 63 papers · best accuracy 93.32
- LIBERO-Spatial 63 papers · best success rate 99.8
- FGVC-Aircraft 62 papers · best accuracy 96.7
- ENZYMES 62 papers · best accuracy 99.9
- SUN RGB-D 62 papers · best mIoU 55
- OlympiadBench 62 papers · best accuracy 75.1
- MLVU 62 papers · best accuracy 81.2
- ImageNet-Sketch 61 papers · best accuracy 68
- MARS 61 papers · best mAP 96.6
- ARC-C 61 papers · best accuracy 94.58
- CIFAR10-DVS 59 papers · best accuracy 90.5
- MPI-INF-3DHP 59 papers · best PCK 99.9
- OCRBench 58 papers · best score 907
- SIDD 58 papers · best PSNR 42.26
- VideoMME 58 papers · best accuracy 83.2
- DAVIS 58 papers · best PSNR 36.66
- MOT17 58 papers · best MOTA 82
- Aircraft 58 papers · best accuracy 94.9
- FB15k-237 58 papers · best MRR 0.53
- LIBERO-Goal 58 papers · best success rate 99.4
- DRealSR 57 papers · best PSNR 29.678
- VizWiz 56 papers · best accuracy 71.5
- EuroSAT 56 papers · best accuracy 99.74
- DBLP 56 papers · best accuracy 95.73
- QM9 56 papers · best validity (%) 100
- LVBench 56 papers · best accuracy 74.2
- PASCAL-S 56 papers · best MAE 0.0535
- UAV123 55 papers · best AUC 72.3
- KITTI 2015 55 papers · best Fl-all 3.56
- NQ 55 papers · best EM 72
- AIME 2025 54 papers · best accuracy 95
- MSVD 54 papers · best CIDEr 189.4
- CamVid 54 papers · best mIoU 84.54
- ARC-Challenge 53 papers · best accuracy 96.33
- HallusionBench 53 papers · best accuracy 92.32
- ImageNet-LT 52 papers · best accuracy 79.17
- Set5 ×4 52 papers · best PSNR 33.21
- IWSLT14 De-En 52 papers · best BLEU 39.49
- Stanford Dogs 52 papers · best accuracy 96.7
- CIFAR-10 (4000 labels) 52 papers · best error rate 3.38
- DIV2K 52 papers · best PSNR (val) 29.35
- NYU Depth V2 52 papers · best RMSE 0.075
- Flowers102 51 papers · best accuracy 99.7
- Adult 51 papers · best accuracy 89
- Urban100 ×4 51 papers · best PSNR 28.6
- LaSOText 51 papers · best AUC 60.9
- NTU RGB+D cross-subject 51 papers · best accuracy 93.4
- ImageNet 256×256 50 papers · best FID 1.35
- AMC23 50 papers · best accuracy 97.5
- EgoSchema 50 papers · best accuracy 82.9
- Penn Treebank 50 papers · best perplexity 19.45
- MNLI 49 papers · best accuracy 90.69
- BBH 49 papers · best accuracy 94
- SNLI 49 papers · best accuracy 92.9
- ScanObjectNN OBJ-ONLY 49 papers · best accuracy 97.76
- Breakfast 49 papers · best accuracy 97.9
- DPG-Bench 49 papers · best overall 88.28
- MathVerse 49 papers · best accuracy 70.08
- Stanford Online Products 49 papers · best R@1 88.4
- UCF-Crime 48 papers · best AUC 91.58
- ScanObjectNN OBJ-BG 48 papers · best accuracy 99.48
- TACoS 47 papers · best mIoU 50
- VBench 47 papers · best Total Score 86.51
- DeepFashion 47 papers · best FID 3.7
- StrategyQA 46 papers · best accuracy 96.3
- HDTF 46 papers · best FID 7.031
- Set14 ×4 46 papers · best PSNR 30.01
- SYNTHIA → Cityscapes 46 papers · best mIoU 73.5
- TNL2K 46 papers · best AUC 68.6
- LIBERO-Object 46 papers · best success rate 99.8
- UCF-QNRF 46 papers · best MAE 58.13
- BLINK 45 papers · best accuracy 85.3
- MMVP 45 papers · best accuracy 83.33
- HKU-IS 45 papers · best MAE 0.0234
- Something-Something V1 45 papers · best accuracy 68.7
- ShanghaiTech Part B 45 papers · best MAE 6.19
- QQP 44 papers · best accuracy 92.9
- R2R test unseen 44 papers · best success rate 80
- Traffic 44 papers · best MSE 0.294
- DAVIS 2016 44 papers · best J&F (val) 93.6
- WN18RR 44 papers · best MRR 0.526
- ShanghaiTech Part A 44 papers · best MAE 49.34
- NeRF Synthetic 43 papers · best PSNR 35.02
- CUB 5-way 5-shot 43 papers · best accuracy 97.92
- SUN397 43 papers · best accuracy 95.94
- CIFAR-10 unconditional 43 papers · best FID 1.38
- AGNews 43 papers · best accuracy 98.19
- Yelp 43 papers · best accuracy 97.31
- Scene Flow 43 papers · best EPE 0.32
- Coauthor-CS 43 papers · best accuracy 96.96
- QNLI 42 papers · best accuracy 95.1
- ETTh1 42 papers · best MSE 0.3557
- NExT-QA 42 papers · best accuracy 86.8
- DanceTrack 42 papers · best HOTA 75.8
- CIRR 42 papers · best R@1 55.52
- SQA3D 42 papers · best EM@1 65.3
- POPE Adversarial 41 papers · best accuracy 90.5
- Reddit 41 papers · best accuracy 97.28
- Cora node classification 41 papers · best accuracy 89.89
- SQuAD 41 papers · best F1 94.5
- LAMBADA 41 papers · best accuracy 88.52
- CFP-FP 41 papers · best accuracy 99.18
- KIT-ML 41 papers · best FID 0.135
- SUN 41 papers · best H 75.2
- iNaturalist 2018 41 papers · best accuracy 84.1
- PTC 41 papers · best accuracy 81.4
- NTU RGB+D cross-view 41 papers · best accuracy 98.2
- DUT-OMRON 41 papers · best MAE 0.0414
- SVHN→MNIST 41 papers · best accuracy 99.4
- NCI109 40 papers · best accuracy 94.3
- CIFAR-10 IPC=10 40 papers · best accuracy 79.1
- Ref-YouTube-VOS 40 papers · best J&F 81.3
- Flowers 40 papers · best accuracy 99.6
- LOL-v1 40 papers · best PSNR 28.603
- PASCAL-5i 1-shot 40 papers · best mIoU 76.8
- ImageNet, ResNet-50 40 papers · best accuracy 80.1
- ZINC 40 papers · best MAE 0.047
- VSI-Bench 39 papers · best accuracy 67.5
- ReasonSeg 39 papers · best gIoU 73.2
- SOP 39 papers · best R@1 88.2
- WebVision 39 papers · best accuracy 87.44
- MNIST → USPS 39 papers · best accuracy 99.88
- SSv2 39 papers · best accuracy 78
- UCF_CC_50 39 papers · best MAE 96.8
- WebShop 39 papers · best success rate 99
- miniImageNet 1-shot 39 papers · best accuracy 84.78
- FC100 5-way 5-shot 39 papers · best accuracy 88.81
- Omniglot 20-way 1-shot 39 papers · best accuracy 99.63
- NUS-WIDE 38 papers · best mAP 79.7
- MeViS 38 papers · best J&F 63.3
- InfoVQA 38 papers · best accuracy 88.9
- VLCS 38 papers · best accuracy 83.6
- PopQA 38 papers · best EM 56
- LOL 38 papers · best PSNR 27.51
- TREC 38 papers · best accuracy 99.2
- mini-ImageNet 5-shot 38 papers · best accuracy 99.63
- CUB GZSL 38 papers · best harmonic mean H (%) 85.4
- CUB 5-way 1-shot 37 papers · best accuracy 96.4
- COD10K 37 papers · best MAE 0.014
- N-Caltech101 37 papers · best accuracy 94.6
- CIFAR-10, ResNet-18 37 papers · best accuracy 96.84
- FC100 5-way 1-shot 37 papers · best accuracy 57.34
- ImageNet 256x256 class-conditional 37 papers · best FID 1.22
- VOT2016 37 papers · best EAO 0.537
- OpenBookQA 36 papers · best accuracy 92.4
- LLFF 36 papers · best PSNR 28.05
- AlpacaEval 2.0 36 papers · best LC win rate (%) 65.7
- ESC-50 36 papers · best accuracy 99.1
- ScienceQA-IMG 36 papers · best accuracy 97.3
- COIL-20 36 papers · best accuracy 99.8
- XD-Violence 36 papers · best AP 88.66
- ImageNet zero-shot 36 papers · best accuracy 88.31
- ARC-Easy 36 papers · best accuracy 94.3
- NYUD-v2 36 papers · best mIoU 60.9
- XSum 36 papers · best ROUGE-1 49.77
- Omniglot 5-way 1-shot 36 papers · best accuracy 99.97
- Amazon 35 papers · best accuracy 99.88
- CIFAR-100, 400 labels 35 papers · best error rate 32.51
- REVERIE val unseen 35 papers · best success rate 68.11
- GLUE 35 papers · best Avg. 90.1
- Office-Home Avg 35 papers · best accuracy 91.2
- WikiCS 35 papers · best accuracy 83.3
- MS COCO 2014 35 papers · best mIoU (val) 56.7
- Scene15 35 papers · best accuracy 99.78
- PTB 35 papers · best perplexity 18.78
- CNN/DailyMail 35 papers · best ROUGE-1 48.54
- CUHK03 35 papers · best mAP 95.6
- LOLv2-real 35 papers · best PSNR 33.3
- LiveCodeBench 35 papers · best pass@1 76.5
- POPE Random 34 papers · best accuracy 99.3
- SEED-Bench 34 papers · best accuracy 81.1
- CAMO 34 papers · best MAE 0.032
- CIFAR-100 IPC=10 34 papers · best accuracy 64.6
- Computers 34 papers · best accuracy 92.41
- MR 34 papers · best accuracy 90.96
- ActivityNet-Captions 34 papers · best mIoU 56.25
- Rain100L 34 papers · best PSNR 40.67
- Minerva 34 papers · best pass@1 55
- CMU-MOSI 34 papers · best MAE 0.448
- WeMath 34 papers · best accuracy 88.1
- PTC-MR 33 papers · best accuracy 78.86
