Update README.md
Browse files
README.md
CHANGED
|
@@ -85,10 +85,10 @@ Our LoRA adapter shows improvements over the base Florence-2 model across all me
|
|
| 85 |
|
| 86 |
| Metric | Base Model | Adapted Model | Improvement |
|
| 87 |
|---------|------------|-----------------------|-------------|
|
|
|
|
| 88 |
| METEOR | 0.213 | 0.240 | +12.7% |
|
| 89 |
| BLEU | 0.110 | 0.150 | +36.4% |
|
| 90 |
| CIDEr | 0.031 | 0.035 | +12.9% |
|
| 91 |
-
| CAPTURE | 0.546 | 0.553 | +1.3% |
|
| 92 |
| ROUGE-L | 0.275 | 0.294 | +6.9% |
|
| 93 |
|
| 94 |
These results demonstrate that our LoRA adapter enhances the image captioning capabilities of the Florence-2 base model, particularly in generating more detailed and accurate captions.
|
|
|
|
| 85 |
|
| 86 |
| Metric | Base Model | Adapted Model | Improvement |
|
| 87 |
|---------|------------|-----------------------|-------------|
|
| 88 |
+
| CAPTURE | 0.546 | 0.553 | +1.3% |
|
| 89 |
| METEOR | 0.213 | 0.240 | +12.7% |
|
| 90 |
| BLEU | 0.110 | 0.150 | +36.4% |
|
| 91 |
| CIDEr | 0.031 | 0.035 | +12.9% |
|
|
|
|
| 92 |
| ROUGE-L | 0.275 | 0.294 | +6.9% |
|
| 93 |
|
| 94 |
These results demonstrate that our LoRA adapter enhances the image captioning capabilities of the Florence-2 base model, particularly in generating more detailed and accurate captions.
|