Merge pull request #18 from HoldenHu/main

Merge From Hengchang

content/authors/hengchang/_index.md

@@ -0,0 +1,73 @@
+---
+# Display name
+title: Hengchang Hu
+
+# Full Name (for SEO)
+first_name: Hengchang
+last_name: Hu
+
+# Is this the primary user of the site?
+superuser: false
+
+# Role/position
+role: Graduate Students
+
+# Organizations/Affiliations
+organizations:
+  - name: National University of Singapore, School of Computing
+    url: 'http://www.comp.nus.edu.sg'
+
+# Short bio (displayed in user profile at end of posts)
+bio: PhD Candidate August 2019 Intake
+
+interests:
+  - Recommender System
+  - MultiModal
+
+education:
+  courses:
+    - course: PhD in Computer Science
+      institution: National University of Singapore
+      year: 2019-Now
+    - course: Bachelor of Software Engineering
+      institution: Sichuan University
+      year: 2015-2019
+
+# Social/Academic Networking
+# For available icons, see: https://docs.hugoblox.com/getting-started/page-builder/#icons
+#   For an email link, use "fas" icon pack, "envelope" icon, and a link in the
+#   form "mailto:[email protected]" or "#contact" for contact widget.
+social:
+  - icon: house
+    icon_pack: fas
+    link: https://holdenhu.github.io/
+  - icon: envelope
+    icon_pack: fas
+    link: '[email protected]'
+  - icon: google-scholar
+    icon_pack: ai
+    link: https://scholar.google.com.sg/citations?user=GOdo_yIAAAAJ&hl=zh-CN
+  - icon: github
+    icon_pack: fab
+    link: https://github.com/HoldenHu
+# Link to a PDF of your resume/CV from the About widget.
+# To enable, copy your resume/CV to `static/files/cv.pdf` and uncomment the lines below.
+# - icon: cv
+#   icon_pack: ai
+#   link: files/cv.pdf
+
+# Enter email to display Gravatar (if Gravatar enabled in Config)
+email: '[email protected]'
+
+# Highlight the author in author lists? (true/false)
+highlight_name: false
+
+# Organizational groups that you belong to (for People widget)
+#   Set this to `[]` or comment out if you are not using People widget.
+user_groups:
+  - Graduate Students
+#  - Researchers
+---
+
+Hengchang is currently a fifth year Ph.D. candidate in School of Computing, National University of Singapore.
+He is currently the member of Web Information Retrieval / Natural Language Processing Group (WING) and uner supervision of associate professor Dr. Min-yen Kan. His current research interest lies in graph neural network, multi-modal and recommendation system.

content/publication/hu-etal-2023-mmsr/cite.bib

@@ -0,0 +1,14 @@
+@inproceedings{hu-etal-2023-mmsr,
+ abstract = {In sequential recommendation, multi-modal information (e.g., text or image) can provide a more comprehensive view of an item’s profile. The optimal stage (early or late) to fuse modality features into item representations is still debated. We propose a graph-based approach (named MMSR) to fuse modality features in an adaptive order, enabling each modality to prioritize either its inherent sequential nature or its interplay with other modalities. MMSR represents each user’s history as a graph, where the modality features of each item in a user’s history sequence are denoted by cross-linked nodes. The edges between homogeneous nodes represent intra-modality sequential relationships, and the ones between heterogeneous nodes represent inter-modality interdependence relationships. During graph propagation, MMSR incorporates dual attention, differentiating homogeneous and heterogeneous neighbors. To adaptively assign nodes with distinct fusion orders, MMSR allows each node’s representation to be asynchronously updated through an update gate. In scenarios where modalities exhibit stronger sequential relationships, the update gate prioritizes updates among homogeneous nodes. Conversely, when the interdependent relationships between modalities are more pronounced, the update gate prioritizes updates among heterogeneous nodes. Consequently, MMSR establishes a fusion order that spans a spectrum from early to late modality fusion. In experiments across six datasets, MMSR consistently outperforms state-of-the-art models, and our graph propagation methods surpass other graph neural networks. Additionally, MMSR naturally manages missing modalities. The code is available at: https://github.com/HoldenHu/MMSR.},
+ address = {Online},
+ author = {Hu, Hengchang and
+Guo, Wei  and
+Liu, Yong  and
+Kan, Min-Yen},
+ booktitle = {Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM '23)},
+ doi = {10.1145/3583780.3614775},
+ month = {Oct},
+ title = {Adaptive Multi-Modalities Fusion in Sequential Recommendation Systems},
+ url = {https://dl.acm.org/doi/pdf/10.1145/3583780.3614775},
+ year = {2023}
+}

content/publication/hu-etal-2023-mmsr/index.md

@@ -0,0 +1,17 @@
+---
+title: Adaptive Multi-Modalities Fusion in Sequential Recommendation Systems
+authors:
+- Hengchang Hu
+- Wei Guo
+- Yong Liu
+- min
+date: '2023-10-21'
+publishDate: '2024-07-11T07:40:56.306034Z'
+publication_types:
+- paper-conference
+publication: '*Proceedings of the 32nd ACM International Conference on Information and Knowledge Management (CIKM '23)*'
+abstract: In sequential recommendation, multi-modal information (e.g., text or image) can provide a more comprehensive view of an item’s profile. The optimal stage (early or late) to fuse modality features into item representations is still debated. We propose a graph-based approach (named MMSR) to fuse modality features in an adaptive order, enabling each modality to prioritize either its inherent sequential nature or its interplay with other modalities. MMSR represents each user’s history as a graph, where the modality features of each item in a user’s history sequence are denoted by cross-linked nodes. The edges between homogeneous nodes represent intra-modality sequential relationships, and the ones between heterogeneous nodes represent inter-modality interdependence relationships. During graph propagation, MMSR incorporates dual attention, differentiating homogeneous and heterogeneous neighbors. To adaptively assign nodes with distinct fusion orders, MMSR allows each node’s representation to be asynchronously updated through an update gate. In scenarios where modalities exhibit stronger sequential relationships, the update gate prioritizes updates among homogeneous nodes. Conversely, when the interdependent relationships between modalities are more pronounced, the update gate prioritizes updates among heterogeneous nodes. Consequently, MMSR establishes a fusion order that spans a spectrum from early to late modality fusion. In experiments across six datasets, MMSR consistently outperforms state-of-the-art models, and our graph propagation methods surpass other graph neural networks. Additionally, MMSR naturally manages missing modalities. The code is available at: https://github.com/HoldenHu/MMSR.
+links:
+- name: URL
+  url: https://dl.acm.org/doi/pdf/10.1145/3583780.3614775
+---