Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Task-Oriented Multi-Agent Communication on MMLU, ARC-C, GSM8K, OpenBookQA, MedQA Average (test)

79.03Accuracy

TodyComm

49.07856.85464.6372.406Feb 3, 2026
Updated 4d ago

Evaluation Results

MethodLinks
2026.02
79.03409.2
2026.02
77.45422.2
2026.02
77.04383
2026.02
75.17607.8
2026.02
73.39430.2
2026.02
70.92758.8
2026.02
68.95606.2
2026.02
60.52444
2026.02
59.05643.4
2026.02
55.78458.2
2026.02
55.38635.4
2026.02
55.19773
2026.02
54.07636.2
2026.02
50.7625
2026.02
50.23756.6