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

Long-form generation on Long-form generation datasets LOO - near OOD

0.3PRR

HUQ-SATRMD

-0.1472-0.03110.0850.2011Apr 13, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.04
0.3
2026.04
0.2
2026.04
0.12
2026.04
0.06
2026.04
0.04
2026.04
0.03
2026.04
0
2026.04
0
2026.04
-0.01
2026.04
-0.01
2026.04
-0.01
2026.04
-0.01
2026.04
-0.01
2026.04
-0.02
2026.04
-0.04
2026.04
-0.05
2026.04
-0.05
2026.04
-0.05
2026.04
-0.06
2026.04
-0.06
2026.04
-0.09
2026.04
-0.09
2026.04
-0.1
2026.04
-0.13