From 86435c8084df9923ce461bea6e32c9e637686b0b Mon Sep 17 00:00:00 2001
From: zshyang <57812785+zshyang@users.noreply.github.com>
Date: Thu, 30 Nov 2023 21:46:39 -0800
Subject: [PATCH] Update index.html

---
 index.html | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/index.html b/index.html
index 2e35c5c..d1a3991 100644
--- a/index.html
+++ b/index.html
@@ -625,7 +625,7 @@ <h3>Framework Overview</h3>
           In (b), human motion is fed into the human motion encoder to produce a semantic-aware, subject-invariant latent code <em>Z</em>. 
           The CLIP feature of the subject-translated sentence and <em>Z</em> are concatenated together and passed into the animal decoders.
           During inference in (c), we generate animal motions based on human motion sequences sampled from generative models.
-          Details on the architecture, loss functions, and inference process are elaborated in paper.
+          Details on the architecture, loss functions, and inference process are elaborated in the paper.
         </p>
     </div>
 
@@ -636,7 +636,8 @@ <h2>Bibtex</h2>
         <p class="description">
 @misc{yang2023omnimotiongpt,
       title={OmniMotionGPT: Animal Motion Generation with Limited Data}, 
-      author={Zhangsihao Yang and Mingyuan Zhou and Mengyi Shan and Bingbing Wen and Ziwei Xuan and Mitch Hill and Junjie Bai and Guo-Jun Qi and Yalin Wang},
+      author={Zhangsihao Yang and Mingyuan Zhou and Mengyi Shan and Bingbing Wen and Ziwei Xuan and Mitch Hill  <br>
+		and Junjie Bai and Guo-Jun Qi and Yalin Wang},
       year={2023},
       eprint={2311.18303},
       archivePrefix={arXiv},