- Pubmed node classification 33 papers · best accuracy 97.46
- IMDB-BINARY 33 papers · best accuracy 79.9
- Ref-DAVIS17 33 papers · best J&F 80.6
- Replica 33 papers · best PSNR 44.87
- KITTI 33 papers · best AbsRel 0.044
- 2WikiMultiHopQA 33 papers · best EM 72
- ETTm1 33 papers · best MSE 0.2906
- Weather 33 papers · best MSE 0.201
- RealEstate10K 33 papers · best PSNR 37.576
- ORL 33 papers · best accuracy 97.5
- VeRi-776 33 papers · best mAP 95.2
- COCO-20i 1-shot 33 papers · best mIoU 71.33
- LibriSpeech test-clean 33 papers · best WER 1.5
- R2R-CE Val-Unseen 33 papers · best success rate 70.3
- Omniglot 20-way 5-shot 33 papers · best accuracy 99.65
- MedQA 32 papers · best accuracy 91.4
- AQuA 32 papers · best accuracy 91.89
- MultiArith 32 papers · best accuracy 99.8
- Citeseer node classification 32 papers · best accuracy 80.2
- Energy 32 papers · best RMSE 0.087
- Tanks and Temples 32 papers · best PSNR 36.243
- GSO 32 papers · best PSNR 30.443
- LRA ListOps 32 papers · best accuracy 62.75
- LRA Image 32 papers · best accuracy 91.12
- SUBJ 32 papers · best accuracy 97.89
- Charades 32 papers · best mAP 66.2
- Arena-Hard 32 papers · best win rate (%) 87
- CPLFW 32 papers · best accuracy 94.57
- hopper-medium 32 papers · best normalized score 109.8
- HICO-DET Default Full 32 papers · best mAP 48.16
- GLUE RTE 32 papers · best accuracy 94.62
- Rain100H 32 papers · best PSNR 34.57
- ARC-E 32 papers · best accuracy 92.4
- AWA2 32 papers · best H 86.39
- COCO-20i 5-shot 32 papers · best mIoU 71.9
- CIFAR-100, 10000 labels 32 papers · best error rate 19.01
- Places365 32 papers · best accuracy 65.1
- PRW 32 papers · best mAP 62
- IFEval 32 papers · best score 85.4
- OVIS 32 papers · best AP (val) 53.2
- NoCaps 31 papers · best CIDEr 124.4
- NAS-Bench-201 CIFAR-100 31 papers · best accuracy 74.25
- ogbn-products 31 papers · best accuracy 87.36
- CIFAR-10 (40 labels) 31 papers · best error rate 4.24
- ActivityNet 31 papers · best accuracy 94.7
- ImageNet-10 31 papers · best accuracy 99.8
- SYSU-MM01 all-search 31 papers · best mAP 90.8
- Oxford Pets 31 papers · best accuracy 99.25
- LRA Text 31 papers · best accuracy 91.1
- LRA Retrieval 31 papers · best accuracy 92.45
- ADE20K (UperNet) 31 papers · best mIoU 54.5
- CUHK03 detected 31 papers · best Rank-1 88.9
- ETTh2 31 papers · best MSE 0.276
- MOT20 31 papers · best IDF1 80.9
- ImageNet, ResNet-18 31 papers · best accuracy 73.1
- SVT 31 papers · best accuracy 97.8
- KITTI Eigen split 31 papers · best Abs Rel 0.0251
- YouTube-VIS 2019 31 papers · best AP (val) 68.9
- AWA2 GZSL 31 papers · best harmonic mean H 92.3
- CIFAR-10 clean 31 papers · best accuracy 96.28
- MME perception 30 papers · best score 1726.77
- NAS-Bench-201 CIFAR-10 30 papers · best accuracy 95.46
- CIFAR-100, 2500 labels 30 papers · best error rate 23.78
- ImageNet-1K IPC=10 30 papers · best accuracy 58.1
- CIFAR-10 IPC=50 30 papers · best accuracy 87
- CIFAR-100 IPC=50 30 papers · best accuracy 72.48
- Manga109 ×4 30 papers · best PSNR 32.65
- Celeb-DF 30 papers · best AUC 99.97
- MIMIC-CXR 30 papers · best BLEU-4 0.193
- ETTm2 30 papers · best MSE 0.164
- OTB-2015 30 papers · best AUC 92.6
- Peptides-struct 30 papers · best MAE 0.2421
- MME-P 30 papers · best score 1803
- WMT14 De-En 30 papers · best BLEU 37.9
- S3DIS 30 papers · best mIoU 80.1
- LSUN Bedroom 30 papers · best FID 2.55
- SVHN (1000 labels) 30 papers · best error rate 1.41
- Vimeo90K 30 papers · best PSNR 37.32
- YTF 30 papers · best accuracy 97.92
- Kinetics-600 30 papers · best accuracy 91.8
- roman-empire 29 papers · best accuracy 93.87
- CommonsenseQA 29 papers · best accuracy 99.67
- CIFAR-10 (250 labels) 29 papers · best error rate 4.21
- Urban100 ×2 29 papers · best PSNR 34.47
- CoLA 29 papers · best Matthews correlation 68.5
- HICO-DET Default Rare 29 papers · best mAP 50.74
- LRA average 29 papers · best accuracy 88.26
- SVTP 29 papers · best accuracy 93.3
- ACM 29 papers · best accuracy 95.3
- ShanghaiTech 29 papers · best AUC 98.35
- LIBERO average 29 papers · best success rate 99.9
- TerraIncognita 29 papers · best accuracy 66.5
- SUN GZSL 29 papers · best harmonic mean H (%) 71.9
- iLIDS-VID 29 papers · best Rank-1 97.2
- Omniglot 5-way 5-shot 29 papers · best accuracy 99.93
- CSQA 28 papers · best accuracy 87.2
- SEED 28 papers · best score 76.5
- MMBench-EN 28 papers · best accuracy 88.9
- ADE20K semantic segmentation 28 papers · best mIoU 58.6
- CSL-Daily 28 papers · best BLEU-4 32.08
- N-Cars 28 papers · best accuracy 99.9
- ESOL 28 papers · best RMSE 0.279
- REDDIT-BINARY 28 papers · best accuracy 95.28
- LRA Pathfinder 28 papers · best accuracy 96.69
- FMNIST 28 papers · best accuracy 99.21
- CIFAR-10, ResNet-20 28 papers · best accuracy 93.58
- VQA 28 papers · best accuracy (test-dev) 81.11
- ImageNet-1K zero-shot 28 papers · best accuracy 84.1
- CIFAR-100, ResNet-18 28 papers · best accuracy 87.6
- Occ3D-nuScenes 28 papers · best mIoU (val) 55.1
- PASCAL-5i 5-shot 28 papers · best mIoU 82.6
- YAGO3-10 28 papers · best MRR 0.696
- DVS-CIFAR10 28 papers · best accuracy 87.63
- ModelNet10 28 papers · best accuracy 98.46
- CLEVR 28 papers · best accuracy 99.6
- CMU-MOSEI 28 papers · best MAE 0.489
- WMT14 En-Fr 28 papers · best BLEU 45.15
- S3DIS 6-fold 27 papers · best mIoU 94.7
- RGBT234 27 papers · best MPR 94
- MMLU 5-shot 27 papers · best accuracy 84.7
- DBPedia 27 papers · best accuracy 99.32
- Minerva Math 27 papers · best pass@1 47.4
- Set5 ×2 27 papers · best PSNR 39.72
- CS 27 papers · best accuracy 96.58
- ScanNet200 27 papers · best mIoU (val) 39.6
- Diving48 27 papers · best accuracy 94.9
- MOT16 27 papers · best MOTA 76.6
- CIFAR-10.1 27 papers · best accuracy 95.5
- AwA 27 papers · best accuracy 91.05
- Flickr 27 papers · best accuracy 93.27
- Coauthor Physics 27 papers · best accuracy 98.85
- VOT2018 27 papers · best EAO 0.489
- OTB100 27 papers · best AUC 73.3
- VGGSound 26 papers · best accuracy 97.22
- Tiny-ImageNet IPC=10 26 papers · best accuracy 56.2
- RealBlur-J 26 papers · best PSNR 34.35
- Gait3D 26 papers · best Rank-1 80.5
- ADE20K val, UperNet 26 papers · best mIoU 54
- CALFW 26 papers · best verification accuracy (%) 97.07
- PC-59 26 papers · best mIoU 64.4
- VideoMME Long 26 papers · best accuracy 78.1
- pMNIST 26 papers · best accuracy 99.1
- DVS128 Gesture 26 papers · best accuracy 99.62
- Synthetic-NeRF 26 papers · best PSNR 34.68
- IU-Xray 26 papers · best BLEU-4 0.266
- SYSU-MM01 indoor-search 26 papers · best Rank-1 90.1
- MS COCO Karpathy 26 papers · best CIDEr 145.4
- ogbg-molhiv 26 papers · best ROC-AUC 87.4
- IMDB-MULTI 26 papers · best accuracy 64.2
- PHOENIX14T 25 papers · best BLEU-4 33.62
- Boston 25 papers · best test RMSE 2.369
- Tox21 25 papers · best ROC-AUC 85.74
- HIDE 25 papers · best PSNR 32.83
- Office-31 Avg 25 papers · best accuracy 96.9
- Set14 ×2 25 papers · best PSNR 34.83
- amazon-ratings 25 papers · best accuracy 66.79
- ScanObjectNN 25 papers · best accuracy 97.2
- Animal-10N 25 papers · best accuracy 96.371
- BBCSport 25 papers · best accuracy 98.37
- Amazon-Photo 25 papers · best accuracy 96.65
- CrowdPose 25 papers · best AP 79.4
- DVS-Gesture 25 papers · best accuracy 99.31
- CUHK-SYSU 25 papers · best mAP 97.8
- tieredImageNet 1-shot 25 papers · best accuracy 96.76
- SYNTHIA->Cityscapes (16 classes) 25 papers · best mIoU 73
- CIFAR-10 (α=0.1) 24 papers · best accuracy 94.74
- PEMS08 24 papers · best MAE 6.59
- Places-LT 24 papers · best accuracy 44.2
- RegDB visible-to-infrared 24 papers · best Rank-1 95.19
- Protein 24 papers · best RMSE 0.522
- Office-31 24 papers · best accuracy 92.7
- CIFAR-100 IPC=1 24 papers · best accuracy 48
- BACE 24 papers · best ROC-AUC 92
