List your company
ResearchFormerly WizWand

Compositional Generalization on Evaluation Dataset Seen Average

62.34ScoreMistral-7B
19.668830.746941.82552.9031Jan 29, 2026
Updated 8mo ago

Evaluation results

MethodLinks
Mistral-7BFine-tuning Strategy=LoRA, Model Backbone=Mistral-7B2026.0162.34
LLaMA-3-8B + COGLMFine-tuning Strategy=LoRA, Model Backbone=LLaMA-3-8B, Method Components=COGLM2026.0162.31
Mistral-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Mistral-7B, Method Components=COGLM2026.0162.07
Gemma-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Gemma-7B, Method Components=COGLM2026.0160.53
Qwen2.5-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Qwen2.5-7B, Method Components=COGLM2026.0159.41
Gemma-7BFine-tuning Strategy=LoRA, Model Backbone=Gemma-7B2026.0158.42
Qwen2.5-7BFine-tuning Strategy=LoRA, Model Backbone=Qwen2.5-7B2026.0158.24
LLaMA-3-8BFine-tuning Strategy=LoRA, Model Backbone=LLaMA-3-8B2026.0157.49
DeepSeek-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=DeepSeek-7B, Method Components=COGLM2026.0147.25
DeepSeek V3 + FS*Evaluation Protocol=Few-shot, Model Backbone=DeepSeek V32026.0143.22
Claude 3.5 Sonnet + FS*Evaluation Protocol=Few-shot, Model Backbone=Claude 3.5 Sonnet2026.0142.95
DeepSeek V3Evaluation Protocol=Zero-shot, Model Backbone=DeepSeek V32026.0141.96
DeepSeek-7BFine-tuning Strategy=LoRA, Model Backbone=DeepSeek-7B2026.0140.38
Claude 3.5 SonnetEvaluation Protocol=Zero-shot, Model Backbone=Claude 3.5 Sonnet2026.0139.8
GPT-4o + FS*Evaluation Protocol=Few-shot, Model Backbone=GPT-4o2026.0139.42
GPT-4oEvaluation Protocol=Zero-shot, Model Backbone=GPT-4o2026.0135.73
Llama 3 (70B) + FS*Evaluation Protocol=Few-shot, Model Backbone=Llama 3 (70B)2026.0130.56
Llama 3 (70B)Evaluation Protocol=Zero-shot, Model Backbone=Llama 3 (70B)2026.0121.31