FinLLM@IJCAI2026: International Symposium on Large Language Models for Financial Services In Conjunction with IJCAI 2026 Bremen, Germany, August 15-21, 2026 |
| Conference website | https://finllm.github.io/workshop/#/ |
| Submission link | https://easychair.org/conferences/?conf=finllmijcai2026 |
| Abstract registration deadline | May 1, 2026 |
| Submission deadline | May 1, 2026 |
International Symposium on Large Language Models for Financial Services (FinLLM@IJCAI2026) In Conjunction with IJCAI2026
Basic Information
Final Submission Deadline: 01 May, 2026 (23:59:59 AoE)
Notification Due: 01 June, 2026 (23:59:59 AoE)
Workshop Date: 15-17 August, 2026
Venue: Bremen, Germany
Conference Website: https://2026.ijcai.org/
Workshop Website: https://finllm.github.io/workshop/#/
Call for Papers
Building on the resounding success of our previous symposiums—FinLLM@IJCAI 2023 in Macau and the dual-city FinLLM@IJCAI 2025 in Montreal & Guangzhou—FinLLM 2026 returns as a core satellite event of IJCAI 2026. Large Language Models (LLMs) have consistently demonstrated transformative potential in the financial sector, automating complex tasks ranging from financial forecasting and risk management to sentiment analysis and insight extraction.
The landscape of financial technology is now experiencing a paradigm shift from conversational LLMs to highly autonomous AI agents. The explosive growth of open-source agentic frameworks, such as OpenClaw, is democratizing complex financial operations. This evolution is enabling the rise of the "One-Person Company" (OPC) organizational paradigm—where lean teams or solo founders can orchestrate multi-agent systems to autonomously handle quantitative analysis, algorithmic trading, and portfolio management at an enterprise scale.
As the deployment of these systems deepens, new critical frontiers have emerged. A major focus is the collaboration between large and small models, seeking the optimal balance between the robust reasoning of cloud-based foundation models and the latency/privacy benefits of local, smaller models. Furthermore, the economic efficiency of token consumption in high-frequency financial applications, the robust security of autonomous execution, and the complex landscape of bias diversity in global finance have become central to sustainable AI integration.
FinLLM 2026 aims to provide a global platform for academic researchers, industry practitioners, and policymakers to share novel research, exchange best practices, and debate the future of AI in finance. We invite original research contributions, position papers, and work-in-progress reports. The workshop topics include but are not limited to:
-
Architectures, Agentic Workflows & Model Collaboration:
-
Synergistic frameworks combining Large Foundation Models (Cloud) and Small Specialized Models (Edge/Local) in finance.
-
Autonomous AI agents (e.g., OpenClaw) and the "One-Person Company" (OPC) organizational paradigm.
-
Multi-agent systems (Manager/Worker frameworks) for complex financial decision-making and smart trading.
-
Multimodal financial data modeling and cross-lingual architectures for global finance.
-
Efficiency, Token Economics & Optimization:
-
Token economics and cost-effective deployment strategies for financial LLM applications.
-
Resource-efficient fine-tuning, inference, and distillation for high-frequency financial tasks.
-
Sustainable scaling of multi-agent interactions in enterprise environments.
-
Security, Trust & Ethics in Autonomous Finance:
-
Advanced security protocols and permission controls for autonomous financial AI agents.
-
Evaluating and managing bias diversity, fairness, and cultural alignment in cross-border financial predictions.
-
Real-time model validation, hallucination mitigation, and systemic stability in algorithmic trading.
-
Cross-border data privacy compliance and regulatory alignment for agentic AI.
Submission Instructions
Each submission can be up to 7 pages of contents plus up to 2 additional pages of references and acknowledgements. The submitted papers must be written in English and in PDF format according to the IJCAI'26 template (https://www.ijcai.org/authors_kit). All submitted papers will be under a single-blind peer review for their novelty, technical quality, and impact. The submissions can contain author details. Submission will be accepted via the Easychair submission website.
Based on the requirement from IJCAI'26, at least one author of each accepted paper must travel to the IJCAI venue in person. In addition, multiple submissions of the same paper to more than one IJCAI workshop are forbidden.
For enquiries, please email to: yanjiangpeng@efunds.com.cn
