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

Multi-agent Question Answering on SocialIQA (first 300 questions)

82.56Average Accuracy

HyLaT

41.188851.929462.6773.4106May 25, 2026
Updated 8d ago

Evaluation Results

MethodLinks
2026.05
82.568372.011.47
2026.05
80.7880.67570.7
2026.05
76.6677.63505.035.47
2026.05
59.89631,717.618.18
2026.05
58.22611,247.513.15
2026.05
58.1160.671,316.214.48
2026.05
57.8960.331,137.8911.58
2026.05
5656113.591.25
2026.05
55.7861.33960.2610.04
2026.05
42.7843324.433.61