Multimodal Reasoning on SEED (accuracy)
44.2Accuracy (SEED)GRIP (Idefics2-8B)
Evaluation Results
| Method | Links | |
|---|---|---|
| GRIP (Idefics2-8B)Target Model=Gemini 2.5 Flash Lite, Retriever Training Source=Idefics2-8B2026.06 | 44.2 | |
| GRIP (Qwen2.5-VL-7B)Target Model=Gemini 2.5 Flash Lite, Retriever Training Source=Qwen2.5-VL-7B2026.06 | 43.7 | |
| CLIP Similarity BasedTarget Model=Gemini 2.5 Flash Lite, Retrieval Strategy=CLIP Similarity2026.06 | 43.5 | |
| Zero ShotTarget Model=Gemini 2.5 Flash Lite, Retrieval Strategy=Zero Shot2026.06 | 40.7 |