List your company

Arithmetic Reasoning on Arithmetics

100AccuracyOracle
41.7656.887287.12Mar 21, 2026Mar 25, 2026Mar 29, 2026Apr 2, 2026Apr 6, 2026Apr 10, 2026Apr 14, 2026

Evaluation results

MethodLinks
OracleModel=Qwen2.5-7B, Black-Box Friendly=true2026.04100
OracleModel=Llama3.2-3B, Black-Box Friendly=true2026.0499.8
OracleModel=Qwen2.5-3B, Black-Box Friendly=true2026.0499.5
OracleModel=Llama3.1-8B, Black-Box Friendly=true2026.0499.3
Majority VoteBackbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0398.3
RCSmedoidModel=Llama3.2-3B, Black-Box Friendly=true2026.0498.3
RCSuniModel=Llama3.2-3B, Black-Box Friendly=true2026.0498.2
RCSfreqModel=Llama3.2-3B, Black-Box Friendly=true2026.0498.2
OracleModel=Gemma2-9B, Black-Box Friendly=true2026.0498.2
SCModel=Llama3.2-3B, Black-Box Friendly=true2026.0497.5
RCSprobModel=Llama3.2-3B, Black-Box Friendly=false2026.0497.5
CEModel=Gemma2-9B, Black-Box Friendly=false2026.0497.5
CEModel=Llama3.2-3B, Black-Box Friendly=false2026.0497.3
SCModel=Gemma2-9B, Black-Box Friendly=true2026.0497.3
RCSfreqModel=Gemma2-9B, Black-Box Friendly=true2026.0497.3
RCSprobModel=Gemma2-9B, Black-Box Friendly=false2026.0497.3
Majority VoteModel=Falcon3-7B2026.0397.2
RCSmedoidModel=Gemma2-9B, Black-Box Friendly=true2026.0497.2
RCSuniModel=Gemma2-9B, Black-Box Friendly=true2026.0497.2
GreedyModel=Qwen2.5-7B, Black-Box Friendly=true2026.0497
DARBackbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0396.7
GreedyModel=Llama3.2-3B, Black-Box Friendly=true2026.0496
GreedyModel=Gemma2-9B, Black-Box Friendly=true2026.0496
ANLLModel=Gemma2-9B, Black-Box Friendly=false2026.0496
Society Of MindModel=Llama3.1-8B2026.0395.5
DARBackbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0395.5
Uncertain PromptBackbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0395.3
Uncertain PromptModel=Falcon3-7B2026.0395
DARModel=Falcon3-7B2026.0395
Society Of MindBackbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0395
Uncertain PromptBackbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0395
NLLModel=Gemma2-9B, Black-Box Friendly=false2026.0494.8
Vote PromptModel=Llama3.1-8B2026.0394.5
DARModel=Llama3.1-8B2026.0394.5
Uncertain PromptModel=Llama3.1-8B2026.0393
Society Of MindBackbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0392.5
Vote PromptBackbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0392.3
Vote PromptModel=Falcon3-7B2026.0391.5
Majority VoteBackbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0391.4
MAD-M^2Backbone Model=Falcon3-7B, Number of Agents=4, Number of Rounds=22026.0391.3
Vote PromptBackbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0391
MAD-M2Model=Falcon3-7B2026.0390.7
Majority VoteModel=Llama3.1-8B2026.0390.7
Society Of MindModel=Falcon3-7B2026.0390
GreedyModel=Llama3.1-8B, Black-Box Friendly=true2026.0489.5
GreedyModel=Qwen2.5-3B, Black-Box Friendly=true2026.0489
ANLLModel=Llama3.2-3B, Black-Box Friendly=false2026.0488.5
RCSuniModel=Llama3.1-8B, Black-Box Friendly=true2026.0488.5
