Compositional Generalization on Evaluation Dataset Seen Average
62.34ScoreMistral-7B
Evaluation Results
| Method | Links | |
|---|---|---|
| Mistral-7BFine-tuning Strategy=LoRA, Model Backbone=Mistral-7B2026.01 | 62.34 | |
| LLaMA-3-8B + COGLMFine-tuning Strategy=LoRA, Model Backbone=LLaMA-3-8B, Method Components=COGLM2026.01 | 62.31 | |
| Mistral-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Mistral-7B, Method Components=COGLM2026.01 | 62.07 | |
| Gemma-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Gemma-7B, Method Components=COGLM2026.01 | 60.53 | |
| Qwen2.5-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=Qwen2.5-7B, Method Components=COGLM2026.01 | 59.41 | |
| Gemma-7BFine-tuning Strategy=LoRA, Model Backbone=Gemma-7B2026.01 | 58.42 | |
| Qwen2.5-7BFine-tuning Strategy=LoRA, Model Backbone=Qwen2.5-7B2026.01 | 58.24 | |
| LLaMA-3-8BFine-tuning Strategy=LoRA, Model Backbone=LLaMA-3-8B2026.01 | 57.49 | |
| DeepSeek-7B + COGLMFine-tuning Strategy=LoRA, Model Backbone=DeepSeek-7B, Method Components=COGLM2026.01 | 47.25 | |
| DeepSeek V3 + FS*Evaluation Protocol=Few-shot, Model Backbone=DeepSeek V32026.01 | 43.22 | |
| Claude 3.5 Sonnet + FS*Evaluation Protocol=Few-shot, Model Backbone=Claude 3.5 Sonnet2026.01 | 42.95 | |
| DeepSeek V3Evaluation Protocol=Zero-shot, Model Backbone=DeepSeek V32026.01 | 41.96 | |
| DeepSeek-7BFine-tuning Strategy=LoRA, Model Backbone=DeepSeek-7B2026.01 | 40.38 | |
| Claude 3.5 SonnetEvaluation Protocol=Zero-shot, Model Backbone=Claude 3.5 Sonnet2026.01 | 39.8 | |
| GPT-4o + FS*Evaluation Protocol=Few-shot, Model Backbone=GPT-4o2026.01 | 39.42 | |
| GPT-4oEvaluation Protocol=Zero-shot, Model Backbone=GPT-4o2026.01 | 35.73 | |
| Llama 3 (70B) + FS*Evaluation Protocol=Few-shot, Model Backbone=Llama 3 (70B)2026.01 | 30.56 | |
| Llama 3 (70B)Evaluation Protocol=Zero-shot, Model Backbone=Llama 3 (70B)2026.01 | 21.31 |