- ListOps 24 papers · best accuracy 99.9
- GLUE SST-2 24 papers · best accuracy 96.4
- Tiny-ImageNet, IPC 50 24 papers · best accuracy 59
- ZINC250k 24 papers · best FCD 0.401
- HICO-DET Default Non-Rare 24 papers · best mAP 47.74
- Handwritten 24 papers · best accuracy 99.98
- CIFAR-100-LT IF=100 24 papers · best accuracy 61.33
- Physics 24 papers · best accuracy 98.78
- REDDIT-B 24 papers · best accuracy 94.5
- SPAQ 24 papers · best SRCC 0.947
- VSR 24 papers · best accuracy 84.3
- CIFAR-10N Worst 24 papers · best accuracy 96.32
- WMT16 En-Ro 24 papers · best BLEU 37.5
- O-HAZE 24 papers · best PSNR 26.32
- YouTube-VIS 2021 24 papers · best AP (val) 64.6
- DND 24 papers · best PSNR 40.25
- USPS→MNIST 24 papers · best accuracy 99.3
- Cityscapes → Foggy Cityscapes 24 papers · best mAP 64.7
- tieredImageNet 5-shot 24 papers · best accuracy 98.67
- MIT67 24 papers · best accuracy 98.58
- Food-101N 24 papers · best accuracy 93.42
- Yale 24 papers · best accuracy 98.18
- Bench2Drive closed-loop 24 papers · best Driving Score 89.42
- SOD 24 papers · best MAE 0.061
- MMHal-Bench 23 papers · best Score 4.26
- NAS-Bench-201 ImageNet-16-120 23 papers · best accuracy 47.59
- NAVSIM navtest 23 papers · best PDMS 92.4
- Wine 23 papers · best RMSE 0.46
- ImageNet-Dogs 23 papers · best accuracy 93.5
- OBQA 23 papers · best accuracy 89.6
- Electricity 23 papers · best MSE 0.097
- PIE-Bench 23 papers · best PSNR 29.44
- Pets 23 papers · best accuracy 97.97
- Arxiv 23 papers · best accuracy 76.45
- ImageNet 16-shot 23 papers · best accuracy 78.48
- halfcheetah-medium 23 papers · best normalized score 98.7
- 3DMatch 23 papers · best RR (%) 95.8
- FSS-1000 1-shot 23 papers · best mIoU 93.4
- HICO-DET Known Object Full 23 papers · best mAP 51.34
- ScanRefer 23 papers · best Acc@0.25 63.9
- FreeSolv 23 papers · best RMSE 0.28
- DukeMTMC 23 papers · best mAP 96.3
- Hopper 23 papers · best return 3810
- HalfCheetah 23 papers · best return 10660.59
- Split CIFAR100 23 papers · best accuracy 89
- LibriSpeech test-other 23 papers · best WER 2.73
- PACS average 23 papers · best accuracy 93.9
- REVERIE test unseen 23 papers · best success rate 57.81
- ModelNet40 classification 23 papers · best accuracy 94.6
- Breast Cancer 23 papers · best accuracy 99
- PPI 23 papers · best micro-F1 99.5
- SST 23 papers · best accuracy 96.6
- LogicVista 23 papers · best accuracy 68.5
- VideoMMMU 23 papers · best accuracy 68.1
- WMT16 Ro-En 23 papers · best BLEU 39.1
- UCF101 split 1 23 papers · best accuracy 98.8
- Kinetics 23 papers · best accuracy 91.2
- Model size 23 papers · best parameters (M) 4.1
- COCO-Panoptic 22 papers · best PQ 61.2
- DreamBench 22 papers · best CLIP-T 0.357
- DiDeMo 22 papers · best R@1 64.1
- BDGP 22 papers · best accuracy 99.2
- CIFAR-10-C 22 papers · best accuracy 94.59
- MEAD 22 papers · best FID 5.54
- ImageNet-1K IPC=50 22 papers · best accuracy 68.4
- WebQSP 22 papers · best F1 87.7
- RealBlur-R 22 papers · best PSNR 41.29
- CIRCO 22 papers · best mAP@5 39.82
- CREMA-D 22 papers · best accuracy 85.15
- sMNIST 22 papers · best accuracy 99.75
- CelebA-HQ-256 22 papers · best FID 5.11
- Bamboogle 22 papers · best EM 63.2
- HIV 22 papers · best ROC-AUC (%) 86
- Mini-ImageNet 22 papers · best accuracy 91.02
- VITON-HD 22 papers · best SSIM 0.928
- SSCBench-KITTI-360 22 papers · best mIoU 22.39
- CSIQ 22 papers · best SRCC 0.983
- TID2013 22 papers · best SRCC 0.972
- CivilComments 22 papers · best worst-group accuracy 80.4
- Permuted MNIST 22 papers · best accuracy 98.18
- VOC 22 papers · best mIoU 96.4
- N-MNIST 22 papers · best accuracy 99.7
- SQuAD 2.0 22 papers · best F1 (dev) 91.2
- SOTS-Indoor 22 papers · best PSNR 42.64
- SST-5 22 papers · best accuracy 58.32
- WiC 22 papers · best accuracy 72.1
- GREW 22 papers · best Rank-1 88.2
- RxR-CE Val-Unseen 22 papers · best success rate 73.6
- Office-31 average 22 papers · best accuracy 93.5
- VQA-RAD 22 papers · best accuracy 87.1
- VOT2020 22 papers · best EAO 0.729
- ImageNet VID 22 papers · best mAP (val) 92.1
- VIPeR 22 papers · best Rank-1 68
- PEMS04 21 papers · best MAE 15.61
- Concrete 21 papers · best Test log-likelihood -2.842
- ImageNet ReaL 21 papers · best accuracy 91.1
- BBBP 21 papers · best ROC-AUC 76.7
- ClinTox 21 papers · best ROC-AUC 95.75
- CelebA 64×64 21 papers · best FID 1.16
- FreiHAND 21 papers · best PA-MPJPE 4.9
- COPA 21 papers · best accuracy 92
- NLVR2 test-P 21 papers · best accuracy 92.58
- CUHK03 labeled 21 papers · best Rank-1 90.39
- Rain200H 21 papers · best PSNR 33.53
- antmaze-umaze-diverse 21 papers · best normalized score 99.4
- CTW1500 21 papers · best F-measure 90.6
- Total-Text 21 papers · best F-measure 90.98
- FSC-147 21 papers · best MAE 5.74
- CCV 21 papers · best accuracy 64.46
- Walker2d-medium 21 papers · best normalized score 111.7
- Video-MME overall 21 papers · best accuracy 77.4
- Beauty 21 papers · best NDCG@10 0.0698
- text8 21 papers · best BPC 1.06
- LOLv2-synthetic 21 papers · best PSNR 31.019
- FSS-1000 5-shot 21 papers · best mIoU 93.5
- Urban100 21 papers · best PSNR 30.48
- Peptides-func 21 papers · best AP 0.7571
- GenEval Position 21 papers · best score 0.99
- Emu Edit 21 papers · best DINO 0.897
- AVE 21 papers · best accuracy 87.41
- CIFAR-10 (IID) 21 papers · best accuracy 94.15
- Office-31 A→W 21 papers · best accuracy 98.6
- LIVE 21 papers · best SRCC 0.986
- Office-Home average 21 papers · best accuracy 87.2
- CIFAR-10-LT IF=100 21 papers · best accuracy 92.33
- Colored MNIST 21 papers · best accuracy 96.75
- PubMedQA 21 papers · best accuracy 95.44
- Amazon Computers 21 papers · best accuracy 92.51
- MSLS 21 papers · best R@1 (val) 94.5
- SIDD benchmark 21 papers · best PSNR (dB) 40.23
- Office-31 D->A 21 papers · best accuracy 89.7
- HMDB51 split 1 21 papers · best accuracy 83.1
- ALFWorld 20 papers · best success rate 98
- CIFAR-100, α=0.1 20 papers · best accuracy 94.39
- In-Shop 20 papers · best R@1 95.14
- MultiNLI 20 papers · best worst-group accuracy 78.95
- WikiText 20 papers · best perplexity 5.55
- LLFF (3 views) 20 papers · best PSNR 23.23
- SYSU-MM01 all-search single-shot 20 papers · best mAP 79.85
- RegDB infrared-to-visible 20 papers · best Rank-1 94.22
- AlpacaEval 20 papers · best win rate 94.78
- CIFAR-10 IPC=1 20 papers · best accuracy 68.9
- Urban100 ×3 20 papers · best PSNR 30.28
- CWQ 20 papers · best F1 84.8
- antmaze-large-play 20 papers · best normalized score 92.3
- antmaze-large-diverse 20 papers · best normalized score 94.2
- gRefCOCO 20 papers · best gIoU (val) 78.4
- VQA 2.0 20 papers · best accuracy (val) 71
- B100 ×4 20 papers · best PSNR 28
- WTQ 20 papers · best accuracy 81.32
- Computer 20 papers · best accuracy 93.99
- ADE20K val (Semantic FPN) 20 papers · best mIoU 50
- KonIQ-10k 20 papers · best SRCC 0.938
- 50Salads 20 papers · best accuracy 89.5
- NABirds 20 papers · best accuracy 92
- UCSD Ped2 20 papers · best AUC 99.1
- CR 20 papers · best accuracy 91.8
- Gigaword 20 papers · best ROUGE-1 45.7
- Nordland 20 papers · best R@1 97.7
- Office-31 W→A 20 papers · best accuracy 78.9
- TuSimple 20 papers · best accuracy 97.18
- FFHQ 256² 20 papers · best FID 3.39
- ICDAR 2015 20 papers · best F-measure 91.34
- Urban100 x4 20 papers · best PSNR (dB) 28.81
- TGIF-QA FrameQA 20 papers · best accuracy 64.2
- ImageNet (ILSVRC2012) 20 papers · best accuracy 77.2
- PASCAL VOC 2007 trainval 20 papers · best CorLoc 83.5
- COCO val2017 Mask R-CNN 1x 20 papers · best box AP 49.9
- HumanML3D text-to-motion 20 papers · best FID 0.032
- QVHighlights 19 papers · best R1@0.5 70.05
- Omniglot 19 papers · best accuracy 99.4
- Caltech-256 19 papers · best accuracy 95.9
- ImageNet 64×64 19 papers · best FID 1.035
- THuman2.0 19 papers · best P2S 0.359
