mixed
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Mixed (test)
95.3Accuracy
50
Mixed
0.59NDCG@10
44
Mixed (AIME, GPQA, HLE, HotpotQA, ALFWorld)
57.74Average Score
24
MIXED (test)
95Task 1 Accuracy
13
MIXED #3000 English to Luxembourgish (test)
0.26CometScore
13
Mixed
45.55F1 Score
12
Mixed In Domain
96.44Accuracy
11
Mixed In Domain
96.16Accuracy
11
Mixed In-Domain
0.031EER
11
Mixed In Domain
91.33Seg-F1
9
Mixed Overall Summary
67.28Overall Success Rate
8
Mixed 20-step f.t. (held-out)
98Resistance
7
Mixed dataset
98.3Accuracy
6
Mixed-1M
0.749Correlation Coefficient (CC)
5
mixed long
364.3OOM Count
5
Mixed
2Coverage Loss
5
Mixed
6.89Time (ms)
5
Mixed BERT-base (test)
83.3Accuracy
5
Mixed
5Recovery Score (Grader 1)
4
mixed SDRW LDRS NYU (test)
71.4F1 Score
2
Mixed (train)
95.82Accuracy
1
Mixed (val)
95.2Accuracy
1
Mixed (test)
95.2Accuracy
1