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

Time-controlled instruction-following on UROBench long (30s–1min)

6.13MAE

TiCo

4.972812.783920.59528.4061Mar 23, 2026
Updated 4mo ago

Evaluation Results

MethodLinks
2026.03
6.1313.4
2026.03
17.4141.1
2026.03
18.3739.9
2026.03
19.2740.4
2026.03
19.4441.7
2026.03
29.7264.2
2026.03
32.2874.1
2026.03
35.0675.6