China Conference on Knowledge Graph and Semantic Computing (CCKS 2025)

September 19–21, 2025 · Fuzhou

http://sigkg.cn/ccks2025/

1. Conference Overview

The China Conference on Knowledge Graph and Semantic Computing (CCKS) is organized by the Technical Committee on Language and Knowledge Computing of the Chinese Information Processing Society of China. It originated from the Chinese Knowledge Graph Symposium (CKGS) and the Chinese Semantic Web and Web Science Conference (CSWS). In 2016, the two conferences were merged. From 2016 to 2024, CCKS has been successfully held in Beijing, Chengdu, Tianjin, Hangzhou, Nanchang, Guangzhou (online), Qinhuangdao, Shenyang, and Chongqing.

CCKS has become the core academic conference in China in the fields of knowledge graph and semantic technology, attracting researchers and professionals from related areas such as knowledge representation and reasoning, natural language understanding and knowledge acquisition, graph data management and graph computing, and intelligent question answering.

CCKS 2025 will be held in Fuzhou from September 19 to 21, 2025. The theme of this year’s conference is “Large Model Agents and Knowledge Computing”, aiming to explore how to enhance the accuracy and reliability of large model agents in knowledge computing with the support of knowledge graphs. Key topics include knowledge representation, storage, mining, fusion, reasoning, explainability, ethics, knowledge graph-enhanced large models, and agent architectures. The conference seeks to guide the paradigm shift in knowledge computing technology in the era of large models and promote the development of knowledge-enhanced large model agents.

The agenda will include workshops, keynote speeches, frontier forums, industrial forums, young scholar forums, evaluations and competitions, paper presentations, posters, and system demonstrations. Renowned scholars from academia and professionals from industry will be invited to share the latest progress, trends, and practical experiences to promote collaboration between academia and industry.

2. Evaluation Tasks

CCKS 2025 will organize a series of evaluation competitions related to knowledge graphs, aiming to provide a platform for researchers to test their technologies, algorithms, and systems. Task organizers may choose their own platforms and evaluation methods.

In CCKS 2024, there were 11 evaluation tasks across 6 themes, proposed by 4 enterprises and 13 universities. Topics included knowledge editing for large models, zero-shot knowledge extraction, complex question answering on person-centric knowledge graphs, document-level event relation extraction and summarization, bilingual Text-to-SQL, and more. Over 2,400 teams and 5,300 participants joined the competitions. A total of 42 teams were awarded a combined RMB 240,000, with the top task award reaching RMB 50,000, demonstrating significant influence across both academia and industry.

We sincerely invite researchers, institutions, and enterprises to submit proposals for evaluation tasks for CCKS 2025.

Important Dates

  • Proposal Submission Deadline: April 16
  • Notification of Acceptance: April 23
  • Official Task Release: April 30
  • Registration Period: May 1 – August 1
  • Training & Validation Data Release: May 14
  • Test Data Release: August 1
  • Result Submission Deadline: August 8
  • Ranking Notification: August 15
  • Evaluation Paper Submission: September 1
  • Conference Dates (Presentations & Awards): September 19–21

Evaluation Chairs

  • Feiliang Ren, Northeastern University (renfeiliang@cse.neu.edu.cn)
  • Sheng Bi, Southeast University (bisheng@seu.edu.cn)

Please send your evaluation task proposals via email to the evaluation chairs. Each proposal should clearly describe the task objectives and the process for data preparation. Refer to the CCKS 2024 task description documents for format and content guidance (https://sigkg.cn/ccks-ijckg2024/evaluation/)

Suggested Topics for Evaluation Tasks (including but not limited to):

Knowledge Representation and Reasoning

  • Knowledge representation and ontology modeling
  • Knowledge representation learning
  • Ontology reuse and evolution
  • Ontology mapping, fusion, and alignment
  • Ontology evaluation
  • Knowledge reasoning
  • Knowledge base completion
  • Knowledge representation and reasoning powered by large models

Knowledge Acquisition and Knowledge Graph Construction

  • Open information extraction
  • Crowdsourced knowledge engineering and collaborative acquisition
  • Human-in-the-loop knowledge base construction
  • Knowledge mining with large models
  • Knowledge acquisition from Wikidata
  • Tools, languages, and systems for automated knowledge base construction
  • Supervised/unsupervised learning-based knowledge acquisition
  • Semi-supervised/distant supervision and text extraction

Linked Data, Knowledge Fusion, and Knowledge Graph Storage Management

  • Entity recognition, disambiguation, and linking
  • Terminology mapping and integration
  • Heterogeneous knowledge linking and integration
  • Ontology-based data integration
  • Knowledge fusion powered by large models
  • Knowledge querying and search
  • Elastic storage and distributed computing for knowledge
  • Graph databases
  • Knowledge querying and analysis with large models

Natural Language Understanding, Semantic Computing, and KG Mining

  • Text understanding
  • Machine reading comprehension
  • Semantic similarity and relatedness computation
  • Synonym discovery
  • Natural language understanding with large models

Knowledge Graph Applications

  • Knowledge graph visualization
  • Semantic search
  • Knowledge-based question answering systems
  • Intelligent personal assistant systems
  • Knowledge-based semantic analysis of natural language/speech/image/video
  • Intelligent recommendation systems
  • Applications of large models and knowledge graphs

Knowledge Graph-Enhanced Large Models

  • Knowledge graph-enhanced large model pre-training
  • Knowledge graph-enhanced large model fine-tuning
  • Generation validation and explanation based on knowledge graphs
  • Knowledge editing for large models based on knowledge graphs
  • Retrieval-augmented generation with knowledge graphs

Knowledge Graph-Enhanced Agent Applications

  • Agent applications enhanced by knowledge graphs
  • Knowledge graph-enhanced question answering and dialogue for agents
  • Retrieval-based agent applications enhanced by knowledge graphs