mlsanityData to 7 Oct 2026

Responsible Federated LLMs via Safety Filtering and Constitutional AI

arXiv 2502.16691 ↗First posted 23 Feb 2025cs.CL cs.DC cs.MAcited by 1

On the map: LLM alignment / RLHF / preference optimisation → Safety alignment and jailbreak attacks

also in 1 more
Abstract (arXiv)

Recent research has increasingly focused on training large language models (LLMs) using federated learning, known as FedLLM. However, responsible AI (RAI), which aims to ensure safe and trustworthy responses, remains underexplored in this context. In FedLLM, client-side training data may contain harmful content, resulting in unsafe LLMs that can generate inappropriate responses. Aggregating such models into a global model and redistributing it to clients risks the widespread deployment of unsafe LLMs. To address this, we incorporate two well-established RAI techniques into FedLLM: safety filtering and constitutional AI. Our experiments show that these methods significantly improve LLM safety, achieving over 20% improvement on AdvBench.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Safety alignment and jailbreak attacks, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 16 of its 34 references that are in the map. First posted on arXiv 23 Feb 2025.

Builds on

16 of 34 references.

References · 16

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  3. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  5. 2019Parameter-Efficient Transfer Learning for NLPcited by 4,195
  6. 2018Federated Optimization in Heterogeneous Networkscited by 2,971
  7. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 2,950
  8. 2020Adaptive Federated Optimizationcited by 1,036
  9. 2021A General Language Assistant as a Laboratory for Alignmentcited by 1,023
  10. 2022Red Teaming Language Models with Language Modelscited by 983
  11. 2021Ethical and social risks of harm from Language Modelscited by 957
  12. 2019Measuring the Effects of Non-Identical Data Distribution for Federated Visual Classificationcited by 836
4 more
  1. 2023Towards Building the Federated GPT: Federated Instruction Tuningcited by 184
  2. 2023FederatedScope-LLM: A Comprehensive Package for Fine-tuning Large Language Models in Federated Learningcited by 146
  3. 2024OpenFedLLM: Training Large Language Models on Decentralized Private Data via Federated Learningcited by 124
  4. 2017Ethical Challenges in Data-Driven Dialogue Systemscited by 86

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025Multi-level Value Alignment in Agentic AI Systems: Survey and Perspectivescited by 4

Similar, not cited

Close in wording, no citation either way

  1. 2024Large Language Model Safety: A Holistic Survey
  2. 2025A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment
  3. 2024ShieldLM: Empowering LLMs as Aligned, Customizable and Explainable Safety Detectors
  4. 2024AEGIS: Online Adaptive AI Content Safety Moderation with Ensemble of LLM Experts
  5. 2024eFedLLM: Efficient LLM Inference Based on Federated Learning
  6. 2024Emerging Safety Attack and Defense in Federated Instruction Tuning of Large Language Models
  7. 2025Can Federated Learning Safeguard Private Data in LLM Training? Vulnerabilities, Attacks, and Defense Evaluation
  8. 2026MASCRDM: Multi-Agent System for Compliance Risk Detection and Mitigation in Training Process of Large Language Models