Benchmarks

Compare state-of-the-art methods across the tasks and datasets used to measure AI progress.

309,131 indexed

Task	Dataset	Domain	Results	Last Update
Sycophancy Mitigation	Held-out	llm chat instructionadversarial safety security	40	Jun 5, 2026
Sycophancy Mitigation	MMLU on-the-fly	llm chat instructionadversarial safety security	40	Jun 5, 2026
Deepfake detection	DeepfakeTIMIT Dataset2 (K-Fold Cross Validation)	computer visionadversarial safety security	40	Jun 5, 2026
Image Deblurring	Standard Images Average blur kernel (test)	computer vision	40	Jun 5, 2026
Image Deblurring	Standard Test Images Gaussian blur kernel	computer vision