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

RAG Robustness on NQ

57.8Robustness RACC (Paradox)

AV Filter

28.05635.77843.551.222Jun 4, 2025
Updated 9d ago

Evaluation Results

MethodLinks
2025.06
57.83.260.8059.83.4
2025.06
56.4456.2251.28
2025.06
54.66.6582575.4
2025.06
546.8572.456.86
2025.06
527.654.83.8519.2
2025.06
51.81.654.60.246.222.4
2025.06
50.67.453.24.448.49.2
2025.06
48.410.652.24.445.618
2025.06
46.210.6524.642.819.8
2025.06
37.422.459.42.856.82.4
2025.06
3228.659.64.844.230.6
2025.06
29.224.256.8540.415.8