Class-level Continual Learning on TACO
73.8AAMasFACT
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| MasFACTSetting=Standard, Replay-based=false2026.05 | 73.8 | -0.11 | |
| MasFACTEvaluation Style=Standard2026.05 | 73.8 | -0.11 | |
| GDesignerSetting=Standard, Replay-based=true2026.05 | 72.35 | 1.06 | |
| GDesigner†Evaluation Style=Standard, Replay=true2026.05 | 72.35 | 1.06 | |
| MasRouterSetting=Standard, Replay-based=true2026.05 | 71.57 | 2.25 | |
| MasRouter†Evaluation Style=Standard, Replay=true2026.05 | 71.57 | 2.25 | |
| MasFACTSetting=Few-shot, Replay-based=false2026.05 | 70.94 | 0.12 | |
| MasFACTEvaluation Style=Few-shot2026.05 | 70.94 | 0.12 | |
| ARGDesignerSetting=Standard, Replay-based=true2026.05 | 70.24 | 1.92 | |
| ARGDesigner†Evaluation Style=Standard, Replay=true2026.05 | 70.24 | 1.92 | |
| MasRouterSetting=Few-shot, Replay-based=true2026.05 | 69.76 | 2.89 | |
| MasRouter†Evaluation Style=Few-shot, Replay=true2026.05 | 69.76 | 2.89 | |
| GDesignerSetting=Standard, Replay-based=false2026.05 | 69.42 | 5.26 | |
| GDesignerEvaluation Style=Standard2026.05 | 69.42 | 5.26 | |
| ARGDesignerSetting=Standard, Replay-based=false2026.05 | 69.41 | 4.28 | |
| ARGDesignerEvaluation Style=Standard2026.05 | 69.41 | 4.28 | |
| ARGDesignerSetting=Few-shot, Replay-based=true2026.05 | 68.82 | 2.21 | |
| ARGDesigner†Evaluation Style=Few-shot, Replay=true2026.05 | 68.82 | 2.21 | |
| MasRouterSetting=Standard, Replay-based=false2026.05 | 68.75 | 3.7 | |
| MasRouterEvaluation Style=Standard2026.05 | 68.75 | 3.7 | |
| MasRouterSetting=Few-shot, Replay-based=false2026.05 | 67.62 | 3.12 | |
| MasRouterEvaluation Style=Few-shot2026.05 | 67.62 | 3.12 | |
| ARGDesignerSetting=Few-shot, Replay-based=false2026.05 | 67.36 | 3.74 | |
| ARGDesignerEvaluation Style=Few-shot2026.05 | 67.36 | 3.74 | |
| GDesignerSetting=Few-shot, Replay-based=true2026.05 | 67.1 | 2.22 | |
| GDesigner†Evaluation Style=Few-shot, Replay=true2026.05 | 67.1 | 2.22 | |
| GDesignerSetting=Few-shot, Replay-based=false2026.05 | 66.41 | 4.81 | |
| GDesignerEvaluation Style=Few-shot2026.05 | 66.41 | 4.81 | |
| AgentDropoutSetting=Standard, Replay-based=true2026.05 | 65.91 | 1.42 | |
| AgentDropout†Evaluation Style=Standard, Replay=true2026.05 | 65.91 | 1.42 | |
| AFLOWSetting=Standard, Replay-based=true2026.05 | 65.48 | 2.13 | |
| AFLOW†Evaluation Style=Standard, Replay=true2026.05 | 65.48 | 2.13 | |
| AFLOWSetting=Few-shot, Replay-based=true2026.05 | 64.75 | 2.64 | |
| AFLOW†Evaluation Style=Few-shot, Replay=true2026.05 | 64.75 | 2.64 | |
| AFLOWSetting=Standard, Replay-based=false2026.05 | 64.12 | 4.21 | |
| AFLOWEvaluation Style=Standard2026.05 | 64.12 | 4.21 | |
| AgentDropoutSetting=Few-shot, Replay-based=true2026.05 | 63.38 | 1.7 | |
| AgentDropout†Evaluation Style=Few-shot, Replay=true2026.05 | 63.38 | 1.7 | |
| AgentDropoutSetting=Standard, Replay-based=false2026.05 | 63.14 | 3.53 | |
| AgentDropoutEvaluation Style=Standard2026.05 | 63.14 | 3.53 | |
| AFLOWSetting=Few-shot, Replay-based=false2026.05 | 62.84 | 3.68 | |
| AFLOWEvaluation Style=Few-shot2026.05 | 62.84 | 3.68 | |
| AgentDropoutSetting=Few-shot, Replay-based=false2026.05 | 62.72 | 3.17 | |
| AgentDropoutEvaluation Style=Few-shot2026.05 | 62.72 | 3.17 | |
| SC(CoT)-ICLSetting=Few-shot, Replay-based=false2026.05 | 59.22 | 3.03 | |
| SC(CoT)-ICLEvaluation Style=Few-shot2026.05 | 59.22 | 3.03 | |
| CompleteSetting=Few-shot, Replay-based=false2026.05 | 59.01 | — | |
| CompleteEvaluation Style=Few-shot2026.05 | 59.01 | — | |
| ChainSetting=Few-shot, Replay-based=false2026.05 | 58.46 | — | |
| ChainEvaluation Style=Few-shot2026.05 | 58.46 | — | |
| CoT-ICLSetting=Few-shot, Replay-based=false2026.05 | 58.04 | 4.38 | |
| CoT-ICLEvaluation Style=Few-shot2026.05 | 58.04 | 4.38 | |
| VanillaSetting=Few-shot, Replay-based=false2026.05 | 55.15 | — | |
| VanillaEvaluation Style=Few-shot2026.05 | 55.15 | — | |
| TreeSetting=Few-shot, Replay-based=false2026.05 | 54.55 | — | |
| TreeEvaluation Style=Few-shot2026.05 | 54.55 | — |