The Singapore ACM SIGKDD Symposium 2024 (“Symposium”) is an in-person event organized by the Singapore ACM SIGKDD Chapter. The Symposium aims to invite local researchers to introduce their recent research works in KDD areas, and connect local researchers with industry companies. The event was last held in 2023.
Date: Wednesday 31 July 2024
Venue: Singapore Management University, School of Computing and Information Systems
The Symposium will be a one-day event consisting of:
- Invited keynote talks from both academia and industry
- Contributed oral and poster presentations.
All presentations adopt a non-archival nature, i.e., there is no proceeding and they are not considered as formal publications. There are two tracks of contributed presentations, namely, Research Track, and Applied Data Science Track.
Awards
We are pleased to announce the winners of the best oral talk and best poster awards from the set of submissions to the Singapore ACM SIGKDD Symposium 2024.
BEST ORAL TALK AWARD
[SIGIR’24] Data-efficient Fine-tuning for LLM-based Recommendation
Presented by Xinyu Lin
BEST ORAL TALK AWARD RUNNER-UP
[KDD’24] A Learned Generalized Geodesic Distance Function-Based Approach for Node Feature Augmentation on Graphs
Presented by Amitoz Azad
BEST POSTER AWARD
[WWW’24] On the Feasibility of Simple Transformer for Dynamic Graph Modeling
Presented by Yuxia Wu
BEST POSTER AWARD RUNNER-UP
[NeurIPS’23] Estimating Propensity for Causality-based Recommendation without Exposure Data
Presented by Zhongzhou Liu
BEST POSTER AWARD RUNNER-UP
[KDD’24] LARP: Language Audio Relational Pre-training for Cold-Start Playlist Continuation
Presented by Xiaohao Liu
How to get there?
The event will kick off at 9.00am, 31 July 2024 at SOE/SCIS2 SR B1.2 at Singapore Management University. Please come 15 mins earlier to allow time for settling in. We look forward to seeing you!
Directions to the venue:
SOE/SCIS2 SR B1.2 is located at the basement concourse (B1), which is accessible to the public and requires NO registration with the security. Once you are in the basement concourse, follow signatures towards School of Economics / School of Computing and Information Systems 2.
By train: Alight at either Bencoolen (Downtown Line) or Bras Basah (Circle Line) stations.
- Bencoolen (Downtown Line), please kindly follow the signage towards Exit C, and then take a right turn at the basement concourse (B1).
- Bras Basah (Circle Line), please kindly follow the signage towards Exit B, and then take a right turn at the basement concourse (B1).
By taxi: The drop off point is at 90 Stamford Road, Singapore Management University, Singapore 178903. Then, take the escalator towards the basement concourse.
By driving: Parking is available at the Lee Kong Chian School of Business. Exit the car park via lift lobby A, which is connected to the basement concourse (B1).
Registration
Please join us and register via this link!
Registration fee per participant:
- Student registration: $15 before 25 July 2024
- Early bird registration: $25 before 25 July 2024
- Regular/On-site registration: $30 after 25 July 2024
The registration fee covers:
- Access to all talks and presentations
- Networking opportunities
- Coffee and tea breaks
- Lunch reception
Programme
| 9.00—9.15am | Opening | Hady W. Lauw, Chair, Singapore ACM SIGKDD Chapter |
| 9.15—10am | Invited Talk: Two perspectives about biases in recommender system: OoD and unfairness | Zhenhua Dong, Huawei Noah’s Ark Lab |
| 10.00—10.30am | Oral talks: Applied Data Science Session | |
| #1: [KDD’24] Class-incremental Learning for Time Series: Benchmark and Evaluation | Zhongzheng Qiao, NTU | |
| #2: [CIKM’23] Voucher Abuse Detection with Prompt-based Fine-tuning on Graph Neural Networks | Zhihao Wen, Huawei | |
| 10.30—11am | Coffee break | |
| 11.00—11.45am | Invited talk: Towards Inclusive AI: Advancing Multilingual Large Language Models | Wenxuan Zhang, DAMO Academy, Alibaba Group |
| 11.45am—12.30pm | Oral talks: Research Session 1 | |
| #3: [SIGIR’24] Data-efficient Fine-tuning for LLM-based Recommendation | Xinyu Lin, NUS | |
| #4: [KDD’24] Hierarchical Neural Constructive Solver for Real-world TSP Scenarios | Goh Yong Liang, NUS | |
| #5: [NeurIPS’23] LMC: Large Model Collaboration with Cross-assessment for Training-Free Open-Set Object Recognition | Haoxuan Qu, SUTD | |
| 12.30—2pm | Lunch reception | |
| 2—2.45pm | Invited Talk: Exit Strategy – Unravelling the Science, Technology and Business of a Successful AI Start-up | Shonali Krishnaswamy, CTO and Co-Founder, AiDA Technologies |
| 2.45—3.45pm | Oral talks: Research Session 2 | |
| #6: [NeurIPS’23] LD²: Scalable Heterophilous Graph Neural Network with Decoupled Embeddings | Ningyi Liao, NTU | |
| #7: [KDD’24] A Learned Generalized Geodesic Distance Function-Based Approach for Node Feature Augmentation on Graphs | Amitoz Azad, SMU | |
| #8: [KDD’24] How to Avoid Jumping to Conclusions: Measuring the Robustness of Outstanding Facts in Knowledge Graphs | XIAO Hanhua, SMU | |
| #9: [IJCAI’24] Federated Learning Optimization: A Comparative Study of Data and Model Exchange Strategies in Dynamic Networks | Alka Luqman, NTU | |
| 3.45—5.15pm | Coffee break and poster session | |
| 5.15—5.30pm | Award presentation and closing |
Invited Talks
Two perspectives about biases in recommender system: OoD and unfairness
