Moved blog to dedicated page
This commit is contained in:
Yan Lin
parent
7b37f017ba
commit
0f4437d78c
36 changed files with 397 additions and 1631 deletions
36
data.yaml
36
data.yaml
|
|
@ -5,6 +5,7 @@ primaryPublications:
|
|||
- "IEEE TKDE"
|
||||
- "2025"
|
||||
links:
|
||||
Paper: "https://ieeexplore.ieee.org/document/11004614"
|
||||
Preprint: "https://arxiv.org/abs/2402.07232"
|
||||
Code: "https://github.com/Logan-Lin/UVTM"
|
||||
|
||||
|
|
@ -115,7 +116,7 @@ secondaryPublications:
|
|||
authors: "Letian Gong, Shengnan Guo, <strong>Yan Lin</strong>, Yichen Liu, Erwen Zheng, Yiwei Shuang, Youfang Lin, Jilin Hu, Huaiyu Wan"
|
||||
tags:
|
||||
- "IEEE TKDE"
|
||||
- "2024"
|
||||
- "2024"
|
||||
links:
|
||||
Paper: "https://ieeexplore.ieee.org/document/10836764"
|
||||
|
||||
|
|
@ -181,13 +182,13 @@ secondaryPublications:
|
|||
Code: "https://github.com/Water2sea/WITRAN"
|
||||
|
||||
primaryProjects:
|
||||
- title: 'Research on <i>Prediction of User Travel Destination and Travel Time Based on Trajectory Representation Learning</i>'
|
||||
- title: "Research on <i>Prediction of User Travel Destination and Travel Time Based on Trajectory Representation Learning</i>"
|
||||
tags:
|
||||
- "Fundamental Research Funds for the Central Universities of China"
|
||||
desc: "Applies representation learning to trajectory data to transform original features into high-level information, improving the performance of downstream tasks such as travel time and destination prediction."
|
||||
links: {}
|
||||
|
||||
- title: 'Development of <i>OverleafCopilot - Empowering Academic Writing in Overleaf with Large Language Models</i>'
|
||||
- title: "Development of <i>OverleafCopilot - Empowering Academic Writing in Overleaf with Large Language Models</i>"
|
||||
tags:
|
||||
- "Personal Interest Project"
|
||||
desc: "This project aims to develop a Browser extension to seamlessly integrate Large Language Models (such as ChatGPT) into the popular online academic writing platform, Overleaf."
|
||||
|
|
@ -195,7 +196,7 @@ primaryProjects:
|
|||
Home: "https://www.overleafcopilot.com/"
|
||||
Install: "https://chromewebstore.google.com/detail/overleaf-copilot/eoadabdpninlhkkbhngoddfjianhlghb"
|
||||
|
||||
- title: 'Development of <i>PromptGenius - All-purpose prompts for LLMs</i>'
|
||||
- title: "Development of <i>PromptGenius - All-purpose prompts for LLMs</i>"
|
||||
tags:
|
||||
- "Personal Interest Project"
|
||||
desc: "This project focuses on developing a website that offers a wide range of prompt categories, enhancing the versatility of LLMs for various tasks and improving their output quality."
|
||||
|
|
@ -204,33 +205,33 @@ primaryProjects:
|
|||
Code: "https://github.com/wenhaomin/ChatGPT-PromptGenius"
|
||||
|
||||
secondaryProjects:
|
||||
- title: 'Research on <i>Inverse Design of Materials Using Diffusion Probabilistic Models</i>'
|
||||
- title: "Research on <i>Inverse Design of Materials Using Diffusion Probabilistic Models</i>"
|
||||
tags:
|
||||
- "Villum Foundation"
|
||||
desc: "This project focuses on developing diffusion probabilistic models to first understand the relationship between chemistry/structure and material properties, then enable the inverse design of new materials with specific properties. This project currently supports my postdoctoral research position."
|
||||
links: {}
|
||||
|
||||
- title: 'Research on <i>Pre-training Representation Learning Methods of Spatial-temporal Trajectory Data for Traffic Prediction</i>'
|
||||
- title: "Research on <i>Pre-training Representation Learning Methods of Spatial-temporal Trajectory Data for Traffic Prediction</i>"
|
||||
tags:
|
||||
- "National Natural Science Foundation of China"
|
||||
desc: "This project aims to propose pre-training representation learning methods for spatial-temporal trajectory data, modeling multiple features to improve traffic prediction tasks. It demonstrates how trajectory representation learning can enhance traffic data mining."
