End-to-end Question Answering on FinSlides
85.5Overall ScoreSlideAgent
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| SlideAgentMethod Type=Multimodal RAG (Type 2) and Agentic Methods (Type 3), Base Model=GPT-4o2025.10 | 85.5 | 85.9 | 79.6 | |
| VDocRAGMethod Type=Multimodal RAG (Type 2) and Agentic Methods (Type 3), Base Model=GPT-4o2025.10 | 83.5 | 83.8 | 64.2 | |
| ViDoRAGMethod Type=Multimodal RAG (Type 2) and Agentic Methods (Type 3), Base Model=GPT-4o2025.10 | 82.2 | 83.3 | 65.1 | |
| COLPALIMethod Type=Multimodal RAG (Type 2) and Agentic Methods (Type 3), Base Model=GPT-4o2025.10 | 80.9 | 81.5 | 62.7 | |
| GPT-4oMethod Type=Multimodal LLMs (Type 1)2025.10 | 80 | 80.8 | 62.1 | |
| VisRAGMethod Type=Multimodal RAG (Type 2) and Agentic Methods (Type 3), Base Model=GPT-4o2025.10 | 79.2 | 81.1 | 75.8 | |
| Gemini-2.5-FlashMethod Type=Multimodal LLMs (Type 1)2025.10 | 76.2 | 75.8 | 100 | |
| Gemini-2.0-FlashMethod Type=Multimodal LLMs (Type 1)2025.10 | 70.8 | 70.6 | 77.8 | |
| SlideAgentType=3, Base Model=InternVL3-8B2025.10 | 63.3 | 62.8 | 68.3 | |
| Qwen2.5-VL-32B-InstructType=12025.10 | 57.4 | 56.6 | 87.5 | |
| Gemini-2.5-Flash-LiteMethod Type=Multimodal LLMs (Type 1)2025.10 | 57 | 56.6 | 68.3 | |
| Claude-4.1-OpusMethod Type=Multimodal LLMs (Type 1)2025.10 | 56.5 | 54.8 | 73.3 | |
| Qwen2.5-VL-7B-InstructType=12025.10 | 53.6 | 52.5 | 85.7 | |
| ViDoRAGType=2, Base Model=InternVL3-8B2025.10 | 52.7 | 55.4 | 66.6 | |
| VDocRAGType=2, Base Model=InternVL3-8B2025.10 | 51.8 | 50.1 | 65.9 | |
| VisRAGType=2, Base Model=InternVL3-8B2025.10 | 51.1 | 49.6 | 65.2 | |
| COLPALIType=2, Base Model=InternVL3-8B2025.10 | 50.4 | 49.3 | 65.7 | |
| InternVL3-8BType=12025.10 | 49.8 | 49.5 | 64.3 | |
| Phi-3-vision-128k-instructType=12025.10 | 48.8 | 48.5 | 64.3 | |
| Claude-3.5-HaikuMethod Type=Multimodal LLMs (Type 1)2025.10 | 48.5 | 49.5 | 29.6 | |
| llava-v1.6-vicuna-13b-hfType=12025.10 | 32 | 31.3 | 64.3 | |
| Llama-3.2-11B-Vision-InstructType=12025.10 | 23.3 | 23.2 | 26.2 | |
| llava-1.5-13b-hfType=12025.10 | 20.6 | 16.5 | 36.7 | |
| llava-v1.6-mistral-7b-hfType=12025.10 | 12.2 | 12.1 | 17.8 | |
| llava-1.5-7b-hfType=12025.10 | 10.7 | 10.1 | 16.6 |