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

Task Representation Accuracy on Task Vector Evaluation Suite Llama2-13B (test)

87.69Accuracy

LTV

-2.238821.108144.45567.8019Sep 29, 2025
Updated 6d ago

Evaluation Results

MethodLinks
2025.09
87.69
2025.09
84.99
2025.09
82.25
2025.09
80.33
2025.09
77.51
2025.09
71.53
2025.09
51.46
2025.09
43.84
2025.09
42.25
2025.09
41.59
2025.09
36.97
2025.09
27.67
2025.09
24.74
2025.09
20.46
2025.09
16.42
2025.09
16.07
2025.09
1.84
2025.09
1.22