Deploying to gh-pages from @ 7b06c85 🚀

GGLAB-KU · May 20, 2024 · 174d632 · 174d632
1 parent fb051f7
commit 174d632
Show file tree

Hide file tree

Showing 32 changed files with 91 additions and 91 deletions.
diff --git a/404.html b/404.html
@@ -175,7 +175,7 @@ <h1 class="post-title">Page not found</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/feed.xml b/feed.xml
@@ -1,2 +1,2 @@
-<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en"><generator uri="https://jekyllrb.com/" version="4.3.3">Jekyll</generator><link href="https://gglab-ku.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="https://gglab-ku.github.io/" rel="alternate" type="text/html" hreflang="en" /><updated>2024-05-16T07:30:32+00:00</updated><id>https://gglab-ku.github.io/feed.xml</id><title type="html">blank</title><subtitle>Homepage for GGLab@Koc
+<?xml version="1.0" encoding="utf-8"?><feed xmlns="http://www.w3.org/2005/Atom" xml:lang="en"><generator uri="https://jekyllrb.com/" version="4.3.3">Jekyll</generator><link href="https://gglab-ku.github.io/feed.xml" rel="self" type="application/atom+xml" /><link href="https://gglab-ku.github.io/" rel="alternate" type="text/html" hreflang="en" /><updated>2024-05-20T11:23:03+00:00</updated><id>https://gglab-ku.github.io/feed.xml</id><title type="html">blank</title><subtitle>Homepage for GGLab@Koc
 </subtitle></feed>
diff --git a/index.html b/index.html
@@ -288,6 +288,23 @@ <h2><a href="/news/" style="color: inherit;">news</a></h2>
       <table class="table table-sm table-borderless">
 
 
+        <tr>
+          <th scope="row" style="width: 100px">May 2024</th>
+          <td>
+
+              Paper accepted to <a href="https://2024.aclweb.org/" rel="external nofollow noopener" target="_blank">Findings of ACL 2024</a>!
+
+<hr>
+Our paper entitled <a href="/assets/pdf/2403.03167.pdf">PARADISE: Evaluating Implicit Planning Skills of Language Models with Procedural Warnings and Tips Dataset</a> is accepted to <a href="https://2024.aclweb.org/" rel="external nofollow noopener" target="_blank">Findings of ACL 2024</a>! Check the <a href="https://github.com/GGLAB-KU/paradise" rel="external nofollow noopener" target="_blank">repo</a> for more details 📣
+
+<blockquote>
+  Abstract: Recently, there has been growing interest within the community regarding whether large language models are capable of planning or executing plans. However, most prior studies use LLMs to generate high-level plans for simplified scenarios lacking linguistic complexity and domain diversity, limiting analysis of their planning abilities. These setups constrain evaluation methods (e.g., predefined action space), architectural choices (e.g., only generative models), and overlook the linguistic nuances essential for realistic analysis. To tackle this, we present PARADISE, an abductive reasoning task using Q\&amp;A format on practical procedural text sourced from wikiHow. It involves warning and tip inference tasks directly associated with goals, excluding intermediary steps, with the aim of testing the ability of the models to infer implicit knowledge of the plan solely from the given goal. Our experiments, utilizing fine-tuned language models and zero-shot prompting, reveal the effectiveness of task-specific small models over large language models in most scenarios. Despite advancements, all models fall short of human performance. Notably, our analysis uncovers intriguing insights, such as variations in model behavior with dropped keywords, struggles of BERT-family and GPT-4 with physical and abstract goals, and the proposed tasks offering valuable prior knowledge for other unseen procedural tasks. The PARADISE dataset and associated resources are publicly available for further research exploration with this https URL.
+</blockquote>
+
+
+          </td>
+        </tr>
+
         <tr>
           <th scope="row" style="width: 100px">Apr 2024</th>
           <td>
@@ -339,23 +356,6 @@ <h2><a href="/news/" style="color: inherit;">news</a></h2>
           </td>
         </tr>
 
