Class-level Continual Learning on MATH
61.68Average Accuracy (AA)MasFACT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MasFACTSetting=Standard, Replay-based=false2026.05 | 61.68 | 0.21 | |
| MasFACTEvaluation Style=Standard2026.05 | 61.68 | 0.21 | |
| MasRouterSetting=Standard, Replay-based=true2026.05 | 58.56 | 1.47 | |
| MasRouter†Evaluation Style=Standard, Replay=true2026.05 | 58.56 | 1.47 | |
| ARGDesignerSetting=Standard, Replay-based=true2026.05 | 57.83 | 1.59 | |
| ARGDesigner†Evaluation Style=Standard, Replay=true2026.05 | 57.83 | 1.59 | |
| ARGDesignerSetting=Standard, Replay-based=false2026.05 | 56.33 | 3.86 | |
| ARGDesignerEvaluation Style=Standard2026.05 | 56.33 | 3.86 | |
| GDesignerSetting=Standard, Replay-based=true2026.05 | 56.18 | 1.94 | |
| GDesigner†Evaluation Style=Standard, Replay=true2026.05 | 56.18 | 1.94 | |
| MasFACTSetting=Few-shot, Replay-based=false2026.05 | 56.07 | -0.45 | |
| MasFACTEvaluation Style=Few-shot2026.05 | 56.07 | -0.45 | |
| MasRouterSetting=Few-shot, Replay-based=true2026.05 | 55.17 | 2.76 | |
| MasRouter†Evaluation Style=Few-shot, Replay=true2026.05 | 55.17 | 2.76 | |
| GDesignerSetting=Few-shot, Replay-based=true2026.05 | 54.7 | 2.36 | |
| GDesigner†Evaluation Style=Few-shot, Replay=true2026.05 | 54.7 | 2.36 | |
| MasRouterSetting=Standard, Replay-based=false2026.05 | 54.28 | 4.46 | |
| MasRouterEvaluation Style=Standard2026.05 | 54.28 | 4.46 | |
| ARGDesignerSetting=Few-shot, Replay-based=true2026.05 | 54.21 | 1.88 | |
| ARGDesigner†Evaluation Style=Few-shot, Replay=true2026.05 | 54.21 | 1.88 | |
| ARGDesignerSetting=Few-shot, Replay-based=false2026.05 | 53.84 | 3.62 | |
| ARGDesignerEvaluation Style=Few-shot2026.05 | 53.84 | 3.62 | |
| GDesignerSetting=Standard, Replay-based=false2026.05 | 53.36 | 4.95 | |
| GDesignerEvaluation Style=Standard2026.05 | 53.36 | 4.95 | |
| MasRouterSetting=Few-shot, Replay-based=false2026.05 | 52.42 | 3.14 | |
| MasRouterEvaluation Style=Few-shot2026.05 | 52.42 | 3.14 | |
| GDesignerSetting=Few-shot, Replay-based=false2026.05 | 51.92 | 3.89 | |
| GDesignerEvaluation Style=Few-shot2026.05 | 51.92 | 3.89 | |
| AgentDropoutSetting=Standard, Replay-based=true2026.05 | 51.75 | 2.42 | |
| AgentDropout†Evaluation Style=Standard, Replay=true2026.05 | 51.75 | 2.42 | |
| AgentDropoutSetting=Standard, Replay-based=false2026.05 | 49.88 | 4.96 | |
| AgentDropoutEvaluation Style=Standard2026.05 | 49.88 | 4.96 | |
| AFLOWSetting=Standard, Replay-based=true2026.05 | 49.63 | 1.75 | |
| AFLOW†Evaluation Style=Standard, Replay=true2026.05 | 49.63 | 1.75 | |
| AFLOWSetting=Standard, Replay-based=false2026.05 | 48.85 | 3.92 | |
| AFLOWEvaluation Style=Standard2026.05 | 48.85 | 3.92 | |
| AgentDropoutSetting=Few-shot, Replay-based=true2026.05 | 47.05 | 1.56 | |
| AgentDropout†Evaluation Style=Few-shot, Replay=true2026.05 | 47.05 | 1.56 | |
| AFLOWSetting=Few-shot, Replay-based=true2026.05 | 44.39 | 2.03 | |
| AFLOW†Evaluation Style=Few-shot, Replay=true2026.05 | 44.39 | 2.03 | |
| AgentDropoutSetting=Few-shot, Replay-based=false2026.05 | 43.54 | 4.18 | |
| AgentDropoutEvaluation Style=Few-shot2026.05 | 43.54 | 4.18 | |
| AFLOWSetting=Few-shot, Replay-based=false2026.05 | 42.18 | 3.37 | |
| AFLOWEvaluation Style=Few-shot2026.05 | 42.18 | 3.37 | |
| SC(CoT)-ICLSetting=Few-shot, Replay-based=false2026.05 | 37.88 | 6.16 | |
| SC(CoT)-ICLEvaluation Style=Few-shot2026.05 | 37.88 | 6.16 | |
| CompleteSetting=Few-shot, Replay-based=false2026.05 | 37.63 | — | |
| CompleteEvaluation Style=Few-shot2026.05 | 37.63 | — | |
| ChainSetting=Few-shot, Replay-based=false2026.05 | 36.92 | — | |
| ChainEvaluation Style=Few-shot2026.05 | 36.92 | — | |
| CoT-ICLSetting=Few-shot, Replay-based=false2026.05 | 36.83 | 4.29 | |
| CoT-ICLEvaluation Style=Few-shot2026.05 | 36.83 | 4.29 | |
| TreeSetting=Few-shot, Replay-based=false2026.05 | 36.73 | — | |
| TreeEvaluation Style=Few-shot2026.05 | 36.73 | — | |
| VanillaSetting=Few-shot, Replay-based=false2026.05 | 35.79 | — | |
| VanillaEvaluation Style=Few-shot2026.05 | 35.79 | — |