| Paper | Code | BLEU-4 | METEOR | ROUGE-L | CIDEr | SPICE | Detection | ACC | #Learning Samples (N) | ModelName | ReleaseDate |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Few-Shot Multimodal Explanation for Visual Question Answering | ✓ Link | 67.91 | 50.55 | 79.41 | 510.44 | 64.09 | 29.09 | 51.45 | 16 | MEAgent | 2024-10-28 |
| GPT-4 Technical Report | ✓ Link | 45.51 | 35.17 | 52.67 | 269.68 | 37.67 | 7.00 | 42.30 | 16 | GPT-4-1106-Vision-Preview | 2023-03-15 |
| Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context | ✓ Link | 41.87 | 34.61 | 55.90 | 276.14 | 40.58 | 1.40 | 40.88 | 16 | Gemini-1.5 Pro | 2024-03-08 |
| Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond | ✓ Link | 24.30 | 23.40 | 34.52 | 201.47 | 26.13 | 1.05 | 40.33 | 16 | Qwen-VL-Max | 2023-08-24 |
| CogVLM: Visual Expert for Pretrained Language Models | ✓ Link | 14.45 | 17.53 | 24.28 | 127.37 | 17.70 | 0.89 | 34.23 | 16 | GLM-4V | 2023-11-06 |
| Variational Causal Inference Network for Explanatory Visual Question Answering | ✓ Link | 9.17 | 19.82 | 33.34 | 4.28 | 13.39 | 0.28 | 17.77 | 16 | VCIN | 2023-01-01 |
| REX: Reasoning-aware and Grounded Explanation | ✓ Link | 0.00 | 4.37 | 23.23 | 0.89 | 0.00 | 0.00 | 17.77 | 16 | REX | 2022-03-11 |