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

Writing flaw detection on BenchMarker Writing - In Domain, NLP

81.5Accuracy

Gemini 2.5

23.36438.45753.5568.643Feb 5, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.02
81.565.853.1
2026.02
80.863.150.1
2026.02
79.462.548.3
2026.02
78.861.947.3
2026.02
77.440.428.5
2026.02
76.753.237.6
2026.02
7624.916.4
2026.02
75.952.536.4
2026.02
7555.338.1
2026.02
74.742.6
2026.02
74.72.92
74.400
2026.02
73.538.622.4
2026.02
73.44.60.4
2026.02
73.449.431.5
2026.02
71.857.638.1
2026.02
71.237.519.1
2026.02
70.655.334.9
2026.02
70.554.534
2026.02
69.854.333.4
2026.02
68.356.334.6
2026.02
68.336.615.5
2026.02
63.946.521.6
2026.02
63.145.620
5033.90
2026.02
46.637.52.5
25.640.80