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

Binary ill-posedness detection on SituatedQA

90.4Accuracy

H0 topology

45.6857.2968.980.51Jun 22, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.06
90.490.4
2026.06
88.488.4
2026.06
86.686.6
2026.06
85.585.5
2026.06
80.280.2
2026.06
73.973.9
2026.06
61.659.5
2026.06
57.757.6
2026.06
57.154.6
2026.06
56.456.3
2026.06
55.246.2
2026.06
54.936.3
2026.06
54.443.1
2026.06
54.346.1
2026.06
52.952.7
2026.06
51.542.5
2026.06
48.647.9
2026.06
47.432.6