Tao Xiang 向涛

Tao Xiang 向涛

M.Sc. Informatics

Technical University of Munich

Biography

Tao Xiang (Leon) boasts an impressive educational background, having earned both his Bachelor’s and Master’s degrees in Computer Science from the Technical University of Munich (TUM), which secured a commendable QS ranking of 37 in 2024. Following his academic achievements at TUM, Leon is currently looking for a Ph.D position and his dream university is the CityU.

Demonstrating his entrepreneurial acumen, Leon co-founded Memomind, an AI startup that harnesses cutting-edge technology to bring forth innovative solutions.

Leon’s linguistic skills are equally impressive, with proficiency in German, English, Cantonese, and Mandarin. This linguistic versatility undoubtedly sets him apart in the global tech community. His exceptional talents have not gone unnoticed, as the Hong Kong government has honored him with the “Top-Talent” designation, further highlighting not only his technical capabilities but also his significant contributions to the broader community.

Interests
  • Machine Learning
  • Deep Learning
  • Natural Language Processing
  • Conversational AI
  • Multimodal AI
Education
  • Ph.D. in NLP, 2023

    Singapore University of Technology and Design

  • M.Sc. Informatics, 2021

    Technical University of Munich

  • B.Sc. Informatics, 2018

    Technical University of Munich

Experience

 
 
 
 
 
Research Assistant
Mar 2023 – Present Germany
Advisor: Kathrin Sessler; Supervisor: Enkelejda Kasneci
 
 
 
 
 
SAP SE
Research Intern
Nov 2022 – Present Germany
Advisor: Anum Afzal; Supervisor: Florian Matthes

Selected Projects

*
Text2Music
Text2Music is a virtual composer that creates style-specific music based on user-defined prompts.
Text2Music
A Task Scheduler App
An app for task scheduler written in Python and implemented async IO.
A Task Scheduler App
TrendFlow
TrendFlow is an advanced framework that uses deep learning techniques to analyze research trends. This powerful framework offers a wide range of analytical capabilities, including literature clustering, trend generation, and trend summarization. With TrendFlow, you can gain insights into emerging research topics and stay up-to-date on the latest advancements in your field.
TrendFlow
Knowledge-based QA for Diagnosis
An QA system for diagnosis, based on knowledge graph.
Knowledge-based QA for Diagnosis
Towards Forever Memory Dialogue Systems
We explore to achieve forever memory dialogue systems based on knowledge graphs.

Selected Awards

Selected Publications

Check out my publications on ORCID.
(2023). PARL: A Dialog System Framework with Prompts as Actions for Reinforcement Learning. ICAART.

PDF Cite Code DOI