Call for Submissions

Workshop Scope

The workshop will address both foundational and emerging challenges in federated learning (FL), with a particular focus on its integration with multi-agent systems, embodied intelligence, and large-scale foundation models. Building on prior workshops (e.g., FedKDD@KDD 2025), this edition (FedKDD/FedMAS 2026) expands the scope toward federated multi-agent collaboration, distributed decision-making, and real-world AI systems that operate under privacy, communication, and autonomy constraints.

We aim to bridge federated learning with multi-agent systems (MAS), where multiple autonomous agents collaboratively learn and interact under decentralized and partially observable environments. This includes scenarios such as autonomous driving fleets, robotic swarms, edge intelligence, and embodied AI systems, where agents must jointly optimize perception, reasoning, and action while preserving data privacy.

We further emphasize federated large models (e.g., LLMs and multimodal foundation models), exploring how decentralized data and computation can support scalable, safe, and personalized model training and alignment. Submissions on emerging application domains such as federated generative AI, decentralized content creation, and AI-native platforms are highly encouraged.

(1) Scaling Laws and System Foundations. Understanding how federated learning systems scale across data, models, agents, and infrastructure:

Scaling laws for federated foundation models (LLMs, vision-language models, multimodal models);
Learning with massive numbers of agents/clients (e.g., million- or billion-scale MAS);
Cross-device and cross-silo FL under heterogeneous data, models, and communication constraints;
System-level co-design for computation, communication, and storage efficiency in large-scale FL;
Adaptive and hierarchical FL for dynamic multi-agent environments.

(2) Safety, Trustworthiness, and Alignment. Security, privacy, and alignment challenges in federated and multi-agent settings:

Privacy leakage and inference attacks in federated and agent-based learning systems;
Backdoor, poisoning, and adversarial attacks in decentralized multi-agent training;
Safety and alignment of federated large language and generative models;
Trust, robustness, and incentive mechanisms in open multi-agent ecosystems;
Governance, auditing, and compliance for federated AI systems.

(3) Federated Multi-Agent Systems (FedMAS). Learning, coordination, and decision-making across distributed intelligent agents:

Federated reinforcement learning and decentralized policy optimization;
Communication-efficient coordination and collaboration among agents;
Game-theoretic and mechanism design approaches for federated MAS;
Learning under partial observability and non-stationary environments;
Applications in autonomous driving, robotics, IoT, and edge intelligence.

(4) Federated Embodied Intelligence. Integrating perception, reasoning, and action in physical or simulated environments:

Federated learning for robotics and embodied agents;
Cross-agent transfer and generalization in embodied tasks;
Simulation-to-real transfer with privacy constraints;
Human-agent interaction and collaborative embodied intelligence;
Distributed learning for real-world sensorimotor systems.

(5) Federated Large Models and Generative AI. Distributed training and deployment of foundation models:

Privacy-preserving pre-training and fine-tuning of LLMs across decentralized data;
Federated alignment, personalization, and instruction tuning;
Multimodal and cross-modal federated learning (e.g., text-image-video);
Synthetic data generation, evaluation, and associated risks in FL;
Efficient training and synchronization of large-scale models.

(6) Graph Analytics and Data Mining. Bridging federated learning with graph-based and data mining techniques:

Federated graph learning and decentralized graph mining;
Scalable and communication-efficient graph analytics;
Privacy-preserving graph representation learning;
Applications in recommendation systems, knowledge graphs, and social networks;
Integration of graph structures into multi-agent and federated systems.

(7) AI-driven Science and Real-world Applications. Enabling secure, cross-institutional collaboration:

Biomedical and healthcare applications (e.g., genomics, drug discovery);
Climate science and environmental modeling with distributed data;
Scientific simulations with privacy-sensitive experimental data;
Industrial and enterprise applications of federated multi-agent systems.

We also encourage submissions on benchmarking, evaluation methodologies, and open challenges in federated and multi-agent learning. Discussions on ethical, societal, and economic implications of decentralized AI systems are particularly welcome.

Submission Guidelines

Format: We invite short technical papers - up to 5 pages including references and unlimited pages of appendix. All manuscripts should be submitted in a single PDF file including all content, figures, tables, and references, following the new Standard ACM Conference Proceedings Template. For LaTeX users: unzip acmart.zip, make, and use sample-sigconf.tex as a template; Additional information about formatting and style files is available online at ACM Proceedings Template. Additionally, papers must be in the two-column format, with the recommended setting for Latex file: \documentclass[sigconf, anonymous, review]{acmart}.

Camera ready: Accepted papers can have up to 5 pages and unlimited pages of references and appendix. Please use the new acmart.cls when preparing submissions.

Submission: Papers should be submitted at the openreview website.

Review: All papers will be double-blinded and peer-reviewed by at least 2 reviewers.

Presentation: While all accepted papers will be presented with posters, high-quality accepted papers will also have the opportunity to participate in the oral/spotlight presentation and win our Best Paper Award(s). We strongly encourage all accepted papers to be presented in person by at least one of the authors. However, we will also help accommodate remote presentations due to strong travel difficulties based on available facilities at the conference.

According to the policy of the KDD 2026 conference, the accepted papers will NOT be included in proceedings or any form of publication.

Important Dates

Submission site open: Mar 10, 2026
Paper submissions: May 28, 2026
Paper notifications: June 25, 2026
Early-bird registration due: TBD
Workshop date: August 9 8am-12pm (Korea time), 2026