-        <tr>
-          <th scope="row" style="width: 100px">Mar 2024</th>
-          <td>
-
-              Our new paper is available on <a href="https://arxiv.org/" rel="external nofollow noopener" target="_blank">arXiv</a>!
-
-<hr>
-Our paper entitled <a href="/assets/pdf/2403.03167.pdf">PARADISE: Evaluating Implicit Planning Skills of Language Models with Procedural Warnings and Tips Dataset</a> is available on <a href="https://arxiv.org/" rel="external nofollow noopener" target="_blank">arXiv</a>! Check the <a href="https://github.com/GGLAB-KU/paradise" rel="external nofollow noopener" target="_blank">repo</a> for more details 📣
-
-<blockquote>
-  Abstract: Recently, there has been growing interest within the community regarding whether large language models are capable of planning or executing plans. However, most prior studies use LLMs to generate high-level plans for simplified scenarios lacking linguistic complexity and domain diversity, limiting analysis of their planning abilities. These setups constrain evaluation methods (e.g., predefined action space), architectural choices (e.g., only generative models), and overlook the linguistic nuances essential for realistic analysis. To tackle this, we present PARADISE, an abductive reasoning task using Q\&amp;A format on practical procedural text sourced from wikiHow. It involves warning and tip inference tasks directly associated with goals, excluding intermediary steps, with the aim of testing the ability of the models to infer implicit knowledge of the plan solely from the given goal. Our experiments, utilizing fine-tuned language models and zero-shot prompting, reveal the effectiveness of task-specific small models over large language models in most scenarios. Despite advancements, all models fall short of human performance. Notably, our analysis uncovers intriguing insights, such as variations in model behavior with dropped keywords, struggles of BERT-family and GPT-4 with physical and abstract goals, and the proposed tasks offering valuable prior knowledge for other unseen procedural tasks. The PARADISE dataset and associated resources are publicly available for further research exploration with this https URL.
-</blockquote>
-
-
-          </td>
-        </tr>
-
         <tr>
           <th scope="row" style="width: 100px">Jan 2024</th>
           <td>
@@ -438,7 +438,7 @@ <h2><a href="/news/" style="color: inherit;">news</a></h2>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_andrew.html b/members/alumni_andrew.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_arda.html b/members/alumni_arda.html
@@ -208,7 +208,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_atakan.html b/members/alumni_atakan.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_farrin.html b/members/alumni_farrin.html
@@ -209,7 +209,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_muge.html b/members/alumni_muge.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_subha.html b/members/alumni_subha.html
@@ -205,7 +205,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/alumni_tilek.html b/members/alumni_tilek.html
@@ -208,7 +208,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/faculty_gg.html b/members/faculty_gg.html
@@ -208,7 +208,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/intern_durhasan.html b/members/intern_durhasan.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/msc_ali.html b/members/msc_ali.html
@@ -205,7 +205,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/msc_hulki.html b/members/msc_hulki.html
@@ -208,7 +208,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/msc_zeynel.html b/members/msc_zeynel.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/phd_abed.html b/members/phd_abed.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/phd_gurkan.html b/members/phd_gurkan.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/members/phd_tamta.html b/members/phd_tamta.html
@@ -204,7 +204,7 @@ <h1 class="post-title">
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/news/2023-07-12-paper-INLG23-accepted/index.html b/news/2023-07-12-paper-INLG23-accepted/index.html
@@ -191,7 +191,7 @@ <h1 class="post-title">Paper Inlg23 Accepted</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/news/2023-09-06-papers-AACL23-accepted/index.html b/news/2023-09-06-papers-AACL23-accepted/index.html
@@ -198,7 +198,7 @@ <h1 class="post-title">Papers Aacl23 Accepted</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/news/2024-01-18-paper-arxiv/index.html b/news/2024-01-18-paper-arxiv/index.html
@@ -191,7 +191,7 @@ <h1 class="post-title">Paper Arxiv</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/news/2024-03-06-paper-arxiv/index.html → .../2024-05-16-paper-ACL-findings/index.html b/news/2024-03-06-paper-arxiv/index.html → .../2024-05-16-paper-ACL-findings/index.html
@@ -10,7 +10,7 @@
     <meta charset="utf-8">
     <meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
     <meta http-equiv="X-UA-Compatible" content="IE=edge">
-    <title>Paper Arxiv | GGLab  </title>
+    <title>Paper Acl Findings | GGLab  </title>
     <meta name="author" content="GGLab  ">
     <meta name="description" content="Homepage for GGLab@Koc
 ">
@@ -39,7 +39,7 @@
     <link rel="shortcut icon" href="/assets/img/favicon144x144.png">
 
