Average
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Average (BCI-IV-2a, PhysionetMI, SEED-IV)
18Avg. Delta
6
Average all datasets
-82.76BD-Rate (%)
6
Average Strawberry and PFuji-Size
96.8IoU
6
Average across benchmarks
62Cluster Accuracy (CAcc)
6
Average All Datasets Harmonic Mean
44.1Accuracy
6
Average All Datasets, Low-End Target
39.3Accuracy
6
Average All Datasets, High-End Source
54.8Accuracy
6
Average
59.5Win Rate
6
Average over 5 datasets (test)
83.18Base Accuracy
6
Average (AVG)
85.8Throughput
6
Average MATH500, MBPP, Text-to-SQL, QuaLITY
61.6P@1
6
Average (Scannet, DL3DV, DTU)
21.078PSNR
6
Average (test)
52.45Hit Score
6
Average (BDD, MV, ACDC, DZ) (val)
55.11Average mIoU
6
Average 494 images (test)
24.08PSNR
6
Average 11 datasets
72.67Avg Top-1 Clean Acc
5
Average ScanNet++, TUM RGB-D, KITTI
0.49AURC
5
Average
78.03AJ
5
Average Math500 AIME24 AIME25 GPQA
41.72Pass@1
5
Average All 7 Datasets
65.13Average Score
5
Average (IEMOCAP, MELD, RAVDESS, CASIA, CAFE, RESD)
70.1Accuracy
5
Average of 5 datasets
79.7Synthetic Score
5
Average (MuSiQue, 2Wiki, MoreHopQA, FRAMES)
0.791NDCG
5
Average Hybrid-weather
22.25PSNR
5
Average
96Accuracy
5