Merge

Author: Boltzmachine

app/layout.tsx

@@ -44,7 +44,7 @@ export default function RootLayout({
         <Providers themeProps={{ attribute: "class", defaultTheme: "dark" }}>
           <div className="relative flex flex-col h-screen">
             <Navbar />
-            <main className="container mx-auto max-w-full pt-4 px-6 flex-grow">
+            <main className="container mx-auto max-w-6xl pt-4 px-6 flex-grow">
               {children}
             </main>
             {/* <footer className="w-full flex items-center justify-center py-3">

app/publications/page.tsx

@@ -10,8 +10,9 @@ export default function PublicationsPage() {
         {publications.map((publication) => (
           <li key={publication.title}>
             [{publication.venue}] {publication.title}. {publication.authors}. 
-
-            <ApplicationTag />
+            <ApplicationTag tag={publication.tag} />
+            <p>{publication.abstract}</p>
+            <p>{publication.impact}</p>
           </li>
         ))}
       </ul>

components/tag.tsx

@@ -2,17 +2,26 @@
 
 import { Chip } from "@nextui-org/chip";
 import React from "react";
+import { Tag } from "@/config/publications"
 
-function Tag({ name, color }: { name: string; color: string }) {
-    return (
-        <Chip style={{ backgroundColor: color }}>
-            {name}
-        </Chip>
-    );
-}
 
-function ApplicationTag() {
-    return <Tag name="Application" color="#FFAA60" />
+function ApplicationTag({ tag } : { tag: Tag }) {
+    let name
+    let color
+    console.log(tag)
+    switch (tag) {
+        case Tag.Application:
+            name = "Application"
+            color = "#ffe119"
+            break
+        case Tag.TrustworthyAI:
+            name = "Trustworthy AI"
+            color = "#3cb444b"
+            break
+    }
+    return <><Chip style={{ backgroundColor: color }}>
+            {name}
+            </Chip></>
 }
 
 export { ApplicationTag };

config/people.ts

@@ -66,9 +66,9 @@ export const peopleList = {
         },
         {
             name: "Yangtian Zhang",
-            intro: "<Bio>",
-            homepage: "https://google.com",
-            photo: "/people/no_avt.png",
+            intro: "My research interests are focused on Generative Models, Graph Algorithms, and, more recently, Multi-Modal Foundation Models. Currently, I am exploring innovative solutions for real-world applications and scientific challenges.",
+            homepage: "https://zytzrh.github.io/",
+            photo: "people/yangtian.png",
         },
     ],
     "masters_and_undergrad": [

config/publications.ts

@@ -1,5 +1,6 @@
 export enum Tag {
-    GraphLearning = "Graph Learning",
+    GraphRepresentationLearning = "Graph Representation Learning",
+    TrustworthyAI = "Trustworthy AI",
     Application = "Application",
 }
 
@@ -22,5 +23,18 @@ export const publications = [
         paper: "https://arxiv.org/abs/2312.02203",
         code: "https://github.com/Graph-and-Geometric-Learning/HyBRiD",
         tag: Tag.Application,
+        abstract: "Traditional methods only focus on pariwise connectivity of brain regions. We proposed a new framework based on information bottleneck that learns high-order relationships of brain regions.",
+        impact: "The learned high-order relationships achieve SOTA performance on predictive tasks and are demonstrated significantly more effective than traditional pairwise methods.",
+    },
+    {
+        title: "D4Explainer: In-distribution explanations of graph neural network via discrete denoising diffusion",
+        authors: "Jialin Chen, Shirley Wu, Abhijit Gupta, Rex Ying",
+        venue: "NeurIPS 2023",
+        page: null,
+        paper: "https://proceedings.neurips.cc/paper_files/paper/2023/hash/f978c8f3b5f399cae464e85f72e28503-Abstract-Conference.html",
+        code: "https://github.com/Graph-and-Geometric-Learning/D4Explainer",
+        tag: Tag.TrustworthyAI,
+        abstract: "We propose D4Explainer, a novel approach that provides in-distribution GNN explanations for both counterfactual and model-level explanation scenarios.",
+        impact: "D4Explainer is the first unified framework that combines both counterfactual and model-level explanations. Empirical evaluations on synthetic and real-world datasets demonstrate the state-of-the-art performance achieved by D4Explainer in terms of explanation accuracy, faithfulness, diversity, and robustness."
     }
 ]