TextGraphs Workshop

A TextGraphs Workshop is an academic workshop series that focuses on textgraph-based algorithms and promotes research at the intersection of graph-based methods and NLP methods.

• Context:
  • It can (typically) explore the synergies between Graph Theory and Natural Language Processing by organizing sessions that focus on graph-based methods for various NLP tasks.
  • It can (often) serve as a venue for presenting new research on graph-based and graph-supported machine learning methods, including graph embeddings and graph-based neural networks.
  • It can range from focusing on fundamental graph algorithms to their applications in large-scale NLP tasks such as text summarization, information retrieval, and semantic relation extraction.
  • It can highlight advances in integrating knowledge graphs with large language models (LLMs) to improve their interpretability and factual accuracy.
  • It can promote the development of novel prompting strategies for LLMs, such as Chain-of-Thought and Graph-of-Thought prompting.
  • ...
• Example(s):
  • TextGraphs-1 Workshop (2006) introduced the concept of graph-based methods in NLP and was co-located with the ACL 2006 conference.
  • TextGraphs-2 Workshop (2007) continued the exploration of graph-based approaches and included topics like graph-based clustering and ranking methods.
  • TextGraphs-3 Workshop (2008) focused on graph-based representation learning and applications in semantic networks.
  • TextGraphs-4 Workshop (2009) included discussions on probabilistic graphical models and their applications in NLP.
  • TextGraphs-5 Workshop (2010) addressed the challenges of large-scale graph processing for web data.
  • TextGraphs-6 Workshop (2011) explored advanced graph-based machine learning techniques for NLP.
  • TextGraphs-7 Workshop (2012) emphasized graph-based methods for text mining and information retrieval.
  • TextGraphs-8 Workshop (2013) focused on the integration of graph theory with big data analytics in NLP.
  • TextGraphs-9 Workshop (2014) covered topics such as graph-based text summarization and entity linking.
  • TextGraphs-10 Workshop (2015) discussed the role of graphs in social network analysis and NLP.
  • TextGraphs-11 Workshop (2016) included research on graph-based neural networks and their applications in NLP.
  • TextGraphs-12 Workshop (2018) co-located with NAACL 2018, highlighted advances in graph embeddings and their use in NLP.
  • TextGraphs-13 Workshop (2019) featured discussions on graph-based methods for deep learning and semantic parsing.
  • TextGraphs-14 Workshop (2020) focused on large-scale graphs and their applications in knowledge acquisition and social networks.
  • TextGraphs-15 Workshop (2021) emphasized the integration of graph-based methods with neural models for enhanced NLP tasks.
  • TextGraphs-16 Workshop (2022) covered graph-based techniques for explainability and interpretability in NLP.
  • TextGraphs-17 Workshop (2024) aims to further explore the synergies between knowledge graphs and large language models to address issues such as interpretability and factual accuracy.
  • ...
• Counter-Example(s):
  • AKBC (Automated Knowledge Base Construction) Conference.
• See: Academic Workshop, Text Graphs.

---

---

References

• Program Webpage: http://www.textgraphs.org

2024

• Perplexity
  • The TextGraphs workshop series focuses on exploring the synergies between Graph Theory and Natural Language Processing (NLP). Specifically, the 17th edition of the TextGraphs workshop in 2024 aims to extend the focus on exploring the rising topics of large language models (LLMs) prompting from the unique perspective of Graph Theory.[1][2
  • Key Focus Areas
    • Knowledge Graphs Meet LLMs: Utilizing graph-based methods for reasoning over Knowledge Graphs (KGs) to overcome limitations of existing LLMs, such as lack of interpretability, factual knowledge, and hallucination problems. Conversely, incorporating LLM knowledge learned from large text collections to aid tasks like KG completion and graph representation learning.[1][2
    • Chain Prompting of LLMs: Developing advanced prompting schemes like Chain-of-Thought and Graph-of-Thought for enhancing language understanding and generation tasks using LLMs and other pre-trained models.[1
    • Learning from Structured Data: Building bridges between NLP and various structured data formats like relational databases, XML, JSON, RDF, etc.[1
    • Interpretability of NLP Systems: Adopting graph-based methods to improve the interpretability of NLP systems, which poses a fundamental challenge for practical applications.[1
  • The workshop aims to foster stronger connections between NLP and structured data, tackling key challenges inherent in each field by leveraging the synergies between Graph Theory and NLP methods.[1][2
  • Citations:

[1] https://sites.google.com/view/textgraphs2024
[2] https://www.aclweb.org/portal/content/17th-workshop-graph-based-natural-language-processing
[3] https://sites.google.com/view/textgraphs2022
[4] http://lit.eecs.umich.edu/textgraphs/ws16/ws16/index.html
[5] http://www.wikicfp.com/cfp/servlet/event.showcfp?copyownerid=114767&eventid=119374 

---