HLE
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
HLE text only
18.4Score
4
HLE text
49.5Score
3
HLE
1.275Perplexity
3
HLE-100
54.7Accuracy
3
HLE All Non-Easy
9.3Recovery Rate
3
HLE Med. + Hard
13.1Recovery Rate
3
HLE
11.2Accuracy
2
HLE
41.1Score
2
HLE
22.6HLE Average
2
HLE (reference)
185Problems Evaluated
1