Postdoctoral researcher. Run for life, drift for dream!
Denmark
Joined September 2018
- Tweets38
- Following353
- Followers74
- Likes260
Pinned Tweet
Glad to share our recently accepted #SIGMOD2027 paper. In this work, we introduce a cost-efficient RAG for entity matching that reduces computational overhead through blocking-based batch processing for retrieval and generation. #EntityMatching #GraphRAG arxiv.org/abs/2602.05708
Everyone hates reading AI-generated content, yet everyone uses it. Collaborator often lean on AI to save time in joint project but obscures responsibility. Some thoughts on how to work efficiently on research collaboration in the #AI era. #ai4science
machuangtao.github.io/blog/h…
Chuangtao Ma retweeted
our #SIGMOD2027 paper studies the role of KG-RAG (knowledge graph based retrieval-augmented generation) in domain-specific entity matching. Preprint: arxiv.org/abs/2602.05708. Lead by @DylanMa2018 (my postdoc) and Zeyu Zhang, in collaboration with Sebastian Schelter and @pgroth
I am incredibly proud to lead this work. In this work, we outline a vision for trustworthy and cost-efficient data integration with #AgenticRAG. Huge thanks to you for your guidance and support. @rijitK
Congratulations to my postdoc Chuangtao Ma @DylanMa2018 at Aalborg University for leading this effort!
We introduce blocking to RAG-based entity matching that reduces the overall cost of context retrieval and LLM inference.
priprint made available by my postdoc @DylanMa2018 "Cost-Efficient RAG for Entity Matching with LLMs: A Blocking-based Exploration" C. Ma, Z. Zhang, A. Khan, S. Schelter, @pgroth. We study: Can blocking help in LLM+RAG based entity matching? arxiv.org/pdf/2602.05708
Chuangtao Ma retweeted
sharing the preprint of our paper "SQL-to-Text Generation with Weighted-AST Few-Shot Prompting" - Sriom Chakrabarti, @DylanMa2018 Chuangtao Ma, Arijit Khan, and Sebastian Link arxiv.org/abs/2511.13907 to be presented at the Australasian Database Conference 2025 (#ADC25, #ADC2025)
The camera-ready version is now available on arxiv.org/abs/2505.20099
The brief introduction of this survey paper is available on machuangtao.github.io/LLM-KG…
Excited to share that our survey paper "Large Language Models Meet Knowledge Graphs for Question Answering: Synthesis and Opportunities" has been accepted to the #EMNLP2025 main conference. Preprint: arxiv.org/abs/2505.20099 Stay tuned for the camera-ready version.
Excited to share that our survey paper "Large Language Models Meet Knowledge Graphs for Question Answering: Synthesis and Opportunities" has been accepted to the #EMNLP2025 main conference. Preprint: arxiv.org/abs/2505.20099 Stay tuned for the camera-ready version.
Chuangtao Ma retweeted
Large Language Models Meet Knowledge Graphs for Question Answering: Synthesis and Opportunities
@DylanMa2018 et al. present a taxonomy categorizing methodologies for synthesizing LLMs and knowledge graphs for question answering.
📝arxiv.org/abs/2505.20099
In this survey, we systematically investigated SOTA advances in synthesizing #LLMs and #KGs for #QA and discuss how these approaches address the challenges of different complex QA.
Preprint of our survey "Large Language Models Meet Knowledge Graphs for Question Answering: Synthesis and Opportunities" - C. Ma @DylanMa2018, Y. Chen, T. Wu, A. Khan, & H. Wang arxiv.org/pdf/2505.20099 -an extension of our #EDBT2025 tutorial machuangtao.github.io/LLM-KG… #LLM #KG #QA
Chuangtao Ma retweeted
EDBT 2025 Tutorial: Unifying Large Language Models and Knowledge Graphs for Question Answering
Slides: machuangtao.github.io/upload…
Paper: openproceedings.org/2025/con…
Website: machuangtao.github.io/LLM-KG…
Materials: github.com/machuangtao/LLM-K…
@DylanMa2018 #EDBT2025 #EDBT25 #LLM #KnowledgeGraph
We're giving tutorial "Unifying Large Language Models and Knowledge Graphs for Question Answering: Recent Advances and Opportunities " -Chuangtao Ma @DylanMa2018, Yongrui Chen, Tianxing Wu, Arijit Khan, & Haofen Wang #EDBT2025 #EDBT25 doi.org/10.48786/edbt.2025.1… #LLM #KnowledgeGraph
Join us if you are attending EDBT/ICDT Conference 2025 and are interested in the intersection of large language models and knowledge graphs for QA!
Details are available at: machuangtao.github.io/LLM-KG…
We're giving tutorial "Unifying Large Language Models and Knowledge Graphs for Question Answering: Recent Advances and Opportunities " -Chuangtao Ma @DylanMa2018, Yongrui Chen, Tianxing Wu, Arijit Khan, & Haofen Wang #EDBT2025 #EDBT25 doi.org/10.48786/edbt.2025.1… #LLM #KnowledgeGraph
In this work, we introduce a novel KG-RAG, with vector-based, graph traversal-based subgraph retrievals, and ranking-based subgraph refinement, as well as a hybrid approach that identifies the most relevant subgraphs from external large KGs to augment LLMs for schema matching.
sharing the preprint of our work "Knowledge Graph-based Retrieval-Augmented Generation for Schema Matching" arxiv.org/abs/2501.08686 #LLM #GraphRAG #KGRAG #SchemaMatching
Chuangtao Ma retweeted
The website and call for papers of our workshop "LLM+KG: Data Management Opportunities in Unifying Large Language Models + Knowledge Graphs" at #VLDB2024 @pvldb @vldbconf is up: seucoin.github.io/workshop/l…
#VLDB24, #LargeLanguageModels, #KnowledgeGraphs #GraphDataManagement
Check out the paper collection on survey of knowledge graphs. It reached 100 starts on GitHub recently. #KnowledgeGraph github.com/machuangtao/KGs-S…
To reduce human intervention and improve the ability to handle complex matching tasks, we present a knowledge-enriched schema matching framework. ieeexplore.ieee.org/abstract…
From Pre-trained Word Embeddings to Pre-trained Language Models — Focus on BERT by Adrien Sieg towardsdatascience.com/from-…
