Call for Papers

We invite researchers, practitioners, and industry experts to submit their work to the Workshop on LLM-Based Agents for Intelligent Systems (LLAIS) at ECAI 2025. This workshop explores the transformative role of LLM-based agents in advancing intelligent systems across various domains, including retrieval-augmented generation (RAG), knowledge graphs, information retrieval, software automation, reproducibility in research, and next-generation recommender systems.

Despite their rapid advancements, LLMs still have untapped potential in these areas. LLAIS provides a collaborative platform to discuss innovations, challenges, and future opportunities in designing, developing, and deploying intelligent agents powered by LLMs.

Submission Guidelines

Submit your paper via EasyChair: https://easychair.org/conferences/?conf=llais1

We welcome contributions aligned with the themes of the workshop. Two types of submissions are invited:

• Regular Papers These should present original research, a critical survey, or applications to real-world scenarios. Length: Up to 12 pages (excluding references and appendices), using the CEURART style format.
• Short Papers These can present early-stage ideas, novel concepts, or promising technologies, including preliminary results or prototypes.Length: Up to 6 pages (excluding references and appendices), using the CEURART style format.

Additional materials (e.g. demos, data, supplementary resources) can be uploaded as a ZIP archive via EasyChair.

All submissions must be original and unpublished, and must not be under review elsewhere.

The review process is double-blind, and papers will be evaluated based on originality, quality, and relevance to the workshop topics.

Accepted papers will be published in an open-access CEUR-WS volume.

At least one author must attend the workshop in person to present the work.

Where relevant, authors are encouraged to bring and demonstrate a working prototype.

List of Topics

LLAIS 2025 invites contributions that advance the understanding, development, and deployment of LLM-based agents across various dynamic, real-world domains. We are particularly interested in research and practical applications of LLM-based agents across the following key domains:

Relevant themes and topics include, but are not limited to, the following:

• Knowledge Graph Enhancement
  • Automated Knowledge Graph Construction
  • Enhanced Knowledge Graph Completion
  • Improved Knowledge Graph Querying and Reasoning
• Developing Next-Generation Recommender Systems
  • Enhanced User Understanding
  • Creative Recommendation Generation
  • Conversational and Interactive Recommendations
• Software and AI Integration
  • Automating Key Development Processes, including: 
    • Code Generation, Automated Debugging, and Code Optimization
  • Improving Code Quality and Maintainability
  • Accelerating the Software Development Lifecycle
  • Addressing the challenges of deploying and scaling AI-powered software applications
• Revolutionizing Information Retrieval
  • Enhancing Query Understanding
  • Personalized Information Synthesis
  • Advanced Ranking and Filtering
  • Multimodal Information Retrieval
• Enhancing Reproducibility in Scientific Research
  • Automating Experiment Tracking and Logging
  • Enhancing Data Management and Sharing
  • Developing Standardized Evaluation Benchmarks
  • Improving Research Communication and Dissemination
• Advancing RAG Systems
  • Improving Retrieval Efficiency and Accuracy
  • Enhancing Contextual Understanding
  • Improving Generation Quality
• Evaluation and Ethical Considerations
  • Evaluating Effectiveness of LLM-based agents in different scenarios
  • Addressing Ethical Challenges including Hallucination and Bias

Important Dates

Paper submission deadline: June 16, 2025 (Anywhere on Earth, AOE) via EasyChair

Notification of acceptance: July 25, 2025

Camera-ready papers due: September 25, 2025

Workshop date: TBA (October 25 or 26, 2025)