|
||||
links: {}
|
||||
|
||||
- title: 'Research on <i>Spatial-temporal Trajectory Generation and Representation Learning Methods for Sparsity Problems</i>'
|
||||
- title: "Research on <i>Spatial-temporal Trajectory Generation and Representation Learning Methods for Sparsity Problems</i>"
|
||||
tags:
|
||||
- "National Natural Science Foundation of China"
|
||||
desc: "This project explores how to generate high-quality spatial-temporal trajectory data and corresponding representations to address sparsity-related issues, thereby supporting a variety of downstream tasks."
|
||||
links: {}
|
||||
|
||||
presentations:
|
||||
- title: 'Self-supervised Learning of Trajectory Data'
|
||||
- title: "Self-supervised Learning of Trajectory Data"
|
||||
tags:
|
||||
- "Guest lecture"
|
||||
- "Aalborg University"
|
||||
links:
|
||||
Slides: "/assets/Self-supervised Learning of Trajectory Data.pdf"
|
||||
|
||||
- title: 'PLM4Traj: Leveraging Pre-trained Language Models for Cognizing Movement Patterns and Travel Purposes from Trajectories'
|
||||
- title: "PLM4Traj: Leveraging Pre-trained Language Models for Cognizing Movement Patterns and Travel Purposes from Trajectories"
|
||||
tags:
|
||||
- "Workshop presentation"
|
||||
- "KDD 2024"
|
||||
|
|
@ -238,21 +239,21 @@ presentations:
|
|||
Slides: "/assets/KDD_2024_Workshop_PLM4Traj.pdf"
|
||||
Paper: "https://arxiv.org/abs/2405.12459"
|
||||
|
||||
- title: 'Origin-Destination Travel Time Oracle for Map-based Services'
|
||||
- title: "Origin-Destination Travel Time Oracle for Map-based Services"
|
||||
tags:
|
||||
- "Paper Oral"
|
||||
- "SIGMOD 2024"
|
||||
links:
|
||||
Slides: "/assets/SIGMOD-Oral-PPT.pdf"
|
||||
|
||||
- title: 'Self-supervised Learning of Spatial-temporal Trajectories'
|
||||
- title: "Self-supervised Learning of Spatial-temporal Trajectories"
|
||||
tags:
|
||||
- "Tutorial"
|
||||
- "SpatialDI 2024"
|
||||
links:
|
||||
Slides: "/assets/Talk on SpatialDI 2024.pdf"
|
||||
|
||||
- title: 'Pre-training Context and Time Aware Location Embeddings from Spatial-Temporal Trajectories for User Next Location Prediction'
|
||||
- title: "Pre-training Context and Time Aware Location Embeddings from Spatial-Temporal Trajectories for User Next Location Prediction"
|
||||
tags:
|
||||
- "Paper Oral"
|
||||
- "AAAI 2021"
|
||||
|
|
@ -264,14 +265,3 @@ services:
|
|||
- "Secretary of IEEE (Denmark Section) Computer Society"
|
||||
- "Reviewer for journals including TIST, TII, and TVT"
|
||||
- "Member of program committees of KDD, ICLR, NeurIPS, AAAI, CVPR, ICCV, IJCAI, and WWW"
|
||||
|
||||
blogs:
|
||||
- title: "One Step Diffusion Models"
|
||||
badge: "May 2025"
|
||||
path: "one-step-diffusion-models"
|
||||
tldr: "Despite the promising performance of diffusion models on continuous modality generation, one deficiency that is holding them back is their requirement for multi-step denoising processes, which can be computationally expensive. In this article, we examine recent works that aim to build diffusion models capable of performing sampling in one or a few steps."
|
||||
|
||||
- title: "Multi-modal and Multi-function Transformers"
|
||||
badge: "April 2025"
|
||||
path: "multi-modal-transformer"
|
||||
tldr: "Multi-modal and multi-function Transformers enables a single architecture to process diverse data types such as language, images, and videos simultaneously. These models employ techniques like vector quantization and lookup-free quantization to map different modalities into a unified embedding space, allowing the Transformer to handle them within the same sequence. Beyond processing multiple data types, these architectures can also combine different functionalities-such as auto-regressive language generation and diffusion-based image creation-within a single model."
|
||||
Loading…
Add table
Add a link
Reference in a new issue