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

Factuality Evaluation on LLM-AggreFact (test)

69.9CNN Score

MiniCheck-FT5

48.5854.11559.6565.185Oct 24, 2024
Updated 15d ago

Evaluation Results

MethodLinks
2024.10
69.974.373.677.372.286.274.674.75985.274.7-
2024.10
66.776.571.479.980.487.867.679.959.283.175.3-
2024.10
65.272.474.182.47583.869.378.858.881.674.1-
2024.10
63.371.367.975.274.886.767.378.358.482.972.6-
2024.10
63.273.466.677.764.686.472.57357.983.171.8-
2024.10
63.272.466.873.468.584.765.270.857.273.869.6-
2024.10
62.172.470.980.37688.674.974.460.386.974.7-
2024.10
59.966.469.272.364.388.269.779.359.878.270.7-
2024.10
59.574.768.48071.78874.374.259.68773.7-
2024.10
58.476.367.378.976.688.467.6796081.773.4-
2024.10
5565.568.573.363.880.864.375.156.370.867.3-
2024.10
50.958.263.972.465.185.263.876.555.972.366.4-
2024.10
49.460.663.865.865.885.561.876.856.875.966.2-
2026.04
-----------77.4
2026.04
-----------57.5
2026.04
-----------78.1
2026.04
-----------79.2