Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

General

Benchmarks

Task NameDataset NameSOTA ResultTrend
Machine Translationgeneral 2023 (test)
BLEU32.64
16
Image Manipulation DetectionGeneral Inference Speed Evaluation Images
FPS31.7
16
Instance ErasureGeneral
FID (General)13.24
13
StabilityGeneral (MMLU, BBH, TyDiQA, BoolQ, PIQA, GSM8K)
General Score55.75
9
Video CompressionGeneral
Parameters (M)18.34
9
SegmentationGeneral Efficiency Evaluation
Latency (ms)7.3
9
Underwater Image EnhancementGeneral Architectural Comparison 1.0 (UEIB-T90)
PSNR22.82
8
General Vision-Language UnderstandingGeneral
Avg Score72.4
8
Average evaluation across 7 tasksGeneral (test)
BERTScore76.5
8
Colon Polyp SegmentationGeneral
Parameters (M)32.55
8
360-degree video saliency predictionGeneral
Params (M)3.7
7
Model Efficiency AnalysisGeneral 16 frames, 512 text tokens (inference)
FPS20.74
6
Interactive SegmentationGeneral Efficiency Benchmarking
Parameters (MB)84.89
6
Novel View SynthesisGeneral
MFLOPs / Pixel13.77
5
Ending event predictionGeneral (test)
MRR0.401
5
Vulnerability AnalysisGeneral
Metric-
0
3D Scene Decomposition Capability AssessmentGeneral Method Capability Comparison
Metric-
0
Conditional Layout GenerationGeneral Literature Comparison
Metric-
0
Showing 18 of 18 rows