First International TEXT2SPARQL Challenge¤
Co-Located with Text2KG at ESWC25¤
Abstract¤
The TEXT2SPARQL Challenge is a competitive and collaborative benchmark designed to push the boundaries of natural language processing (NLP) and Semantic Web technologies, with a particular emphasis on Neurosymbolic AI approaches. This challenge focuses on translating natural language questions into SPARQL queries, a structured query language for interacting with data in Resource Description Framework (RDF) formats, which are often used in knowledge graphs. Neurosymbolic AI - a hybrid approach combining neural network-based models with symbolic reasoning techniques—plays a crucial role in this challenge by addressing both the interpretative power of neural networks and the logical precision of symbolic methods. By incorporating Neurosymbolic AI, participants can develop models that better capture complex linguistic structures and the logical syntax required for accurate SPARQL query generation. This integration not only enhances model performance but also opens up new possibilities for human-computer interaction, as it enables more intuitive and accessible interfaces for querying large-scale knowledge graphs. Success in the TEXT2SPARQL Challenge has significant implications, from improving knowledge graph accessibility to advancing the development of intelligent systems capable of nuanced reasoning, thus marking an important step forward in AI-driven data interaction.
Latest News¤
2025-05-12 - Second Test Dataset and missing Submission Guidelines Links published
In addition to the missing Submission Guideline Links, we released the DB25 Test Dataset today.
2025-05-08 - Program published
The program is out!
2025-05-08 - Endpoint Responses published
As a first part of the Challenge Results we publish the plain endpoint responses.
2025-05-06 - First Test Dataset released
We just released one of the test datasets: The The CK25 Corporate Knowledge Reference Dataset for Benchmarking Text 2 SPARQL Question Answering Approaches is available on eccenca/ck25-dataset.
2025-05-05 - Evaluation phase is over
Many thanks to all participants in the challenge. We received 12 endpoint registrations and have surveyed these endpoints with 250 questions across the two datasets over the past few days. The next step will be to analyze the results.
2025-04-22 - Challenge registration and evaluation dates extended
We thought you gonna be happy with a few more days to work, check out the new deadlines 
2025-04-17 - DBpedia Mirror available
Our friends from DBpedia shut down their download server for maintenance, so we provided and documented our own temporary mirror for the DBpedia dataset.
2025-04-17 - Evaluator Agent IP Address published
We published the IPs which will be used for fetching the answers in this issue.
2025-04-16 - TEXT2SPARQL Client Application published
We published text2sparql-client, the command line client which we will use to communicate with the TEXT2SPARQL endpoints of the participants. You can use this application to self-evaluate your endpoint.
2025-01-22 - Knowledge Graphs for Evaluation available
We selected and provided the Knowledge Graphs, which will be used for the evaluation.
Please have a look at the Challenge page.
2024-12-16 - Website online
text2sparql.aksw.org is online and can be managed over this repository.