Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

generation

Benchmarks

Task NameDataset NameSOTA ResultTrend
Natural language generation (Table-to-text, summarization)Generation OOD
Score (Full Output)34.1
13
Generation w/ citationsGeneration w/ citations
Citation Quality (8k Context)34.8
13
Inference Speed Evaluationgeneration 128 tokens
Inference Time (ms)2,747.88
8
Joint-level motion trackingGeneration (test)
MPJPE0.5671
3
Showing 4 of 4 rows