Spatial Reasoning on VSR zero-shot (test)
63.67Accuracy (zero-shot)Llama-2 7B
Evaluation Results
| Method | Links | |
|---|---|---|
| Llama-2 7BLLM backbone=Llama-2, Model Size=7B2025.12 | 63.67 | |
| Llama-2 Chat 7BLLM backbone=Llama-2, Model Size=7B, Variant=Chat2025.12 | 61.8 | |
| Moxin-7BLLM backbone=Moxin-7B, Model Size=7B2025.12 | 60.8 | |
| Mistral v0.1 7BLLM backbone=Mistral, Model Size=7B, Variant=v0.12025.12 | 58.5 | |
| Mistral Instruct v0.1 7BLLM backbone=Mistral, Model Size=7B, Variant=Instruct v0.12025.12 | 57.8 | |
| LLaVa v1.5 7B (Base)LLM backbone=LLaVa v1.5, Model Size=7B, Variant=Base2025.12 | 51.47 |