Abstract: The goal of a recommender system is to deliver the right information to the right people. Most researches focus on optimizing accuracy, but this alone is insufficient for building a trustworthy system. Trustworthy recommender systems involve various aspects; in this talk, we focus on addressing biases from two perspectives: Out of Distribution (OoD) and unfairness. OoD biases arise from the gap between expected user preferences and observed user behaviors, leading to issues such as position bias, exposure bias, and trust bias. We will discuss solutions, such as causality-inspired methods and the information bottleneck method, to mitigate these biases. For unfairness, we will discuss the issue from the perspective of two stakeholders, i.e., users and content providers (CPs). To achieve user fairness, we propose counterfactual data augmentation methods to create fair data distributions. For CP fairness, we employ Max-Min fairness to improve exposure opportunities for weak CPs. Some of these de-biasing methods have been successfully implemented and verified in industrial products such as app market, news feeds, and advertisement systems. We will share practical experiences and online results.
Speaker: Zhenhua Dong is a technology expert and project manager of Huawei Noah’s ark lab. He is leading a research team focused on recommender system and causal inference. His team has launched significant improvements of recommender systems for several applications, such as news feeds, App store, instant services and advertising. With more than 40 applied patents and 60 research articles in TKDE, SIGIR, RecSys, KDD, WWW, AAAI, CIKM etc., he is known for research on recommender system, causal inference and counterfactual learning. He is also serving as PC or SPC members of SIGKDD, SIGIR, RecSys, WSDM, CIKM, industry chair of RecSys 2024. He translated the book “the singularity is near” into Chinese, named “奇点临近”. He received the Ph.D. degree from Nankai University in 2012. He was a visiting scholar at GroupLens lab in the University of Minnesota during 2010-2011.
Towards Inclusive AI: Advancing Multilingual Large Language Models
Abstract: Large Language Models (LLMs) are now widely used globally given their remarkable capabilities across diverse tasks. However, most of the focus has been on high-resource languages such as English, leaving their multilingual proficiency as an area still under exploration. This talk will delve into the multilingual aspects of LLMs, highlighting a novel benchmark, M3Exam, for evaluating their multilingual capabilities. We will also address the safety concerns arising from potential multilingual attacks. Additionally, methodologies for enhancing multilingual capabilities through an analysis of their internal processing mechanisms will be discussed. Finally, I will present the SeaLLMs project, a significant step forward in mitigating linguistic biases by introducing an LLM specialized in Southeast Asian languages. Compared to models with similar parameter sizes, SeaLLMs has achieved state-of-the-art performance across a diverse array of tasks such as world knowledge, mathematical reasoning, translation, and instruction following. Furthermore, it has been specifically enhanced to be more trustworthy.
Speaker: Dr. Wenxuan Zhang is currently a research scientist at Alibaba DAMO Academy. He received his Ph.D. degree from the Chinese University of Hong Kong under the supervision of Professor Wai Lam, and then joined Alibaba Singapore with the Ali Star award. His primary research areas are natural language processing and trustworthy AI. His research aims to advance NLP models that are inclusive, supporting diverse languages and cultures through multilingual language models, while also trustworthy by improving their safety and robustness. He has published over 30 papers in top-tier AI conferences and journals, including ICLR, NeurIPS, ACL, EMNLP, SIGIR, WWW, TOIS, and TKDE. He is a core tech lead of the SeaLLMs project (LLMs specialized for Southeast Asian languages), which has receivd significant community attention with over 120k downloads. He also regularly serves on the program committees of multiple leading conferences and journals and is the area chair of EMNLP 2024. He organized a tutorial at IJCAI 2023 and received the outstanding speaker award of MLNLP 2023.
Exit Strategy – Unravelling the Science, Technology and Business of a Successful AI Start-up
Abstract: AiDA (www.aidatech.io) is a multi-award winning Machine Learning (ML) start-up from Singapore, delivering AI-based Claims Processing technologies for Health Insurers in Singapore and the ASEAN region. AiDA was acquired by Amplify Health in 2023: https://www.businesstimes.com.sg/startups-tech/startups/aias-amplify-health-acquires-ai-startup-aida-plans-hiring-spree
AiDA’s marquee product SMART-CLAIMS is an AI-based solution for automating the processing of Health Insurance Claims and detecting/preventing Fraud, Waste, and Abuse. SMART-CLAIMS has been adopted by several major insurers in Singapore and increasingly in the ASEAN region. As of today, SMART-CLAIMS processes 75% of the Individual Life Health Claims in Singapore (over 1 million claims per annum) and has created a transformational impact for policyholders submitting claims. The process of getting reimbursed for hospitalisation and doctor visits which typically would take days can now happen in the order of minutes and payment within the hour for a majority of policyholders.