- DFDC 19 papers · best AUC 91.33
- Corel5k 19 papers · best average precision 0.604
- Rain200L 19 papers · best PSNR 42.3
- MSRA-TD500 19 papers · best F-measure 90.3
- RACE 19 papers · best accuracy 89.69
- ISOLET 19 papers · best accuracy 97.45
- psMNIST 19 papers · best accuracy 98.55
- AVA v2.2 19 papers · best mAP 45.1
- TabMWP 19 papers · best accuracy 99.2
- Reuters 19 papers · best accuracy 0.951
- COCO Object 19 papers · best mIoU 58
- Set14 19 papers · best PSNR 32.93
- Horse->Zebra 19 papers · best FID 34
- PROTEINS graph classification 19 papers · best accuracy 80.7
- ToxCast 19 papers · best ROC-AUC 75.51
- ECL 19 papers · best MSE 0.118
- MOVi-E 19 papers · best FG-ARI 84.1
- NFS 19 papers · best AUC 71.3
- LRS2 19 papers · best WER (%) 4.9
- SOTS-Outdoor 19 papers · best PSNR 39.18
- PATTERN 19 papers · best accuracy 89.177
- CIFAR-100 zero-shot 19 papers · best accuracy 97.9
- IIIT5K 19 papers · best accuracy 98.5
- LLFF 3-view 19 papers · best PSNR 22.39
- ImageNet 256x256 19 papers · best FID 1.15
- CIFAR-100 (ResNet32x4 -> ResNet8x4) 19 papers · best accuracy 78.76
- ImageNet-1K linear probing 19 papers · best accuracy 83.6
- DUTS 19 papers · best MAE 0.027
- USPS 19 papers · best accuracy 0.97
- Sintel Clean 19 papers · best EPE 0.79
- Sintel Final 19 papers · best EPE 1.88
- TempCompass 19 papers · best accuracy 77.2
- UCF101 5-way 5-shot 19 papers · best accuracy 99.9
- GTA5 to Cityscapes 19 papers · best mIoU 59.3
- PRID-2011 19 papers · best Rank-1 96.5
- CIFAR-100 (from scratch) 19 papers · best accuracy 86.4
- ASDiv 18 papers · best accuracy 97.24
- PEMS03 18 papers · best MAE 11.66
- PEMS07 18 papers · best MAE 17.19
- FOLIO 18 papers · best accuracy 95.6
- LasHeR 18 papers · best PR 78.1
- CoraFull 18 papers · best accuracy 73.6
- MJHQ-30K 18 papers · best FID 1.46
- MMMU-Pro 18 papers · best accuracy 55.5
- TAP-Vid DAVIS 18 papers · best AJ 68.1
- ModelNet40 z/SO(3) 18 papers · best accuracy 92.6
- Bench2Drive 18 papers · best Driving Score 95.2
- NoisyMNIST 18 papers · best accuracy 99.84
- VOC20 18 papers · best mIoU 98
- ShanghaiTech A 18 papers · best MAE 33.85
- COIN 18 papers · best accuracy 96.4
- CIFAR-10 (α=0.5) 18 papers · best accuracy 94.39
- GTEA 18 papers · best accuracy 86.1
- StackOverflow 18 papers · best accuracy 88.2
- PC-459 18 papers · best mIoU 25
- Pitts30k 18 papers · best R@1 95.8
- Wiki 18 papers · best accuracy 65.42
- COCO Caption 18 papers · best CIDEr 143.1
- COX2 18 papers · best accuracy 92.76
- LSUN Church 18 papers · best FID 2.15
- IC13 18 papers · best accuracy 98.7
- AMC 18 papers · best accuracy 96.6
- A2D-Sentences 18 papers · best mAP 61.4
- Set14 ×3 18 papers · best PSNR 31.13
- ADE20K Semantic FPN 18 papers · best mIoU 49.4
- VideoMME (w/o sub) 18 papers · best accuracy 72
- CIFAR-100 Sym-50% 18 papers · best accuracy 82.07
- 300W Full 18 papers · best NME 2.84
- UCSD 18 papers · best MAE 0.75
- Visual Genome PredCls 18 papers · best mR@100 42.2
- LIBERO (average of 4 suites) 18 papers · best success rate 98.5
- ZINC-12K 17 papers · best MAE 0.056
- MathQA 17 papers · best accuracy 85.4
- SYSU-MM01 indoor-search single-shot 17 papers · best mAP 89.37
- BSD100 ×4 17 papers · best PSNR 28.17
- SIDER 17 papers · best ROC-AUC 68.52
- 2Wiki 17 papers · best F1 82.1
- CALVIN ABC->D 17 papers · best average length 4.77
- FEVER 17 papers · best accuracy 97.82
- CIFAR-10 conditional 17 papers · best FID 1.37
- UCF-101 zero-shot 17 papers · best accuracy 91.5
- CIFAR-10 Sym-50% 17 papers · best accuracy 96.6
- DukeMTMC-VideoReID 17 papers · best mAP 97.1
- Scene 17 papers · best Average Precision 0.9048
- CHAIR 17 papers · best CHAIR_S 16.8
- YouTube-VOS 2019 17 papers · best G (val) 87.4
- TikTok 17 papers · best FVD 117.81
- Flower102 17 papers · best accuracy 98.2
- TabFact 17 papers · best accuracy 91.16
- Synthetic 17 papers · best accuracy 99.91
- COCO Karpathy 17 papers · best CIDEr 152.1
- PA100K 17 papers · best mA 91.09
- ImageNet (ResNet34 -> ResNet18) 17 papers · best accuracy 72.89
- MOVi-C 17 papers · best FG-ARI 81.9
- ImageNette 17 papers · best accuracy 99.59
- TextCaps 17 papers · best CIDEr 163.7
- AG's News 17 papers · best accuracy 95.9
- CIFAR-100 clean 17 papers · best accuracy 81.55
- SVHN, 250 labels 17 papers · best error rate 1.43
- VisDA-2017 17 papers · best accuracy 98.4
- Twitter 17 papers · best accuracy 95.2
- ImageNet-1K fine-tuning 17 papers · best accuracy 88.1
- Sintel test final 17 papers · best EPE 1.48
- CelebA 64x64 17 papers · best FID 1.34
- Set5 ×3 17 papers · best PSNR 35.15
- ModelNet40 z/z 17 papers · best accuracy 92.7
- ModelNet40 (SO(3)/SO(3)) 17 papers · best accuracy 92.6
- ShapeNet Chair 17 papers · best 1-NNA CD (%) 45.78
- ImageNet 1% labels 17 papers · best accuracy 82.1
- MPQA 17 papers · best accuracy 93.7
- CUB 5-shot 17 papers · best accuracy 95.96
- MLVU M-Avg 17 papers · best accuracy 83.9
- HMDB51 5-way 1-shot 17 papers · best accuracy 88.4
- HMDB51 5-way 5-shot 17 papers · best accuracy 93.9
- UCF101 5-way 1-shot 17 papers · best accuracy 99.3
- MNIST → SVHN 17 papers · best accuracy 97
- Market-1501 (single query) 17 papers · best mAP 90.1
- LIP 17 papers · best mIoU (val) 61.17
- CIFAR-100 transfer 17 papers · best accuracy 93.1
- Set5 x4 17 papers · best PSNR (dB) 33.15
- POPE Popular 16 papers · best accuracy 93.2
- HAR 16 papers · best accuracy 98.1
- AIST++ 16 papers · best FID_k 11.67
- NC4K 16 papers · best MAE 0.02
- CIFAR-100-C 16 papers · best accuracy 77
- NExT-GQA 16 papers · best mIoU 43
- STL-10 (1000 labels) 16 papers · best error rate 4.01
- ImageNet 64x64 16 papers · best FID 1.07
- ARC 16 papers · best accuracy 96.2
- DFD 16 papers · best AUC 99.64
- GLUE QNLI 16 papers · best accuracy 94.8
- antmaze-medium-play 16 papers · best normalized score 100
- antmaze-medium-diverse 16 papers · best normalized score 94.6
- T2I-CompBench 16 papers · best Color 0.8753
- MagicBrush 16 papers · best DINO 0.971
- Kinetics-Sounds 16 papers · best accuracy 91.44
- hopper-medium-replay 16 papers · best normalized score 111.1
- Criteo 16 papers · best AUC 0.8477
- ALOI-100 16 papers · best accuracy 88.7
- GenEval Counting 16 papers · best score 0.99
- Iris 16 papers · best accuracy 99.7
- ModelNet40 5-way 10-shot 16 papers · best accuracy 97.9
- ModelNet40 5-way 20-shot 16 papers · best accuracy 99.2
- C4 16 papers · best perplexity 6.38
- MUV 16 papers · best ROC-AUC 85
- ImageNet-Segmentation 16 papers · best mIoU 71.04
- KADID 16 papers · best PLCC 0.972
- RGBNT100 16 papers · best mAP 91.6
- Split MNIST 16 papers · best accuracy 99.834
- hopper-medium-v2 16 papers · best normalized score 103.7
- ADNI 16 papers · best accuracy 96.19
- HR-Bench 4K 16 papers · best accuracy 86.7
- FFHQ 256×256 16 papers · best FID 2.24
- AndroidWorld 16 papers · best success rate 80.2
- CIFAR-100 (VGG13 -> VGG8) 16 papers · best accuracy 76.19
- Berkeley 16 papers · best NoC@90 1.43
- SLAKE 16 papers · best accuracy 88.1
- CIFAR-100 Sym-20% 16 papers · best accuracy 92.67
- RAF-DB 16 papers · best accuracy 94.04
- ImageNet linear probe 16 papers · best accuracy 86.4
- Caltech101-20 16 papers · best accuracy 92.28
- Sintel test clean 16 papers · best EPE 0.68
- Human3.6M Protocol 1 16 papers · best MPJPE (mm) 33.5
- Occluded-Duke 16 papers · best mAP 76.4
- Rotated MNIST 16 papers · best accuracy 98.5
- SNLI-VE 16 papers · best accuracy 84.95
- ReferItGame 16 papers · best accuracy 77.5
- Flickr30K Entities 16 papers · best accuracy 81.89
- CIFAR-10 Sym-20% 16 papers · best accuracy 97.2
- ETH-UCY AVG 16 papers · best ADE 0.11
- Amazon-670K 16 papers · best P@1 53.02
- D4RL hopper-medium-replay 16 papers · best normalized score 110.7
