Fs Mevqa On Sme
Metriken
#Learning Samples (N)
ACC
BLEU-4
CIDEr
Detection
METEOR
ROUGE-L
SPICE
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Modellname | #Learning Samples (N) | ACC | BLEU-4 | CIDEr | Detection | METEOR | ROUGE-L | SPICE | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|---|---|
| REX | 16 | 17.77 | 0.00 | 0.89 | 0.00 | 4.37 | 23.23 | 0.00 | REX: Reasoning-aware and Grounded Explanation | |
| VCIN | 16 | 17.77 | 9.17 | 4.28 | 0.28 | 19.82 | 33.34 | 13.39 | Variational Causal Inference Network for Explanatory Visual Question Answering | |
| Qwen-VL-Max | 16 | 40.33 | 24.30 | 201.47 | 1.05 | 23.40 | 34.52 | 26.13 | Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyond | |
| MEAgent | 16 | 51.45 | 67.91 | 510.44 | 29.09 | 50.55 | 79.41 | 64.09 | Few-Shot Multimodal Explanation for Visual Question Answering | |
| Gemini-1.5 Pro | 16 | 40.88 | 41.87 | 276.14 | 1.40 | 34.61 | 55.90 | 40.58 | Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context | |
| GLM-4V | 16 | 34.23 | 14.45 | 127.37 | 0.89 | 17.53 | 24.28 | 17.70 | CogVLM: Visual Expert for Pretrained Language Models | |
| GPT-4-1106-Vision-Preview | 16 | 42.30 | 45.51 | 269.68 | 7.00 | 35.17 | 52.67 | 37.67 | GPT-4 Technical Report |
0 of 7 row(s) selected.