NLLModel=Llama3.2-3B, Black-Box Friendly=false2026.0486.5
RCSmedoidModel=Llama3.1-8B, Black-Box Friendly=true2026.0486.5
RCSuniModel=Qwen2.5-3B, Black-Box Friendly=true2026.0485.7
RCSprobModel=Llama3.1-8B, Black-Box Friendly=false2026.0485.2
RCSprobModel=Qwen2.5-3B, Black-Box Friendly=false2026.0484.2
RCSfreqModel=Llama3.1-8B, Black-Box Friendly=true2026.0483.7
RCSfreqModel=Qwen2.5-3B, Black-Box Friendly=true2026.0482.8
RCSmedoidModel=Qwen2.5-3B, Black-Box Friendly=true2026.0481.7
Single-agentBackbone Model=Llama3.1-8B2026.0381
CEModel=Llama3.1-8B, Black-Box Friendly=false2026.0481
CEModel=Qwen2.5-3B, Black-Box Friendly=false2026.0480.3
MAD-M2Model=Llama3.1-8B2026.0380
DARBackbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0379.7
Single-agentModel=Llama3.1-8B2026.0379
SCModel=Qwen2.5-3B, Black-Box Friendly=true2026.0478.7
SCModel=Llama3.1-8B, Black-Box Friendly=true2026.0478.3
MAD-M^2Backbone Model=Llama3.1-8B, Number of Agents=4, Number of Rounds=22026.0378
RCSprobModel=Qwen2.5-7B, Black-Box Friendly=false2026.0477.7
RCSmedoidModel=Qwen2.5-7B, Black-Box Friendly=true2026.0477.5
RCSuniModel=Qwen2.5-7B, Black-Box Friendly=true2026.0477.5
RCSfreqModel=Qwen2.5-7B, Black-Box Friendly=true2026.0476.7
CEModel=Qwen2.5-7B, Black-Box Friendly=false2026.0475.5
DARModel=Qwen2.5-1.5B2026.0375
DARBackbone Model=Qwen2.5-1.5B, Number of Agents=4, Number of Rounds=22026.0374.7
SCModel=Qwen2.5-7B, Black-Box Friendly=true2026.0474.3
ANLLModel=Llama3.1-8B, Black-Box Friendly=false2026.0473.5
NLLModel=Llama3.1-8B, Black-Box Friendly=false2026.0473.3
Majority VoteBackbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0373
Society Of MindModel=Qwen2.5-1.5B2026.0372
Vote PromptModel=Qwen2.5-1.5B2026.0372
MAD-M^2Backbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0371.3
DARModel=Qwen2.5-3B2026.0370.5
Vote PromptModel=Qwen2.5-3B2026.0370
Uncertain PromptBackbone Model=Qwen2.5-1.5B, Number of Agents=4, Number of Rounds=22026.0369.7
MAD-M2Model=Qwen2.5-3B2026.0369
Vote PromptBackbone Model=Qwen2.5-1.5B, Number of Agents=4, Number of Rounds=22026.0368
Uncertain PromptModel=Qwen2.5-1.5B2026.0367.5
Majority VoteModel=Qwen2.5-3B2026.0365.8
Uncertain PromptModel=Qwen2.5-3B2026.0365
Society Of MindBackbone Model=Qwen2.5-1.5B, Number of Agents=4, Number of Rounds=22026.0363.7
Society Of MindModel=Qwen2.5-3B2026.0363
Vote PromptBackbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0361.3
Single-agentBackbone Model=Falcon3-7B2026.0360.5
Majority VoteBackbone Model=Qwen2.5-1.5B, Number of Agents=4, Number of Rounds=22026.0359.9
Majority VoteModel=Qwen2.5-1.5B2026.0359.2
Uncertain PromptBackbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0358.7
Single-agentModel=Falcon3-7B2026.0358
Single-agentBackbone Model=Qwen2.5-3B2026.0355
NLLModel=Qwen2.5-7B, Black-Box Friendly=false2026.0447.3
ANLLModel=Qwen2.5-7B, Black-Box Friendly=false2026.0447
Society Of MindBackbone Model=Qwen2.5-3B, Number of Agents=4, Number of Rounds=22026.0346.3
MAD-M2Model=Qwen2.5-1.5B2026.0344