- DomainNet Avg. 16 papers · best accuracy 74.5
- Kinetics 5-way 5-shot 16 papers · best accuracy 98.4
- CV-Bench 16 papers · best accuracy 94.77
- 300W Common 16 papers · best NME 2.51
- FB15k 16 papers · best MRR 0.861
- Mini-Kinetics 16 papers · best accuracy 82.6
- CIFAR-10, 1000 labels 16 papers · best error rate 4.48
- Cityscapes to Foggy Cityscapes 16 papers · best mAP 47.9
- Visual Genome SGCls 16 papers · best mR@100 27.6
- SYNTHIA->Cityscapes (13 classes) 16 papers · best mIoU 75.1
- NAVSIM v1 navtest 16 papers · best PDMS 95.1
- A-OKVQA 15 papers · best accuracy 87.5
- MS-COCO 30K 15 papers · best FID 4.42
- ALOI 15 papers · best accuracy 97
- Yacht 15 papers · best RMSE 0.13
- GTSRB 15 papers · best accuracy 99.2
- ImageNet-1K val, 224² 15 papers · best accuracy 85.57
- ObjectNet 15 papers · best accuracy 77.9
- HMDB-51 zero-shot 15 papers · best accuracy 61.4
- Rain800 15 papers · best PSNR 30.75
- Camelyon17 15 papers · best accuracy 96
- SMD 15 papers · best F1 99.95
- BSDS500 15 papers · best ODS 0.848
- YouCook2 15 papers · best CIDEr 136.1
- LJSpeech 15 papers · best MOS 4.48
- CIFAR-20 15 papers · best accuracy 71.6
- walker2d-medium-replay 15 papers · best normalized score 97.4
- ImageNet 512×512 15 papers · best FID 1.45
- ImageNet-256 15 papers · best FID 1.02
- V-COCO 15 papers · best AP_role 66.6
- ModelNet40 10-way 10-shot 15 papers · best accuracy 94.8
- ModelNet40 10-way 20-shot 15 papers · best accuracy 96.8
- SportsMOT 15 papers · best HOTA 85.6
- Occluded-REID 15 papers · best mAP 89.34
- GSM-Hard 15 papers · best accuracy 76.1
- LRA Path-X 15 papers · best accuracy 98.65
- RDT-B 15 papers · best accuracy 96.3
- MSVR310 15 papers · best mAP 71.7
- CVUSA 15 papers · best R@1 99.67
- CVACT 15 papers · best R@1 (val) 97.65
- HW 15 papers · best accuracy 99.8
- UMLS 15 papers · best MRR 0.972
- GaoFen2 15 papers · best PSNR 49.34
- COCO detection 15 papers · best AP 56.7
- CLUSTER 15 papers · best accuracy 80.113
- CUHK-PEDES 15 papers · best Rank-1 79.31
- Office-31 A→D 15 papers · best accuracy 98
- refCOCOg val(U) 15 papers · best cIoU 80.6
- YouTube-VOS 2018 15 papers · best G (val) 87.6
- GrabCut 15 papers · best NoC@90 1.2
- SBD 15 papers · best NoC@90 3.96
- CIFAR10 zero-shot 15 papers · best accuracy 99.6
- AMC 2023 15 papers · best pass@1 87.5
- IEMOCAP 15 papers · best accuracy 87.5
- VIP 15 papers · best mIoU 45.9
- ML-1M 15 papers · best NDCG@10 0.8392
- DTU (3-view) 15 papers · best PSNR 23.79
- FFHQ 256x256 15 papers · best FID 2.58
- Sudoku 15 papers · best accuracy 99.9
- ShanghaiTech B 15 papers · best MAE 5.6
- ImageNet, MobileNetV2 15 papers · best accuracy 74.3
- DUTS-TE 15 papers · best MAE 0.0281
- CIFAR-FS 1-shot 15 papers · best accuracy 88.67
- ZINC-full 15 papers · best MAE 0.016
- REAL275 15 papers · best 5°2cm 61.7
- Oxford-IIIT Pets 15 papers · best accuracy 99.6
- DrawBench 15 papers · best ImageReward 1.71
- ImageNet 10% labels 15 papers · best accuracy 85.9
- MathVista (testmini) 15 papers · best accuracy 83.1
- miniImageNet 5-shot 5-way 15 papers · best accuracy 87.4
- NTU RGB+D X-Sub 15 papers · best accuracy 95
- D4RL hopper-medium 15 papers · best normalized score 109.4
- WN18 15 papers · best MRR 0.96
- FCVID 15 papers · best mAP 86.4
- Paris6k 15 papers · best mAP 97.8
- aPY 15 papers · best H 56.2
- Market-1501 -> DukeMTMC-reID 15 papers · best mAP 73.2
- DukeMTMC-reID -> Market-1501 15 papers · best mAP 85.3
- CIFAR-10 (from scratch) 15 papers · best accuracy 98.28
- ImageNet-1K val (224x224) 15 papers · best accuracy 86.3
- LLaVA-Bench 14 papers · best score 91.4
- CIFAR10 α=0.05 14 papers · best accuracy 83.31
- ImageNet-1k linear probe 14 papers · best accuracy 89.6
- HICO-DET (Full) 14 papers · best mAP 46.01
- AFHQ 14 papers · best FID 6.19
- 100Leaves 14 papers · best accuracy 96.31
- DTU (3 views) 14 papers · best PSNR 21.86
- TinyImageNet IPC=1 14 papers · best accuracy 39.2
- NTU RGB+D 120 X-Sub 14 papers · best accuracy 95.7
- VoiceBank+DEMAND 14 papers · best PESQ 3.35
- TAP-Vid Kinetics 14 papers · best AJ 75.1
- DROP 14 papers · best accuracy 91.9
- COMPAS 14 papers · best accuracy 0.881
- Text 14 papers · best accuracy 91.02
- Dogs 14 papers · best accuracy 89.21
- antmaze-umaze 14 papers · best normalized score 100
- SWaT 14 papers · best F1 98.9
- Road Anomaly 14 papers · best AP 98.66
- IWSLT14 En-De 14 papers · best BLEU 31.88
- WSC 14 papers · best accuracy 90.1
- CIFAR-10, ResNet-56 14 papers · best accuracy 94.85
- RTTS 14 papers · best FADE 0.417
- Avazu 14 papers · best AUC 0.7981
- UDM10 14 papers · best LPIPS 0.13
- SPMCS 14 papers · best PSNR 25.32
- MMB 14 papers · best score 86.3
- LIVEC 14 papers · best SRCC 0.918
- PETA 14 papers · best mA 92.25
- JHU-Crowd++ 14 papers · best MAE 47.3
- CIFAR-10 asymmetric noise 0.4 14 papers · best accuracy 95.22
- REDS4 14 papers · best PSNR 34.35
- MAWPS 14 papers · best accuracy 95.7
- Set5 14 papers · best PSNR 37.37
- CIFAR-100, WRN-40-2 → WRN-16-2 14 papers · best accuracy 77.5
- Exchange 14 papers · best MSE 0.298
- 11-dataset average, 1-shot 14 papers · best accuracy 76.1
- KonIQ 14 papers · best SRCC 0.967
- RainDrop 14 papers · best PSNR 33.81
- CIFAR100-LT IF=50 14 papers · best accuracy 65.23
- RGBNT201 14 papers · best mAP 85
- pokec 14 papers · best accuracy 86.46
- minesweeper 14 papers · best ROC AUC 98.99
- WorldView-II 14 papers · best PSNR 42.578
- A-847 14 papers · best mIoU 18.3
- halfcheetah-medium-v2 14 papers · best normalized score 74.1
- walker2d-medium-replay-v2 14 papers · best normalized score 98.5
- HDM05 14 papers · best accuracy 97.52
- MNLI-m 14 papers · best accuracy 90.35
- SOTS 14 papers · best PSNR 37.32
- CIFAR-100 (ResNet-56 -> ResNet-20) 14 papers · best accuracy 73.21
- Tokyo24/7 14 papers · best R@1 98.7
- DIOR-R 14 papers · best mAP 80.7
- SIQA 14 papers · best accuracy 83.1
- ICFG-PEDES 14 papers · best Rank-1 70.05
- Composition-1K 14 papers · best SAD 18.63
- DomainBed average 14 papers · best accuracy 84.4
- PathVQA 14 papers · best accuracy 92.7
- PASCAL VOC 2012 train 14 papers · best mIoU 81.8
- IC15 14 papers · best accuracy 89.3
- 20News 14 papers · best accuracy 84.6
- MuirBench 14 papers · best accuracy 81.8
- CoNLL-2003 14 papers · best F1 95.4
- SPA-Data 14 papers · best PSNR 49.76
- CIFAR-100 (IID) 14 papers · best accuracy 73.56
- FF++ 14 papers · best AUC 99.96
- ADE20K val (single-scale) 14 papers · best mIoU 56.7
- IAM 14 papers · best CER 2.89
- ImageNet classification 14 papers · best accuracy 82
- CUB ZSL 14 papers · best accuracy 87.7
- UMIST 14 papers · best accuracy 97.39
- OGB-Arxiv 14 papers · best accuracy 76.31
- IIW 14 papers · best WHDR (%) 11.9
- NH-HAZE 14 papers · best PSNR 21.7
- ILSVRC12 14 papers · best accuracy (val) 78.2
- WFLW 14 papers · best NME 3.95
- CIFAR-100 (T=4) 14 papers · best accuracy 84.88
- HMDB 14 papers · best accuracy 90.84
- ImageNette IPC=10 14 papers · best accuracy 82
- UT-Zappos closed-world 14 papers · best AUC 52.5
- C-GQA closed-world 14 papers · best AUC 18.8
- CIFAR-10, VGG-16 14 papers · best accuracy 94.98
- miniImageNet 1-shot 5-way 14 papers · best accuracy 77.8
- Kinetics 5-way 1-shot 14 papers · best accuracy 95.6
- D4RL halfcheetah-medium 14 papers · best normalized score 85.4
- D4RL walker2d-medium 14 papers · best normalized score 111.6
- D4RL halfcheetah-medium-expert 14 papers · best normalized score 113.2
- D4RL hopper-medium-expert 14 papers · best normalized score 116.5
- Covtype 14 papers · best accuracy 98.53
- CUB 1-shot 14 papers · best accuracy 91.82
- PASCAL-Person-Part 14 papers · best mIoU 74.96
- YouTube-VIS 14 papers · best AP (val) 47.6
- Oxford5k 14 papers · best mAP 96.2
- Visual Genome SGDet 14 papers · best mR@100 20.3
