List your company

Massive Multitask Language Understanding on MMLU (Accuracy and Latency)

82.57AccuracyAgentDropout
46.773256.066665.3674.6534May 28, 2026

Evaluation results

MethodLinks
AgentDropoutBackbone=Qwen2.5-72B-Instruct, Topology=Star, Training-Free (TF)=false2026.0582.57180.62
AgentDropoutBackbone=Qwen2.5-72B-Instruct, Topology=FullConnected, Training-Free (TF)=false2026.0581.92184.71
Vanilla MASBackbone=Qwen2.5-72B-Instruct, Topology=Star, Training-Free (TF)=true2026.0581.7235.17
Vanilla MASBackbone=Qwen2.5-72B-Instruct, Topology=Random, Training-Free (TF)=true2026.0581.48148.03
AgentDropoutBackbone=Qwen2.5-72B-Instruct, Topology=Layered, Training-Free (TF)=false2026.0581.48121.07
Vanilla MASBackbone=Qwen2.5-72B-Instruct, Topology=Chain, Training-Free (TF)=true2026.0581.05217.53
AgentDropoutBackbone=Qwen2.5-72B-Instruct, Topology=Random, Training-Free (TF)=false2026.0580.83116.22
Vanilla MASBackbone=Qwen2.5-72B-Instruct, Topology=Layered, Training-Free (TF)=true2026.0580.39156.18
SC-CoTBackbone=Qwen2.5-72B-Instruct, Topology=-, Training-Free (TF)=true2026.0579.7484.06
LLM-DebateBackbone=Qwen2.5-72B-Instruct, Topology=Debate, Training-Free (TF)=true2026.0579.52149.85
CONCATBackbone=Qwen2.5-72B-Instruct, Topology=Hybrid, Training-Free (TF)=true2026.0579.52107.59
Vanilla MASBackbone=Qwen2.5-72B-Instruct, Topology=FullConnected, Training-Free (TF)=true2026.0579.3196.93
AgentDropoutBackbone=Qwen2.5-72B-Instruct, Topology=Chain, Training-Free (TF)=false2026.0578.43176.02
CONCATBackbone=Qwen2.5-14B-Instruct, Topology=Hybrid, Training-Free (TF)=true2026.0577.7825.76
CoTBackbone=Qwen2.5-72B-Instruct, Topology=-, Training-Free (TF)=true2026.0577.5623.85
AgentDropoutBackbone=Qwen2.5-14B-Instruct, Topology=Chain, Training-Free (TF)=false2026.0576.9134.17
Vanilla MASBackbone=Qwen2.5-14B-Instruct, Topology=Star, Training-Free (TF)=true2026.0576.2551.32
Vanilla MASBackbone=Qwen2.5-14B-Instruct, Topology=Random, Training-Free (TF)=true2026.0576.0348.65
AgentDropoutBackbone=Qwen2.5-14B-Instruct, Topology=FullConnected, Training-Free (TF)=false2026.0575.8239.77
LLM-DebateBackbone=Qwen2.5-14B-Instruct, Topology=Debate, Training-Free (TF)=true2026.0575.650
Vanilla MASBackbone=Qwen2.5-14B-Instruct, Topology=Chain, Training-Free (TF)=true2026.0575.1641.82
AgentDropoutBackbone=Qwen2.5-14B-Instruct, Topology=Star, Training-Free (TF)=false2026.0574.9537.05
Vanilla MASBackbone=Qwen2.5-14B-Instruct, Topology=FullConnected, Training-Free (TF)=true2026.0574.7355.12
AgentDropoutBackbone=Qwen2.5-14B-Instruct, Topology=Random, Training-Free (TF)=false2026.0574.5136.57
AgentDropoutBackbone=Qwen2.5-14B-Instruct, Topology=Layered, Training-Free (TF)=false2026.0574.2937.71
SC-CoTBackbone=Qwen2.5-14B-Instruct, Topology=-, Training-Free (TF)=true2026.0573.8620.78
Vanilla MASBackbone=Qwen2.5-14B-Instruct, Topology=Layered, Training-Free (TF)=true2026.0573.4251.87
CoTBackbone=Qwen2.5-14B-Instruct, Topology=-, Training-Free (TF)=true2026.0573.26.07
AgentDropoutBackbone=Llama3-8B-Instruct, Topology=Random, Training-Free (TF)=false2026.0560.3550.33
Vanilla MASBackbone=Llama3-8B-Instruct, Topology=Layered, Training-Free (TF)=true2026.0559.2671.17
Vanilla MASBackbone=Llama3-8B-Instruct, Topology=FullConnected, Training-Free (TF)=true2026.0557.371.81
AgentDropoutBackbone=Llama3-8B-Instruct, Topology=Chain, Training-Free (TF)=false2026.0556.8639.62
LLM-DebateBackbone=Llama3-8B-Instruct, Topology=Debate, Training-Free (TF)=true2026.0556.4363.32
AgentDropoutBackbone=Llama3-8B-Instruct, Topology=Star, Training-Free (TF)=false2026.0555.9948.67
CONCATBackbone=Llama3-8B-Instruct, Topology=Hybrid, Training-Free (TF)=true2026.0555.5629.39
AgentDropoutBackbone=Llama3-8B-Instruct, Topology=Layered, Training-Free (TF)=false2026.0554.948.52
Vanilla MASBackbone=Llama3-8B-Instruct, Topology=Chain, Training-Free (TF)=true2026.0554.6856.95
AgentDropoutBackbone=Llama3-8B-Instruct, Topology=FullConnected, Training-Free (TF)=false2026.0554.2552.92
Vanilla MASBackbone=Llama3-8B-Instruct, Topology=Star, Training-Free (TF)=true2026.0553.5965.9
Vanilla MASBackbone=Llama3-8B-Instruct, Topology=Random, Training-Free (TF)=true2026.0551.6362.62
SC-CoTBackbone=Llama3-8B-Instruct, Topology=-, Training-Free (TF)=true2026.0550.5417.66
CoTBackbone=Llama3-8B-Instruct, Topology=-, Training-Free (TF)=true2026.0548.155.69