List your company

Mathematical Word Problem Solving on SVAMP

96.94AccuracyAgentRevive
82.17286.00689.8493.674Aug 15, 2025Oct 1, 2025Nov 17, 2025Jan 3, 2026Feb 19, 2026Apr 7, 2026May 25, 2026

Evaluation results

MethodLinks
AgentReviveVar. NS=True, Flex. State=True, Base model=Qwen2.5-72B-Instruct2026.0596.94
SafeSieveParadigm=post-prune, Base model=deepseek-V3-671B2025.0896.6
AgentDropoutParadigm=post-prune, Base model=deepseek-V3-671B2025.0896.01
AgentPruneParadigm=post-prune, Base model=deepseek-V3-671B2025.0895.4
AgentDropoutVar. NS=True, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0595.34
ARG-DesignerVar. NS=True, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0594.63
AgentPruneVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0594.33
MASround=TVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0594.07
CoTParadigm=single, Base model=deepseek-V3-671B2025.0893.94
AutogenVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0593.86
G-DesignerParadigm=pre-design, Base model=deepseek-V3-671B2025.0893.79
SC (CoT)Var. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0593.79
VanillaParadigm=single, Base model=deepseek-V3-671B2025.0893.67
CoTVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0593.35
SafeSieveParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0893.29
AgentPruneParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0893.05
AgentDropoutParadigm=post-prune, Base model=gpt-4o-mini (~8B)2025.0892.9
G-DesignerVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0592.81
VanillaVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0592.67
AgentVerseVar. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0592.45
GPTSwarmParadigm=pre-design, Base model=deepseek-V3-671B2025.0892.43
MASround=1Var. NS=False, Flex. State=False, Base model=Qwen2.5-72B-Instruct2026.0592.36
G-DesignerParadigm=pre-design, Base model=gpt-4o-mini (~8B)2025.0890.29
VanillaParadigm=single, Base model=gpt-4o-mini (~8B)2025.0888.26
ATOMBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0587.4
RandomBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0587.29
ARG-DesignerBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0587.29
LLM-DebateBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0587.19
G-DesignerBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0587.08
GPTSwarmParadigm=pre-design, Base model=gpt-4o-mini (~8B)2025.0887.02
AgentPruneBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0586.77
CoTParadigm=single, Base model=gpt-4o-mini (~8B)2025.0886.24
AgentDropoutBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0586.1
ChainBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0585.36
CoTBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0584.69
VanillaBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0584.39
CompleteBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0583.13
StarBackbone Model=Meta-Llama-3.1-8B-Instruct2026.0582.74