Aixin Sun

I'm a Professor in Computer Science at Nanyang Technological University currently working on information retrieval, recommender systems, text mining and natural language processing. My research focuses on developing algorithms and models for understanding and processing unstructured text data. I previously worked extensively on search and recommendation systems, collaborating with industry partners like PayPal on fraud detection and risk management. My work has covered a wide range of topics including named entity recognition, temporal information extraction, dialog systems, and representation learning. Currently, I'm particularly interested in large language models and their applications in information extraction and retrieval tasks. I'm exploring ways to improve the faithfulness and reliability of language models through methods like contrastive learning and knowledge injection. My research aims to develop more robust and interpretable NLP systems that can effectively handle real-world text processing challenge.

The above introduction is fully generated by AI

Publications

Long Context vs. RAG for LLMs: An Evaluation and Revisits

Xinze Li, Yixin Cao, Yubo Ma, Aixin Sun

ABS HTML PDF

Hesitation and Tolerance in Recommender Systems

Kuan Zou, Aixin Sun, Xuemeng Jiang, Yitong Ji, Hao Zhang, Jing Wang, Ruijie Guo

ABS HTML PDF

Highly Transparent Conductive Gallium-Doped Zinc Oxide Thin Films Grown by Reactive Plasma Deposition for Silicon Heterojunction Solar Cells

Xinliang Chen, Xiaofeng Wang, Bingquan Liang, Aixin Sun, Diannan Li, Zheng Wang, Liyuan Hu, Dekun Zhang, H. Ren, G. Hou, Ying Zhao, Xiaodan Zhang, M. Qu, Shi Yin, X. Ru, Miao Yang, Xixiang Xu

ACS Applied Electronic Materials 2024

Aixin Sun

Publications

Long Context vs. RAG for LLMs: An Evaluation and Revisits

Hesitation and Tolerance in Recommender Systems

Highly Transparent Conductive Gallium-Doped Zinc Oxide Thin Films Grown by Reactive Plasma Deposition for Silicon Heterojunction Solar Cells

Right Tool, Right Job: Recommendation for Repeat and Exploration Consumption in Food Delivery

Sketch: A Toolkit for Streamlining LLM Operations

Not All Videos Become Outdated: Short-Video Recommendation by Learning to Deconfound Release Interval Bias

Open-domain Implicit Format Control for Large Language Model Generation

The Crowd in MOOCs: A Study of Learning Patterns at Scale

Knowledge-Enhanced Conversational Recommendation via Transformer-Based Sequential Modeling

TVR-Ranking: A Dataset for Ranked Video Moment Retrieval with Imprecise Queries

52B to 1T: Lessons Learned via Tele-FLM Series

MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations

Cultural Value Differences of LLMs: Prompt, Language, and Model Size

Highly sensitive flexible dielectric-response-based taste sensor for generic liquid identification

Tele-FLM Technical Report

The geography of corporate fake news

Beyond Collaborative Filtering: A Relook at Task Formulation in Recommender Systems

Not all Layers of LLMs are Necessary during Inference

DP-GCN: Node Classification by Connectivity and Local Topology Structure on Real-World Network

Recommender for Its Purpose: Repeat and Exploration in Food Delivery Recommendations

SciAgent: Tool-augmented Language Models for Scientific Reasoning

Parameter-Efficient Conversational Recommender System as a Language Processing Task

Multi-scale 2D Temporal Map Diffusion Models for Natural Language Video Localization

The 1st Workshop on Recommendation with Generative Models

On Context Utilization in Summarization with Large Language Models

Towards Verifiable Generation: A Benchmark for Knowledge-aware Language Model Attribution

On Challenges of Evaluating Recommender Systems in an Offline Setting

An Image Dataset for Benchmarking Recommender Systems with Raw Pixels

FLM-101B: An Open LLM and How to Train It with $100K Budget

Our Model Achieves Excellent Performance on MovieLens: What Does It Mean?

A Preference Learning Decoupling Framework for User Cold-Start Recommendation

Research Explosion: More Effort to Climb onto Shoulders of the Giant

Dipping PLMs Sauce: Bridging Structure and Text for Effective Knowledge Graph Completion via Conditional Soft Prompting

MS-DETR: Natural Language Video Localization with Sampling Moment-Moment Interaction

Take a Break in the Middle: Investigating Subgoals towards Hierarchical Script Generation

Shall We Trust All Relational Tuples by Open Information Extraction? A Study on Speculation Detection

Retraining A Graph-based Recommender with Interests Disentanglement

Open Information Extraction via Chunks

Few-shot Event Detection: An Empirical Study and a Unified View

FreeLM: Fine-Tuning-Free Language Model

DiffuRec: A Diffusion Model for Sequential Recommendation

A user study on asking clarifying questions in web search

A Survey on Deep Learning for Named Entity Recognition : Extended Abstract

Large Language Model Is Not a Good Few-shot Information Extractor, but a Good Reranker for Hard Samples!

GCRE-GPT: A Generative Model for Comparative Relation Extraction

Natural language processing in urology: Automated extraction of clinical information from histopathology reports of uro-oncology procedures.

DOCoR: Document-level OpenIE with Coreference Resolution

Dataset versus reality: Understanding model performance from the perspective of information need

EDU-Capsule: aspect-based sentiment analysis at clause level

Syntactic Multi-view Learning for Open Information Extraction

Perplexity from PLM Is Unreliable for Evaluating Text Quality

Take a Fresh Look at Recommender Systems from an Evaluation Standpoint

CofeNet: Context and Former-Label Enhanced Net for Complicated Quotation Extraction

An Attribute-Driven Mirror Graph Network for Session-based Recommendation

Improving Conversational Recommender Systems via Transformer-based Sequential Modelling

Automatic Expert Selection for Multi-Scenario and Multi-Task Search

A Survey on Neural Open Information Extraction: Current Status and Future Directions

A Study on Transformer Configuration and Training Objective

Do Loyal Users Enjoy Better Recommendations?: Understanding Recommender Accuracy from a Time Perspective

Chat-Capsule: A Hierarchical Capsule for Dialog-level Emotion Analysis

Aspect-based Sentiment Analysis through EDU-level Attentions

Temporal Sentence Grounding in Videos: A Survey and Future Directions

Towards Reducing Manual Workload in Technology-Assisted Reviews: Estimating Ranking Performance

Mirror Matching: Document Matching Approach in Seed-driven Document Ranking for Medical Systematic Reviews

DP-GCN: Node Classification Based on Both Connectivity and Topology Structure Convolutions for Risky Seller Detection

Towards Debiasing Temporal Sentence Grounding in Video

Real-time dynamic network learning for location inference modelling and computing

Pre-training Graph Transformer with Multimodal Side Information for Recommendation

A Dual-Channel Framework for Sarcasm Recognition by Detecting Sentiment Conflict

SelfCF: A Simple Framework for Self-supervised Collaborative Filtering

Incremental Community Detection on Large Complex Attributed Network

Parallel Attention Network with Sequence Matching for Video Grounding

Video Corpus Moment Retrieval with Contrastive Learning

DocOIE: A Document-level Context-Aware Dataset for OpenIE

Generative Adversarial Reward Learning for Generalized Behavior Tendency Inference

Tweet-aware News Summarization with Dual-Attention Mechanism

Neural Named Entity Boundary Detection

Natural Language Video Localization: A Revisit in Span-Based Question Answering Framework