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

Generative Hallucination Evaluation on AMBER (test)

7.9CHAIR Score

SID

3.6364.7435.856.957May 27, 2026
Updated 6d ago

Evaluation Results

MethodLinks
2026.05
7.964.739.12.1
2026.05
7.863.940.42
2026.05
6.864.436.82.4
2026.05
6.663.735.62.1
2026.05
6.662.932.81.8
2026.05
5.157.521.51.6
2026.05
4.861.524.31.4
2026.05
4.660.723.71.1
2026.05
4.658.120.21
2026.05
4.661.525.21.8
2026.05
4.459.520.81.2
2026.05
4.457.819.81
2026.05
4.457.921.61.5
2026.05
4.156181.1
2026.05
459.321.21.2
2026.05
458.318.31
2026.05
3.956.216.21
2026.05
3.856.815.21