- SYSU-MM01 all-search multi-shot 14 papers · best Rank-1 85.31
- Inference speed 14 papers · best FPS 165
- Video-Holmes 14 papers · best accuracy 60.5
- ImageNet-1K val (from scratch) 14 papers · best accuracy 85.7
- LLaVA-Bench (In-the-Wild) 13 papers · best score 89.8
- DynaMath 13 papers · best accuracy 83.4
- molhiv 13 papers · best ROC-AUC 81.12
- PHOENIX14 13 papers · best WER (%) (dev) 16.9
- TCGA-NSCLC 13 papers · best AUC 98
- ROxf Medium 13 papers · best mAP 91.4
- RPar Medium 13 papers · best mAP 93
- ROxf Hard 13 papers · best mAP 78.2
- GLUE average 13 papers · best average score 91.3
- Cora link prediction 13 papers · best AUC 99.1
- ImageNet-S 13 papers · best accuracy 77.41
- QuickDraw 13 papers · best accuracy 88.3
- DiDeMo text-to-video 13 papers · best R@1 56.9
- Tanks and Temples intermediate 13 papers · best Mean F-score 68.63
- COIL100 13 papers · best accuracy 99.89
- NTU RGB+D 120 X-Set 13 papers · best accuracy 92.2
- Domain generalization, ImageNet-R 13 papers · best accuracy 78.89
- ReVOS overall 13 papers · best J&F 68.4
- MS-COCO 5K 13 papers · best rSum 478.3
- InterHuman 13 papers · best R-Precision Top-1 0.529
- halfcheetah-medium-replay 13 papers · best normalized score 91.5
- Manga109 ×2 13 papers · best PSNR 40.34
- Sports 13 papers · best NDCG@20 0.0547
- CelebA-HQ 256×256 13 papers · best FID 3.15
- YoutubeFace 13 papers · best accuracy 96.17
- GSM8K (8-shot) 13 papers · best accuracy 93.48
- Blender 13 papers · best PSNR 35.02
- Vid4 13 papers · best PSNR 29.507
- CIFAR-100-LT IF=10 13 papers · best accuracy 76.13
- Dense-Haze 13 papers · best PSNR 17.6
- MUTAG graph classification 13 papers · best accuracy 98.94
- 11-dataset average, 16-shot 13 papers · best accuracy 87.6
- ImageNet-LT (All) 13 papers · best accuracy 59.9
- PoseTrack2017 13 papers · best mAP (val) 88
- Kinship 13 papers · best MRR 0.874
- SciTail 13 papers · best accuracy 96.8
- ImageNet, ResNet-34 13 papers · best accuracy 74.38
- SHD 13 papers · best accuracy 96.41
- Real20 13 papers · best PSNR 27.58
- ImageNet-64 13 papers · best FID 1.01
- LogiQA 13 papers · best accuracy 81
- C-GQA open-world 13 papers · best AUC 5.6
- MIT-States open-world 13 papers · best AUC 9.6
- Countdown 13 papers · best accuracy 90.5
- Deepglobe 1-shot 13 papers · best mIoU 51.7
- Deepglobe 5-shot 13 papers · best mIoU 63.4
- CIFAR-FS 5-shot 13 papers · best accuracy 94.55
- S3DIS 6-fold cross 13 papers · best mIoU (val) 78.4
- CIFAR-10 (T=4) 13 papers · best accuracy 97.81
- UT-Zappos open-world 13 papers · best AUC 43.7
- AIME 13 papers · best accuracy 90
- SNU-FILM Hard 13 papers · best PSNR 31.698
- CUB-200-2011 5-way 5-shot 13 papers · best accuracy 94.9
- NTU-60 X-Sub 13 papers · best accuracy 97
- VideoMME w/o subtitles 13 papers · best accuracy 77.7
- NTU RGB+D X-View 13 papers · best accuracy 98.4
- D4RL walker2d-medium-expert 13 papers · best normalized score 125.1
- COCO test-dev2017 13 papers · best AP 78.3
- PASCAL VOC 2012 trainval 13 papers · best CorLoc 73.7
- Kinetics-700 13 papers · best accuracy 91.6
- TGIF-QA 12 papers · best accuracy 79.5
- POPE MS-COCO Random 12 papers · best accuracy 90.72
- MME Cognition 12 papers · best score 548.9
- Camelyon16 12 papers · best AUC 0.987
- ogbn-papers100M 12 papers · best accuracy 76
- RPar Hard 12 papers · best mAP 84.1
- EgoSchema subset 12 papers · best accuracy 79.4
- EMNIST 12 papers · best accuracy 98.7
- AID 12 papers · best accuracy 97.4
- Perception 12 papers · best accuracy 70.2
- DepthTrack 12 papers · best F-score 67.9
- KITTI test car (moderate) 12 papers · best AP3D 82.09
- Domain generalization, ImageNet-A 12 papers · best accuracy 51.97
- Image 12 papers · best accuracy 89.36
- FEMNIST 12 papers · best accuracy 92.78
- GLUE MRPC 12 papers · best accuracy 93.68
- GLUE MNLI 12 papers · best accuracy 90.9
- OV-COCO 12 papers · best AP50 novel 52.4
- OmniObject3D 12 papers · best PSNR 31.929
- D-NeRF 12 papers · best PSNR 35.24
- COCO-30K 12 papers · best FID 2.06
- ActivityNet v1.3 12 papers · best average mAP (val) 38.6
- hopper-medium-expert 12 papers · best normalized score 113.7
- GoPro deblurring 12 papers · best PSNR 34.14
- SICE 12 papers · best SSIM 0.893
- Mfeat 12 papers · best accuracy 99.9
- O&M&I to C 12 papers · best HTER (%) 7.59
- O&C&M to I 12 papers · best HTER (%) 6.5
- O&C&I to M 12 papers · best HTER (%) 4.16
- I&C&M to O 12 papers · best HTER (%) 6.7
- CUB 1-shot 5-way 12 papers · best accuracy 82.2
- LoveDA 12 papers · best mIoU 57.17
- Potsdam 12 papers · best mIoU 87.71
- GenEval Overall 12 papers · best score 0.91
- ImageNet-1K 1% labels 12 papers · best accuracy 81.9
- NWPU-Crowd 12 papers · best MAE 57.8
- NICO 12 papers · best accuracy 98.33
- CIFAR-10 symmetric noise 0.8 12 papers · best accuracy 94.62
- BirdSong 12 papers · best accuracy 75.3
- HOTEL 12 papers · best ADE 0.11
- ogbn-proteins 12 papers · best ROC-AUC 89.04
- MSR-VTT text-to-video 12 papers · best R@1 55.6
- Manga109 12 papers · best PSNR 33.36
- WV3 reduced-resolution 12 papers · best SAM 2.69
- Cat->Dog 12 papers · best FID 26.24
- COIN T=3 12 papers · best success rate 33.99
- YouTube-VOS 12 papers · best PSNR 35.36
- Enwik8 12 papers · best BPC 0.898
- ETTh1 Avg 12 papers · best MSE 0.379
- CIFAR10-LT IF=50 12 papers · best accuracy 93.87
- Fashion 12 papers · best accuracy 99.74
- MetaShift 12 papers · best worst-group accuracy 94.8
- walker2d-medium-v2 12 papers · best normalized score 97.6
- halfcheetah-medium-expert-v2 12 papers · best normalized score 108.7
- HR-Bench 8K 12 papers · best accuracy 79.2
- V* Bench (overall) 12 papers · best accuracy 95.6
- CIFAR-100 WRN-40-2 -> WRN-40-1 12 papers · best accuracy 76.68
- Pitts250k 12 papers · best R@1 97.3
- MSLS-challenge 12 papers · best R@1 84
- BZR 12 papers · best accuracy 90.13
- Cityscapes semantic segmentation 12 papers · best mIoU 79.9
- RSTPReid 12 papers · best Rank-1 72.55
- Mip-NeRF 360 (outdoor) 12 papers · best PSNR 25.47
- Mip-NeRF 360 (indoor) 12 papers · best PSNR 32.09
- Office-31 D→W 12 papers · best accuracy 99.3
- DVD 12 papers · best PSNR 34.95
- SVHN (40 labels) 12 papers · best error rate 1.81
- ImageNet 512x512 12 papers · best FID 1.63
- HICO-DET Known Object Rare 12 papers · best mAP 45.18
- HICO-DET Known Object Non-Rare 12 papers · best mAP 51.38
- CUTE 12 papers · best accuracy 96.5
- Yeast 12 papers · best Average Precision 0.786
- CIFAR-10 vs SVHN 12 papers · best AUROC 99.9
- ISIC 1-shot 12 papers · best mIoU 72
- MSRC 12 papers · best accuracy 1
- Digits 12 papers · best accuracy 99.76
- SUN ZSL 12 papers · best accuracy 80.2
- Argoverse 2 12 papers · best mAP (val) 68.3
- GeoQA 12 papers · best accuracy 85.02
- CB 12 papers · best accuracy 96.43
- ShapeNet Airplane 12 papers · best COV-CD 64.98
- VOC21 12 papers · best mIoU 79.6
- HQF 12 papers · best SSIM 0.693
- Speech Commands 12 papers · best accuracy 98.4
- ImageNet top-1 12 papers · best accuracy 78.9
- REDDIT-MULTI-5K 12 papers · best accuracy 57.5
- LIBERO Avg. 12 papers · best success rate 98.8
- MIT-States closed-world 12 papers · best AUC 24.1
- MMVU 12 papers · best accuracy 73.1
- MedMCQA 12 papers · best accuracy 78.94
- Xiph-4K 12 papers · best PSNR 35.4
- SNU-FILM Extreme 12 papers · best PSNR 26.04
- MSRVTT-QA (zero-shot) 12 papers · best accuracy 67.5
- MSVD-QA (zero-shot) 12 papers · best accuracy 75.8
- CUB-200-2011 5-way 1-shot 12 papers · best accuracy 93.2
- NAVSIM 12 papers · best PDMS 94.89
- NTU-60 xview 12 papers · best accuracy 99.6
- NTU-120 X-Sub 12 papers · best accuracy 95.3
- NTU120-XSet 12 papers · best accuracy 96.4
- AWA1 12 papers · best H 89.1
- D4RL halfcheetah-medium-replay 12 papers · best normalized score 74.7
