Datasets:

ArXiv:
License:
LbhYqh commited on
Commit
a1fd99f
·
verified ·
1 Parent(s): ff05cae

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -2
README.md CHANGED
@@ -151,9 +151,9 @@ We use the following _statistical-based metrics_:
151
  - **Image Recall** measures the percentage of correct images in the multimodal answer relative to the total number of images in the ground truth, evaluating whether the answer effectively includes useful image information.
152
  It is computed as:
153
 
154
- \[
155
  \text{Image Recall} = \frac{\text{True Positives}}{\text{True Positives} + \text{False Negatives}}
156
- \]
157
 
158
  where False Negatives are the images in the ground truth that were omitted in the generated multimodal answer.
159
 
 
151
  - **Image Recall** measures the percentage of correct images in the multimodal answer relative to the total number of images in the ground truth, evaluating whether the answer effectively includes useful image information.
152
  It is computed as:
153
 
154
+ ![\[
155
  \text{Image Recall} = \frac{\text{True Positives}}{\text{True Positives} + \text{False Negatives}}
156
+ \]](https://cdn-uploads.huggingface.co/production/uploads/67571051d39ac252085797ca/dJCpcB_oeZDmjn0AAGbex.png)
157
 
158
  where False Negatives are the images in the ground truth that were omitted in the generated multimodal answer.
159