     <link rel="stylesheet" href="/assets/css/main.css">
-    <link rel="canonical" href="https://gglab-ku.github.io/news/2024-03-06-paper-arxiv/">
+    <link rel="canonical" href="https://gglab-ku.github.io/news/2024-05-16-paper-ACL-findings/">
 
     <!-- Dark Mode -->
 
@@ -158,8 +158,8 @@
 <div class="post">
 
   <header class="post-header">
-    <h1 class="post-title">Paper Arxiv</h1>
-    <p class="post-meta">March 6, 2024</p>
+    <h1 class="post-title">Paper Acl Findings</h1>
+    <p class="post-meta">May 16, 2024</p>
     <p class="post-tags">
       <a href="/blog/2024"> <i class="fas fa-calendar fa-sm"></i> 2024 </a>
 
@@ -169,10 +169,10 @@ <h1 class="post-title">Paper Arxiv</h1>
   <article class="post-content">
 
     <div id="markdown-content">
-      <p>Our new paper is available on <a href="https://arxiv.org/" rel="external nofollow noopener" target="_blank">arXiv</a>!</p>
+      <p>Paper accepted to <a href="https://2024.aclweb.org/" rel="external nofollow noopener" target="_blank">Findings of ACL 2024</a>!</p>
 
 <hr>
-<p>Our paper entitled <a href="/assets/pdf/2403.03167.pdf">PARADISE: Evaluating Implicit Planning Skills of Language Models with Procedural Warnings and Tips Dataset</a> is available on <a href="https://arxiv.org/" rel="external nofollow noopener" target="_blank">arXiv</a>! Check the <a href="https://github.com/GGLAB-KU/paradise" rel="external nofollow noopener" target="_blank">repo</a> for more details 📣</p>
+<p>Our paper entitled <a href="/assets/pdf/2403.03167.pdf">PARADISE: Evaluating Implicit Planning Skills of Language Models with Procedural Warnings and Tips Dataset</a> is accepted to <a href="https://2024.aclweb.org/" rel="external nofollow noopener" target="_blank">Findings of ACL 2024</a>! Check the <a href="https://github.com/GGLAB-KU/paradise" rel="external nofollow noopener" target="_blank">repo</a> for more details 📣</p>
 
 <blockquote>
   <p>Abstract: Recently, there has been growing interest within the community regarding whether large language models are capable of planning or executing plans. However, most prior studies use LLMs to generate high-level plans for simplified scenarios lacking linguistic complexity and domain diversity, limiting analysis of their planning abilities. These setups constrain evaluation methods (e.g., predefined action space), architectural choices (e.g., only generative models), and overlook the linguistic nuances essential for realistic analysis. To tackle this, we present PARADISE, an abductive reasoning task using Q\&amp;A format on practical procedural text sourced from wikiHow. It involves warning and tip inference tasks directly associated with goals, excluding intermediary steps, with the aim of testing the ability of the models to infer implicit knowledge of the plan solely from the given goal. Our experiments, utilizing fine-tuned language models and zero-shot prompting, reveal the effectiveness of task-specific small models over large language models in most scenarios. Despite advancements, all models fall short of human performance. Notably, our analysis uncovers intriguing insights, such as variations in model behavior with dropped keywords, struggles of BERT-family and GPT-4 with physical and abstract goals, and the proposed tasks offering valuable prior knowledge for other unseen procedural tasks. The PARADISE dataset and associated resources are publicly available for further research exploration with this https URL.</p>
@@ -191,7 +191,7 @@ <h1 class="post-title">Paper Arxiv</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>
 

diff --git a/news/TA-award-abed/index.html b/news/TA-award-abed/index.html
@@ -191,7 +191,7 @@ <h1 class="post-title">Ta Award Abed</h1>
     <footer class="fixed-bottom">
       <div class="container mt-0">
         © Copyright 2024 GGLab  . Photos from <a href="https://www.freepik.com/" target="_blank" rel="external nofollow noopener">Freepik</a>.
-Last updated: May 16, 2024.
+Last updated: May 20, 2024.
       </div>
     </footer>