- D4RL walker2d-medium-replay 12 papers · best normalized score 106.3
- MIT Indoor 12 papers · best accuracy 97.33
- VOC07 12 papers · best mAP 92.8
- WMT14 EN-DE newstest2014 12 papers · best BLEU 32.56
- PASCAL VOC 2012 segmentation 12 papers · best mIoU 78.54
- DOTA 12 papers · best mAP 79.155
- TGIF-QA Action 12 papers · best accuracy 79.7
- TGIF-QA Transition 12 papers · best accuracy 97.6
- ImageNet val (ResNet-50) 12 papers · best accuracy 78.52
- VOT2017 12 papers · best EAO 0.38
- Urban100 x2 12 papers · best PSNR (dB) 34.37
- CIFAR-10 linear eval 12 papers · best accuracy 94
- ImgEdit-Bench 12 papers · best overall 4.31
- ETTh1 (avg over horizons) 12 papers · best MSE 0.375
- NAVSIM v1 12 papers · best PDMS 93.5
- NAVSIM v2 navtest 12 papers · best EPDMS 90.1
- AddSub 11 papers · best accuracy 96.71
- SingleEq 11 papers · best accuracy 99.01
- DOTA-v1.0 11 papers · best AP50 80.7
- Sleep-EDF 11 papers · best accuracy 88.01
- LibriSpeech 11 papers · best WER 1.72
- ADHD-200 11 papers · best accuracy 77.82
- STL-10 40 labels 11 papers · best error rate 9.32
- LSUN 11 papers · best FID 6.1
- Tanks-and-Temples advanced 11 papers · best Mean F-score 44.82
- NTU RGB+D 60 X-View 11 papers · best accuracy 99.7
- METR-LA 11 papers · best MAE 0.456
- BRCA 11 papers · best C-index 0.775
- Penn94 11 papers · best accuracy 85.94
- DeepGlobe 11 papers · best mIoU 80.22
- WebPhoto 11 papers · best FID 74.21
- Kinetics-600 zero-shot 11 papers · best accuracy 92.2
- NLVR2 11 papers · best accuracy (dev) 91
- Make3D 11 papers · best Abs Rel 0.254
- Pathfinder 11 papers · best accuracy 95.93
- kitchen-partial 11 papers · best normalized score 92.5
- DAVIS17 11 papers · best J&F (val) 88.8
- CARPK 11 papers · best MAE 2.79
- COCO-Stuff-10K 11 papers · best mIoU 44.88
- PACS Avg. 11 papers · best accuracy 93.7
- Walker2d-medium-expert 11 papers · best normalized score 117.2
- TAP-Vid RGB-Stacking 11 papers · best AJ 77.5
- Pascal Context-59 11 papers · best mIoU 73.7
- CIFAR-100, ResNet-20 11 papers · best accuracy 77.92
- PHOENIX-2014 11 papers · best WER 16.7
- B100 ×2 11 papers · best PSNR 32.62
- VideoLQ 11 papers · best MUSIQ 68.23
- Oxford Flowers 11 papers · best accuracy 99.64
- Chikusei 11 papers · best PSNR 47.814
- Sintel 11 papers · best Abs Rel 0.21
- Flower17 11 papers · best accuracy 98.6
- iSAID 11 papers · best mIoU 71.87
- Vaihingen 11 papers · best mIoU 91.74
- Diabetes 11 papers · best accuracy 77.4
- LSUN Church 256×256 11 papers · best FID 2.66
- T2I-CompBench Color 11 papers · best score 0.8677
- Products 11 papers · best accuracy 84.55
- HM3D 11 papers · best SPL 41.5
- CIFAR-10, BadNets 11 papers · best ASR 0.2
- CIFAR-100 symmetric noise 0.8 11 papers · best accuracy 71.92
- CIFAR-100 asymmetric noise 0.4 11 papers · best accuracy 76.31
- ETH 11 papers · best ADE 0.09
- Film 11 papers · best accuracy 41.84
- MOT15 11 papers · best MOTA 77.6
- MSR Action3D 11 papers · best accuracy 97.3
- MSR-Action3D (24 frames) 11 papers · best accuracy 97.21
- NuScenes Mean 11 papers · best Success 61.04
- SDD 11 papers · best FDE 10.13
- Shiny Blender 11 papers · best PSNR 39.24
- CIFAR-100, ResNet-50 → MobileNetV2 11 papers · best accuracy 72.57
- DomainNet average 11 papers · best accuracy 82.14
- Split CIFAR10 11 papers · best accuracy 91
- Split MiniImageNet 11 papers · best accuracy 65.9
- VisEvent 11 papers · best PR (%) 82.4
- tolokers 11 papers · best ROC AUC 86.08
- questions 11 papers · best ROC AUC 80.14
- ScanRefer overall 11 papers · best Acc@0.25 66.2
- WorldView-III 11 papers · best PSNR 31.0295
- A-150 11 papers · best mIoU 41.5
- CIFAR-10N Aggregate 11 papers · best accuracy 95.87
- halfcheetah-medium-replay-v2 11 papers · best normalized score 72.8
- hopper-medium-replay-v2 11 papers · best normalized score 104.6
- walker2d-medium-expert-v2 11 papers · best normalized score 116.1
- Credit 11 papers · best AUC 98.6
- SemanticKITTI hidden 11 papers · best mIoU 19.2
- CIFAR-10 32×32 11 papers · best FID 1.96
- Matterport3D 11 papers · best RMSE 0.3305
- iNaturalist 11 papers · best accuracy 75.3
- HICO-DET test, Default Full 11 papers · best mAP 38.4
- CIFAR-100 (ResNet-110 -> ResNet-32) 11 papers · best accuracy 75.33
- OGBN-MAG 11 papers · best accuracy 55.54
- WV3 full-resolution 11 papers · best HQNR 0.959
- LEVIR-CD 11 papers · best F1 94.83
- BBC 11 papers · best accuracy 95.77
- KITTI Eigen 11 papers · best Abs Rel 0.015
- TUM-dynamics camera pose 11 papers · best ATE 0.01
- sCIFAR10 11 papers · best accuracy 93.08
- VSPW 11 papers · best mIoU (val) 64.9
- Mutagenicity 11 papers · best accuracy 83.8
- SBU 11 papers · best BER 2.54
- CULane 11 papers · best F1 81.83
- Moving MNIST 11 papers · best MSE 14.9
- MELD 11 papers · best accuracy 71.5
- MSL 11 papers · best F1 95.7
- ABIDE 11 papers · best accuracy 84.67
- AdvBench 11 papers · best ASR 100
- CIFAR-100 (fine-tuning) 11 papers · best accuracy 91.9
- K400 11 papers · best accuracy 87.7
- MalNet-Tiny 11 papers · best accuracy 94.6
- LIBERO-10 11 papers · best success rate 95.8
- COCO Mask R-CNN 1x 11 papers · best box AP 52
- Chest X-ray 1-shot 11 papers · best mIoU 87
- ISIC 5-shot 11 papers · best mIoU 79.8
- DPDD 11 papers · best PSNR 26.81
- ImageNet 128x128 11 papers · best FID 1.48
- COCO Caption Karpathy 11 papers · best CIDEr 145.6
- 3Sources 11 papers · best accuracy 94.67
- CROHME 2014 11 papers · best ExpRate 82.05
- iNaturalist 2018 (All) 11 papers · best accuracy 76.2
- ScreenSpot-v2 11 papers · best accuracy 95.1
- ScreenSpot-Pro 11 papers · best accuracy 74
- Pendigits 11 papers · best accuracy 0.9908
- AIME24 (Avg@32) 11 papers · best accuracy 79.9
- N3DV 11 papers · best PSNR (dB) 33.51
- MMSI-Bench 11 papers · best accuracy 43.3
- Rain1400 11 papers · best PSNR 34.84
- 20NG 11 papers · best accuracy 99.8
- PMC-VQA 11 papers · best accuracy 74.1
- Letter 11 papers · best accuracy 98.43
- FUNSD 11 papers · best F1 92.86
- Moments in Time 11 papers · best accuracy 53.1
- CUHK Avenue 11 papers · best AUC 90.3
- Birds 11 papers · best accuracy 91.74
- nocaps validation overall 11 papers · best CIDEr 120.5
- SearchQA 11 papers · best F1 84.4
- Market1501→MSMT17 11 papers · best mAP 32.8
- Set14 x4 11 papers · best PSNR (dB) 29.06
- CelebA-64 11 papers · best FID 2.71
- AR 11 papers · best accuracy 98.7
- MovieLens 1M 11 papers · best RMSE 0.823
- VOC 2007 trainval 11 papers · best CorLoc 80.8
- Human3.6M Protocol 2 11 papers · best MPJPE (mm) 32.1
- CUHK01 11 papers · best Rank-1 90.4
- 300W Challenging 11 papers · best NME 4.21
- Cosal2015 11 papers · best MAE 0.0454
- CIFAR-10, 48 bits 11 papers · best mAP 0.9456
- Paris106k 11 papers · best mAP 96.2
- TGIF-QA Count 11 papers · best MSE 3.41
- CIFAR-100-LT (imbalance 100) 11 papers · best accuracy 83.3
- ModelNet40 few-shot 5-way 10-shot 11 papers · best accuracy 98
- LIBERO average (4 suites) 11 papers · best success rate 99.1
- NAVSIM v2 11 papers · best EPDMS 90.6
- Cityscapes val (single-scale) 11 papers · best mIoU 84.09
- NTU-60 x-sub (linear eval) 11 papers · best accuracy 87.9
- NTU-120 x-sub (linear eval) 11 papers · best accuracy 81.5
- HPatches homography 11 papers · best AUC@3px 73.9
- MathVerse (Vision Only) 10 papers · best accuracy 67
- RESISC45 10 papers · best accuracy 97.48
- Cancer 10 papers · best accuracy 97.9
- Cora-CA 10 papers · best accuracy 86.54
- ProofWriter 10 papers · best accuracy 97
- WikiCS node classification 10 papers · best accuracy 85.52
- CIFAR-100, 10 tasks 10 papers · best average accuracy 94.5
- SAMSum 10 papers · best ROUGE-L 52.7
- Blender (8 views) 10 papers · best PSNR 26.21
- ASL-DVS 10 papers · best accuracy 99.91
- PEMS-BAY 10 papers · best MAE 0.224
- BSD100 10 papers · best PSNR 27.41