This talk will provide an overview of the technology innovations that underpin the SMART-CLAIMS solution including the underlying algorithms, architecture, and AI governance framework, which have been key to its successful adoption across multiple insurers and markets. The talk also will delve into the critical success factors for AI Start-Up including the science, technology and business model considerations – and the mistakes to avoid in a deep technology entrepreneurial journey.
Speaker: Dr Shonali Krishnaswamy is the CTO and Co-Founder of AiDA Technologies (www.aidatech.io).
Prior to co-founding AiDA, Shonali was Head of the Data Analytics Department at the Institute for Infocomm Research (I2R), which is part of the Singapore Government’s national R&D arm, the Agency for Science, Technology and Research (A*STAR). As Head of Data Analytics at I2R, Shonali led a research team of 70 Data Science researchers and engineers and focused on R&D innovation labs and collaboration projects across multiple industry sectors including Financial Services, Telecommunications, Healthcare and Aerospace.
She has also previously held academic/professorial appointments in Australia at Monash University and Swinburne University. Shonali is the recipient of several national and international awards including an ARC Australian Postdoctoral Fellowship, Monash University Vice-Chancellor’s Award for Excellence in Research by an Early Career Researcher, the Institute of Engineers Singapore Prestigious Engineering Award, the ASEAN Outstanding Engineering Award, and an IBM Innovation Award. In 2020, Shonali was recognised as one of Singapore’s Inaugural 100 Women in Technology by the Infocomm Media Development Authority (IMDA) and the Singapore Computer Society (SCS). More recently, in March 2024, Shonali was recognised as one of Asia’s Top 50 Women Technology Leaders.
Research Posters
| Title | Accepted Venue | Presenter |
| Graph principal flow network for conditional graph generation | WWW’24 | Zhanfeng Mo, NTU |
| Don’t Forget Your Reward Values: Language Model Alignment via Value-based Calibration | Work-in-progress | Mao Xin, NTU |
| LARP: Language Audio Relational Pre-training for Cold-Start Playlist Continuation | KDD’24 | Xiaohao Liu, NUS |
| Think Twice Before Trusting: Self-Detection for Large Language Models through Comprehensive Answer Reflection | Work-in-progress | Moxin Li, NUS |
| Selecting Comparative Sets of Reviews Across Multiple Items | EDBT’25 | Trung-Hoang Le, SMU |
| GliDe with a CaPE: A Low-Hassle Method to Accelerate Speculative Decoding | ICML’24 | Cunxiao Du, SMU |
| Augmenting Decision with Hypothesis in Reinforcement Learning | ICML’24 | NGUYEN Minh Quang, SMU |
| On the Feasibility of Simple Transformer for Dynamic Graph Modeling | WWW’24 | Yuxia Wu, SMU |
| Estimating Propensity for Causality-based Recommendation without Exposure Data | NeurIPS’23 | Zhongzhou Liu, SMU |
| Contrastive General Graph Matching with Adaptive Augmentation Sampling | IJCAI’24 | Bo Jianyuan, SMU |
| Generative Cross-Modal Retrieval: Memorizing Images in Multimodal Language Models for Retrieval and Beyond | ACL’24 | Yongqi Li, The Hong Kong Polytechnic University |
Call for Oral Presentation
Research Track
We are soliciting oral presentations for papers in the areas of knowledge discovery and data mining recently accepted by KDD 2024 as well as similar venues, such as (but not limited to) WSDM, SIGIR, IJCAI, ICML, CIKM, NeurIPS, and ACL.
Applied Data Science Track
We are soliciting oral presentations for papers in the areas of knowledge discovery and data mining recently accepted by the KDD 2024 Applied Data Science track. We also welcome papers from other related conferences in 2024 that meet the ADS scope.
Submission
Authors are required to:
- Submit the accepted version of their papers through this form (choose the relevant track) by 11 July 2024, 11.59pm SGT.
- At least one of the authors registers by the early-bird deadline upon notification of acceptance into the Symposium, and attend the Symposium to present the work.
Non-Proceeding
While this call focuses on papers already accepted in conferences, we also welcome submissions of work-in-progress papers in the areas of knowledge discovery and data mining. As the presentations are non-archival without any proceeding publication, it does not impact the future publication of these papers.
Presentation
There will be two modes of presentation. A subset of the accepted presentations will be presented orally. The others will be presented as posters. We will be awarding certificates for the Best Presentation Awards as well as Best Poster Awards based on inputs from the audience as well as the selection committee.ection committee.
Key Organizers
- General Chairs: Hady W. Lauw, Gao Cong
- Program Chairs (Research Track): Yuan Fang, Cheng Long
- Program Chairs (Applied Data Science Track): Yong Liu, Wee Siong Ng
- Treasurer: Roy Ka-Wei Lee
Singapore ACM SIGKDD Chapter 