# Safe-FCL: Continual Risk Awareness and Fine-Grained Safety Aggregation for Federated Continual Learning of LLMs

arXiv: https://arxiv.org/abs/2610.19968 · HTML: https://mlsanity.com/paper/2610.19968/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 29 of its 55 references that are in the map.
- First posted on arXiv 14 Oct 2026.

## Builds on (most cited first, 29 of 29)

- [Communication-Efficient Learning of Deep Networks from Decentralized Data](https://mlsanity.com/paper/1602.05629/) (2016)
- [Overcoming catastrophic forgetting in neural networks](https://mlsanity.com/paper/1612.00796/) (2016)
- [Learning without Forgetting](https://mlsanity.com/paper/1606.09282/) (2016)
- [iCaRL: Incremental Classifier and Representation Learning](https://mlsanity.com/paper/1611.07725/) (2016)
- [Gradient Episodic Memory for Continual Learning](https://mlsanity.com/paper/1706.08840/) (2017)
- [Dark Experience for General Continual Learning: a Strong, Simple Baseline](https://mlsanity.com/paper/2004.07211/) (2020)
- [BeaverTails: Towards Improved Safety Alignment of LLM via a Human-Preference Dataset](https://mlsanity.com/paper/2307.04657/) (2023)
- [WildChat: 1M ChatGPT Interaction Logs in the Wild](https://mlsanity.com/paper/2405.01470/) (2024)
- [FinQA: A Dataset of Numerical Reasoning over Financial Data](https://mlsanity.com/paper/2109.00122/) (2021)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- [LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset](https://mlsanity.com/paper/2309.11998/) (2023)
- [Orthogonal Gradient Descent for Continual Learning](https://mlsanity.com/paper/1910.07104/) (2019)
- [Gradient Projection Memory for Continual Learning](https://mlsanity.com/paper/2103.09762/) (2021)
- [FLTrust: Byzantine-robust Federated Learning via Trust Bootstrapping](https://mlsanity.com/paper/2012.13995/) (2020)
- [PKU-SafeRLHF: Towards Multi-Level Safety Alignment for LLMs with Human Preference](https://mlsanity.com/paper/2406.15513/) (2024)
- [Red-Teaming Large Language Models using Chain of Utterances for Safety-Alignment](https://mlsanity.com/paper/2308.09662/) (2023)
- [Federated Continual Learning with Weighted Inter-client Transfer](https://mlsanity.com/paper/2003.03196/) (2020)
- [Aegis2.0: A Diverse AI Safety Dataset and Risks Taxonomy for Alignment of LLM Guardrails](https://mlsanity.com/paper/2501.09004/) (2025)
- [OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learning](https://mlsanity.com/paper/2402.06954/) (2024)
- [FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning](https://mlsanity.com/paper/2505.08054/) (2025)
- [PolicyQA: A Reading Comprehension Dataset for Privacy Policies](https://mlsanity.com/paper/2010.02557/) (2020)
- [HarmAug: Effective Data Augmentation for Knowledge Distillation of Safety Guard Models](https://mlsanity.com/paper/2410.01524/) (2024)
- [Achieving Byzantine-Resilient Federated Learning via Layer-Adaptive Sparsified Model Aggregation](https://mlsanity.com/paper/2409.01435/) (2024)
- [TAME: A Trustworthy Test-Time Evolution of Agent Memory with Systematic Benchmarking](https://mlsanity.com/paper/2602.03224/) (2026)
- [Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchor](https://mlsanity.com/paper/2412.18355/) (2024)
- [SAIDO: Generalizable Detection of AI-Generated Images via Scene-Aware and Importance-Guided Dynamic Optimization in Continual Learning](https://mlsanity.com/paper/2512.00539/) (2025)
- [Face-D(^2)CL: Multi-Domain Synergistic Representation with Dual Continual Learning for Facial DeepFake Detection](https://mlsanity.com/paper/2604.08159/) (2026)
- [Adversarial Arena: Crowdsourcing Data Generation through Interactive Competition](https://mlsanity.com/paper/2604.17803/) (2026)
- [SafeCoEvo: Co-Evolving Safety Harnesses and Guards for LLM Agents at Test-Time](https://mlsanity.com/paper/2609.36580/) (2026)

## Abstract (arXiv)

Federated fine-tuning enables large language models (LLMs) to learn collaboratively from distributed client data, but malicious client updates can compromise global model safety. As federated tasks evolve, changing malicious update patterns challenge the ability of existing defenses to maintain reliable risk detection over time. To address this challenge, we propose Safe-FCL, a framework integrating continual risk awareness with fine-grained safety aggregation for federated continual learning of LLMs. Specifically, the Continual Risk Awareness (CRA) mechanism learns emerging risk patterns across sequential tasks and constrains detector updates using a protective subspace derived from historical features, mitigating forgetting of previously learned malicious patterns while adapting to new risks. To translate risk awareness into aggregation-level safety control, the Fine-Grained Safety Aggregation (FGSA) mechanism combines client-level risk assessments from CRA with module-level deviations from historical update behavior to adaptively regulate aggregation contributions, suppressing potentially malicious updates while minimizing unnecessary restrictions on benign updates. Extensive experiments demonstrate that Safe-FCL outperforms current SOTA methods, improving global-model Rule and MD-Judge scores in federated continual learning by 0.41% and 0.64%, respectively, while increasing average risk recognition accuracy by 2.01%.