List your company

Class-level Continual Learning on MMLU-PRO

50.72Average Accuracy (AA)MasFACT
25.18831.816538.44545.0735May 17, 2026

Evaluation results

MethodLinks
MasFACTSetting=Standard, Replay-based=false2026.0550.720.13
MasFACTEvaluation Style=Standard2026.0550.720.13
MasFACTSetting=Few-shot, Replay-based=false2026.0549.18-0.19
MasFACTEvaluation Style=Few-shot2026.0549.18-0.19
ARGDesignerSetting=Standard, Replay-based=true2026.0548.691.33
ARGDesigner†Evaluation Style=Standard, Replay=true2026.0548.691.33
GDesignerSetting=Standard, Replay-based=true2026.0547.871.22
GDesigner†Evaluation Style=Standard, Replay=true2026.0547.871.22
MasRouterSetting=Standard, Replay-based=true2026.0547.530.83
MasRouter†Evaluation Style=Standard, Replay=true2026.0547.530.83
MasRouterSetting=Few-shot, Replay-based=true2026.0547.41.43
MasRouter†Evaluation Style=Few-shot, Replay=true2026.0547.41.43
ARGDesignerSetting=Few-shot, Replay-based=true2026.0546.871.27
ARGDesigner†Evaluation Style=Few-shot, Replay=true2026.0546.871.27
GDesignerSetting=Standard, Replay-based=false2026.0546.723.61
GDesignerEvaluation Style=Standard2026.0546.723.61
MasRouterSetting=Standard, Replay-based=false2026.0546.172.81
MasRouterEvaluation Style=Standard2026.0546.172.81
GDesignerSetting=Few-shot, Replay-based=true2026.0546.161.49
GDesigner†Evaluation Style=Few-shot, Replay=true2026.0546.161.49
MasRouterSetting=Few-shot, Replay-based=false2026.0545.942.27
MasRouterEvaluation Style=Few-shot2026.0545.942.27
AgentDropoutSetting=Standard, Replay-based=true2026.0545.841.49
AgentDropout†Evaluation Style=Standard, Replay=true2026.0545.841.49
ARGDesignerSetting=Standard, Replay-based=false2026.0545.523.39
ARGDesignerEvaluation Style=Standard2026.0545.523.39
GDesignerSetting=Few-shot, Replay-based=false2026.0545.043.28
GDesignerEvaluation Style=Few-shot2026.0545.043.28
ARGDesignerSetting=Few-shot, Replay-based=false2026.0544.723.25
ARGDesignerEvaluation Style=Few-shot2026.0544.723.25
AgentDropoutSetting=Few-shot, Replay-based=true2026.0543.031.72
AgentDropout†Evaluation Style=Few-shot, Replay=true2026.0543.031.72
AFLOWSetting=Standard, Replay-based=true2026.0542.892.19
AFLOW†Evaluation Style=Standard, Replay=true2026.0542.892.19
AFLOWSetting=Few-shot, Replay-based=true2026.0542.521.21
AFLOW†Evaluation Style=Few-shot, Replay=true2026.0542.521.21
AgentDropoutSetting=Standard, Replay-based=false2026.0542.483.83
AgentDropoutEvaluation Style=Standard2026.0542.483.83
AFLOWSetting=Standard, Replay-based=false2026.0542.374.32
AFLOWEvaluation Style=Standard2026.0542.374.32
AgentDropoutSetting=Few-shot, Replay-based=false2026.0541.073.13
AgentDropoutEvaluation Style=Few-shot2026.0541.073.13
SC(CoT)-ICLSetting=Few-shot, Replay-based=false2026.0540.524.95
SC(CoT)-ICLEvaluation Style=Few-shot2026.0540.524.95
AFLOWSetting=Few-shot, Replay-based=false2026.0540.123.97
AFLOWEvaluation Style=Few-shot2026.0540.123.97
CoT-ICLSetting=Few-shot, Replay-based=false2026.0539.64.01
CoT-ICLEvaluation Style=Few-shot2026.0539.64.01
CompleteSetting=Few-shot, Replay-based=false2026.0538.64—
CompleteEvaluation Style=Few-shot2026.0538.64—
TreeSetting=Few-shot, Replay-based=false2026.0536.82—
TreeEvaluation Style=Few-shot2026.0536.82—
ChainSetting=Few-shot, Replay-based=false2026.0536.48—
ChainEvaluation Style=Few-shot2026.0536.48—
VanillaSetting=Few-shot, Replay-based=false2026.0526.17—
VanillaEvaluation Style=Few-shot2026.0526.17—