- AQUA-RAT 10 papers · best accuracy 86.2
- NTU RGB+D 60 X-Sub 10 papers · best accuracy 97.9
- AudioSet 10 papers · best mAP 49.6
- SYSU-MM01 10 papers · best mAP 83.6
- RegDB 10 papers · best mAP 95.8
- GTA5 → BDD100K 10 papers · best mIoU 62.82
- GTA5 → Mapillary 10 papers · best mIoU 70.1
- QM8 10 papers · best MAE 0.009
- MathVista mini 10 papers · best accuracy 82.1
- Weibo 10 papers · best accuracy 0.9663
- MMLU-STEM 10 papers · best accuracy 87
- WIDER 10 papers · best FID 31.49
- YCB-V 10 papers · best AUC of ADD(-S) 95
- Airport 10 papers · best accuracy 98.7
- NarrativeQA 10 papers · best F1 65
- Shapes3D 10 papers · best DCI 0.973
- MMLU zero-shot 10 papers · best accuracy 82.4
- Sr3D 10 papers · best accuracy 91.4
- CIFAR100-10 10 papers · best accuracy 78.35
- RoadScene 10 papers · best VIF 0.969
- halfcheetah-medium-expert 10 papers · best normalized score 97.9
- ImageNet zero-shot classification 10 papers · best accuracy 82.2
- Phoenix-2014T 10 papers · best WER 18.5
- DFEW 10 papers · best WAR 78.29
- ClevrTex 10 papers · best FG-ARI 94.66
- FFHQ-1024 10 papers · best FID 3.01
- arXiv-year 10 papers · best accuracy 66.16
- DICM 10 papers · best NIQE 2.49
- AD 10 papers · best accuracy 92.7
- Harvard 10 papers · best PSNR 49.28
- CIFAR-100 symmetric noise 20% 10 papers · best accuracy 83.1
- CUB 5-shot 5-way 10 papers · best accuracy 90.85
- NGs 10 papers · best accuracy 99.6
- QM9 mu 10 papers · best MAE 0.13
- CV-Bench-3D 10 papers · best accuracy 97.58
- VITON-HD unpaired 10 papers · best FID 7.89
- Retrieval 10 papers · best accuracy 91.44
- Yahoo!News 10 papers · best accuracy 0.833
- UNIV 10 papers · best ADE 0.24
- ZARA1 10 papers · best ADE 0.15
- ZARA2 10 papers · best ADE 0.1
- VATEX text-to-video 10 papers · best R@1 68.4
- MegaDepth-1500 10 papers · best AUC@5 70.1
- KITTI Car 10 papers · best Success 73.6
- HellaSwag (zero-shot) 10 papers · best accuracy 86.7
- PIQA (zero-shot) 10 papers · best accuracy 81.8
- Test100 10 papers · best PSNR 32.064
- CrossTask T=3 10 papers · best success rate 40.45
- NIV T=3 10 papers · best success rate 32.37
- CIFAR-10 IDN 20% 10 papers · best accuracy 96.98
- HICO-DET (Rare) 10 papers · best mAP 50.82
- Lipo 10 papers · best RMSE 0.454
- EuroSAT, 5-way 5-shot 10 papers · best accuracy 93.41
- AMBER 10 papers · best CHAIR 2.1
- PAS-20 10 papers · best mIoU 98.5
- Weather Avg 10 papers · best MSE 0.227
- MUSIC-AVQA 10 papers · best accuracy 85.01
- AudioCaps 10 papers · best FAD 1.03
- VisDA-C 10 papers · best accuracy 90
- Cityscapes 1/16 (186) 10 papers · best mIoU 81.4
- IRSTD-1k 10 papers · best IoU (%) 73.69
- PoseTrack2018 10 papers · best mAP (val) 85.4
- Tiny-ImageNet, α=0.1 10 papers · best accuracy 82.31
- Traffic (avg) 10 papers · best MSE 0.386
- VTAB-1k Natural 10 papers · best accuracy 87.18
- VTAB-1k Specialized 10 papers · best accuracy 90.25
- VTAB-1k Structured 10 papers · best accuracy 69.32
- PG19 10 papers · best perplexity 8.6499
- CIFAR-100N Noisy 10 papers · best accuracy 74.67
- ImageNet-O 10 papers · best AUPR 54.61
- hopper-medium-expert-v2 10 papers · best normalized score 113.5
- 10-Split CIFAR-100 10 papers · best accuracy 81.66
- NVOS 10 papers · best mIoU 94.2
- HICO-DET test, Default Rare 10 papers · best mAP 37.52
- IJB-C 10 papers · best TAR@FAR=1e-4 98.42
- CIFAR-100 VGG13 -> MobileNetV2 10 papers · best accuracy 72.26
- CIFAR-100, WRN-40-2 → ShuffleNetV1 10 papers · best accuracy 78.62
- MPII 10 papers · best PCKh@0.5 (val) 92.5
- UCI-Digit 10 papers · best accuracy 99.95
- S3DIS Area 5 semantic segmentation 10 papers · best mIoU 76
- VidSTG declarative 10 papers · best m_vIoU 35.09
- Sintel camera pose 10 papers · best ATE 0.05
- MP3D 10 papers · best SPL (%) (val) 29.6
- Office-31 W→D 10 papers · best accuracy 100
- LS-VID 10 papers · best mAP 93.4
- CIFAR-10 IDN-50% 10 papers · best accuracy 95.88
- Kodak 10 papers · best PSNR 90.84
- ImageNet-1K 224×224 10 papers · best accuracy 86.2
- TVQA 10 papers · best accuracy 89.1
- AFHQ-Cat 10 papers · best FID 2.4
- CIFAR-10, symmetric 20% noise 10 papers · best accuracy 96.9
- CIFAR-100, symmetric 20% noise 10 papers · best accuracy 81.9
- SMAP 10 papers · best F1 96.72
- SPair-71k 10 papers · best PCK@0.1 88.9
- JHMDB 10 papers · best PCK@0.1 68.3
- AffNIST 10 papers · best accuracy 98.3
- ACE2005 10 papers · best F1 90
- CMNIST 10 papers · best worst-group accuracy 84.98
- LLFF 9-view 10 papers · best PSNR 26.731
- German 10 papers · best accuracy 0.77
- LIBERO-90 10 papers · best success rate 97.7
- LOL-v2-syn 10 papers · best PSNR 32.06
- KITTI-15 10 papers · best Fl-all 2.85
- PascalVOC-SP 10 papers · best F1 0.4912
- mirflickr 10 papers · best average precision 0.835
- CelebA-HQ 256x256 10 papers · best FID 3.39
- GAIA 10 papers · best accuracy 72.8
- Ionosphere 10 papers · best accuracy 98.6
- CIFAR-100 (ResNet32x4 -> ShuffleNetV2) 10 papers · best accuracy 79.83
- REDDIT-5K 10 papers · best accuracy 59.5
- PCQM4Mv2 10 papers · best MAE (val) 0.0734
- RealLQ250 10 papers · best MANIQA 0.7029
- LVIS 10 papers · best AP 42.3
- Food101 zero-shot 10 papers · best accuracy 96.2
- Duke→Market 10 papers · best mAP 83.8
- POPE MSCOCO Adversarial 10 papers · best accuracy 84.32
- CIFAR10, NFE 10 10 papers · best FID 1.97
- LVIS minival zero-shot 10 papers · best AP 62.5
- mini WebVision 10 papers · best accuracy 83.1
- COCO captioning 10 papers · best CIDEr 149
- WISE 10 papers · best overall 0.76
- KITTI 2012 10 papers · best EPE 0.3657
- Sintel Clean train 10 papers · best EPE 1.12
- Glass 10 papers · best accuracy 77.7
- MNIST → MNIST-M 10 papers · best accuracy 99.4
- GEOM-Drugs 10 papers · best validity (%) 99.7
- Flickr30k text-to-image 10 papers · best R@1 97.8
- LSUN-Church 256 10 papers · best FID 1.59
- Scan2Cap 10 papers · best CIDEr@0.5 87.2
- Pokec-n 10 papers · best accuracy 70.12
- CIFAR-10, ResNet-32 10 papers · best accuracy 93.5
- R2R 10 papers · best SPL 62.2
- Chest X-Ray 5-shot 10 papers · best mIoU 88.91
- CORD 10 papers · best F1 98.75
- Breakout 10 papers · best score 532.8
- OCRVQA 10 papers · best accuracy 73.2
- FMD 10 papers · best accuracy 98.2
- VQA v2.0 10 papers · best accuracy (test-dev) 86.5
- Birdsnap 10 papers · best accuracy 85.4
- 3DLoMatch 10 papers · best Registration Recall (%) 81.7
- HR-Bench 8K Overall 10 papers · best accuracy 79.5
- COCO val2017 (RetinaNet) 10 papers · best AP 47.2
- TUM RGB-D fr1/desk 10 papers · best ATE RMSE (cm) 1.3
- SYNTHIA to Cityscapes 10 papers · best mIoU 72.8
- VOT-2015 10 papers · best accuracy 0.63
- YCB-Video 10 papers · best ADD(-S) AUC 96.5
- NLVR² test-P 10 papers · best accuracy 85.52
- MSRA-B 10 papers · best MAE 0.029
- AgeDB-30 10 papers · best verification accuracy (%) 98.467
- Netflix 10 papers · best RMSE 0.7973
- CoSOD3k 10 papers · best S-measure 0.874
- CoCA 10 papers · best MAE 0.077
- Kin8nm 10 papers · best test log-likelihood 2.49
- Oxford105k 10 papers · best mAP 95.2
- MSCOCO 1K 10 papers · best image retrieval R@1 69.1
- SVHN, 500 labels 10 papers · best error rate 1.85
- AFLW-Full 10 papers · best NME 1.27
- AWA1 GZSL 10 papers · best harmonic mean H 72.6
- CIFAR-10 classification 10 papers · best accuracy 95.47
- VOT2019 10 papers · best EAO 0.397
- ImageNet semi-supervised (1% labels) 10 papers · best accuracy 59.9
- In-Shop Clothes Retrieval 10 papers · best R@1 92.1
- CIFAR-100 linear eval 10 papers · best accuracy 72.4
- Weather (avg over horizons) 10 papers · best MSE 0.186
- VSI-Bench Rel. Dir. 10 papers · best accuracy 89.5
- WorldSense 10 papers · best accuracy 65.8
- DailyOmni 10 papers · best accuracy 82.8
- Efficiency 10 papers · best parameters (M) 9.5
- NTU-60 x-view (linear eval) 10 papers · best accuracy 93.2
