List your company

Reasoning on GPQA (80% test)

54.22AccuracySelf-Refine
36.675241.230145.78550.3399May 21, 2025

Evaluation results

MethodLinks
Self-RefineLLM Backbone=Llama3.3-70B2025.0554.22
ADASLLM Backbone=Llama3.3-70B2025.0553.6
MADLLM Backbone=Llama3.3-70B2025.0552.61
MAS-ZEROLLM Backbone=Llama3.3-70B2025.0552.41
MADLLM Backbone=GPT-4o2025.0552.01
MAS-ZEROLLM Backbone=Qwen2.5-32B2025.0551.81
CoT-SCLLM Backbone=Llama3.3-70B2025.0551.2
MAS-ZEROLLM Backbone=GPT-4o2025.0550.6
CoTLLM Backbone=Llama3.3-70B2025.0550.6
DebateLLM Backbone=Llama3.3-70B2025.0550.6
Self-RefineLLM Backbone=Qwen2.5-32B2025.0550.6
CoTLLM Backbone=Qwen2.5-32B2025.0550
SPPLLM Backbone=GPT-4o2025.0549.8
CoT-SCLLM Backbone=Qwen2.5-32B2025.0549.4
DebateLLM Backbone=Qwen2.5-32B2025.0549.4
ReConcileLLM Backbone=GPT-4o2025.0548.43
AFlowLLM Backbone=Llama3.3-70B2025.0547.59
ReConcileLLM Backbone=Llama3.3-70B2025.0547.17
ReConcileLLM Backbone=Qwen2.5-32B2025.0547.17
ADASLLM Backbone=Qwen2.5-32B2025.0547
DebateLLM Backbone=GPT-4o2025.0546.99
DyLANLLM Backbone=GPT-4o2025.0546.99
AFlowLLM Backbone=GPT-4o2025.0546.99
MaASLLM Backbone=Qwen2.5-32B2025.0546.99
Self-RefineLLM Backbone=GPT-4o2025.0546.39
AFlowLLM Backbone=Qwen2.5-32B2025.0546.39
SPPLLM Backbone=Llama3.3-70B2025.0546.18
CoTLLM Backbone=GPT-4o2025.0545.78
ADASLLM Backbone=GPT-4o2025.0545.2
MAS-GPTLLM Backbone=GPT-4o2025.0543.98
MaASLLM Backbone=Llama3.3-70B2025.0543.98
MADLLM Backbone=Qwen2.5-32B2025.0543.57
CoT-SCLLM Backbone=GPT-4o2025.0543.37
MaASLLM Backbone=GPT-4o2025.0543.37
SPPLLM Backbone=Qwen2.5-32B2025.0542.77
DyLANLLM Backbone=Qwen2.5-32B2025.0542.57
DyLANLLM Backbone=Llama3.3-70B2025.0541.57
MAS-GPTLLM Backbone=Llama3.3-70B2025.0540
MAS-GPTLLM Backbone=Qwen2.5-32B2025.0537.35