update

index.html

@@ -139,7 +139,7 @@ <h2 class="subtitle has-text-centered">
 <section class="hero teaser">
   <div class="container is-max-desktop">
     <div class="hero-body">
-      <img src="static/images/overview.png" alt="Overview" style="width: 80%; height: auto;"/>
+      <img src="static/images/overview.png" alt="Overview" style="width: 80%; height: auto; display: block; margin: auto;"/>
       <h2 class="subtitle has-text-centered">
         <strong>TL;DR.</strong>
         RITUAL is a simple yet effective anti-hallucination approach for LVLMs. Our RITUAL method leverages basic im- age transformations (e.g., vertical and horizontal flips) to enhance LVLM accuracy without external models or training. By integrating transformed and original images, RITUAL significantly reduces hallucinations in both discriminative tasks and descriptive tasks. Using both versions together enables the model to refine predictions, reducing errors and boosting correct responses.
@@ -187,7 +187,7 @@ <h2 class="subtitle has-text-centered">
 <section class="hero teaser">
   <div class="container is-max-desktop">
     <div class="hero-body">
-      <img src="static/images/ritual+.png" alt="RITUAL+" style="width: 80%; height: auto;"/>
+      <img src="static/images/ritual+.png" alt="RITUAL+" style="width: 80%; height: auto; display: block; margin: auto;"/>
       <h2 class="subtitle has-text-centered">
         <strong>RITUAL+.</strong>
         In <strong>RITUAL</strong>, the original image V undergoes random transformations, generating a transformed image. In <strong>RITUAL+</strong>, the model evaluates various potential transformations and selects the most beneficial one to improve answer accuracy within the given context, further refining reliability. These transformed images serve as complementary inputs, enabling the model to incorporate multiple visual perspectives to reduce hallucinations.