List your company

Logical Reasoning on GeoShape BBEH

43AccuracyMemAPO
1.412.22333.8Mar 23, 2026

Evaluation results

MethodLinks
MemAPOBackbone=GPT-4o-mini2026.0343
RaRBackbone=GPT-4o-mini2026.0330
SPOBackbone=Qwen3-8B2026.0325
Step-BackBackbone=GPT-4o-mini2026.0325
MemAPOBackbone=Qwen3-8B2026.0324
SPOBackbone=GPT-4o-mini2026.0324
TextGradBackbone=GPT-4o-mini2026.0323
OPROBackbone=GPT-4o-mini2026.0318
CoTBackbone=GPT-4o-mini2026.0317
TextGradBackbone=Qwen3-8B2026.0314
ProTeGiBackbone=GPT-4o-mini2026.0314
IOBackbone=GPT-4o-mini2026.0313
PromptBreederBackbone=GPT-4o-mini2026.0313
PromptBreederBackbone=Qwen3-8B2026.0310
OPROBackbone=Qwen3-8B2026.039
Step-BackBackbone=Qwen3-8B2026.038
ProTeGiBackbone=Qwen3-8B2026.037
IOBackbone=Qwen3-8B2026.036
CoTBackbone=Qwen3-8B2026.036
RaRBackbone=Qwen3-8B2026.033