Our new X account is live! Follow @wizwand_team for updates

Compositional Generalization on Evaluation Dataset (Fold 1 Seen)

0.6191Score

Mistral-7B

Updated 4d ago

Evaluation Results

Method	Links
Mistral-7B 2026.01		0.6191
Mistral-7B + COGLM 2026.01		0.5794
Gemma-7B 2026.01		0.5747
LLaMA-3-8B + COGLM 2026.01		0.5678
Gemma-7B + COGLM 2026.01		0.5626
Qwen2.5-7B + COGLM 2026.01		0.5465
Qwen2.5-7B 2026.01		0.5203
LLaMA-3-8B 2026.01		0.5095
DeepSeek-7B + COGLM 2026.01		0.4425
Claude 3.5 Sonnet + FS* 2026.01		0.4062
GPT-4o + FS* 2026.01		0.3908
DeepSeek V3 + FS* 2026.01		0.3885
DeepSeek V3 2026.01		0.3806
Claude 3.5 Sonnet 2026.01		0.3671
DeepSeek-7B 2026.01		0.3261
GPT-4o 2026.01		0.3161
Llama 3 (70B) + FS* 2026.01		0.2503
Llama 3 (70B) 2026.01		0.1797