mlsanityData to 14 Oct 2026

SafeLM: Unified Privacy-Aware Optimization for Trustworthy Federated Large Language Models

arXiv 2604.16606 ↗First posted 17 Apr 2026cs.CR cs.LGcited by 0

On the map: LLM safety and security → Membership inference attacks

Abstract (arXiv)

Large language models (LLMs) are increasingly deployed in high-stakes domains, yet a unified treatment of their overlapping safety challenges remains lacking. We present SafeLM, a framework that jointly addresses four pillars of LLM safety: privacy, security, misinformation, and adversarial robustness. SafeLM combines federated training with gradient smartification and Paillier encryption for privacy, integrates defenses against training and inference-time attacks, employs contrastive grounding with calibrated decoding to reduce hallucinations, and introduces alignment-aware binarized aggregation to enhance robustness while maintaining bounded reconstruction quality. Across benchmarks on factuality, toxicity, and membership inference, SafeLM achieves 98.0% harmful content detection accuracy, reduces communication by 96.9%, and lowers gradient inversion PSNR from 31.7 dB to 15.1 dB. Ablations show that each component contributes independently, whereas their integration yields a strong privacy utility efficiency trade-off for deploying trustworthy LLMs.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Membership inference attacks, one of 12 directions in LLM safety and security. Builds on 25 of its 30 references that are in the map. First posted on arXiv 17 Apr 2026.

Builds on

25 of 30 references.

References · 25

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  5. 2021TruthfulQA: Measuring How Models Mimic Human Falsehoodscited by 2,975
  6. 2016Deep Learning with Differential Privacycited by 2,728
  7. 2020Extracting Training Data from Large Language Modelscited by 2,189
  8. 2022Language Models (Mostly) Know What They Knowcited by 1,856
  9. 2019Certified Adversarial Robustness via Randomized Smoothingcited by 1,340
  10. 2017Get To The Point: Summarization with Pointer-Generator Networkscited by 1,292
  11. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  12. 2017Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoningcited by 1,055
13 more
  1. 2019Deep Leakage from Gradientscited by 1,021
  2. 2020On Faithfulness and Factuality in Abstractive Summarizationcited by 900
  3. 2022Ignore Previous Prompt: Attack Techniques For Language Modelscited by 832
  4. 2019Adversarial NLI: A New Benchmark for Natural Language Understandingcited by 807
  5. 2020Inverting Gradients -- How easy is it to break privacy in federated learning?cited by 712
  6. 2018signSGD: Compressed Optimisation for Non-Convex Problemscited by 571
  7. 2020What Neural Networks Memorize and Why: Discovering the Long Tail via Influence Estimationcited by 497
  8. 2018Generating Natural Language Adversarial Examplescited by 424
  9. 2020iDLG: Improved Deep Leakage from Gradientscited by 382
  10. 2016Adversarial Training Methods for Semi-Supervised Text Classificationcited by 280
  11. 2017TernGrad: Ternary Gradients to Reduce Communication in Distributed Deep Learningcited by 256
  12. 2019Can You Really Backdoor Federated Learning?cited by 255
  13. 2021Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Modelscited by 232

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Privacy-Preserving Large Language Models: Mechanisms, Applications, and Future Directions
  2. 2024Privacy in Fine-tuning Large Language Models: Attacks, Defenses, and Future Directions
  3. 2025A Survey on Privacy Risks and Protection in Large Language Models
  4. 2023Gradient-Free Privacy Leakage in Federated Language Models through Selective Weight Tampering
  5. 2026Toward Efficient Membership Inference Attacks against Federated Large Language Models: A Projection Residual Approach
  6. 2024Analysis of Privacy Leakage in Federated Large Language Models
  7. 2026Towards Privacy-Preserving Large Language Model: Text-free Inference Through Alignment and Adaptation
  8. 2025A Survey: Towards Privacy and Security in Mobile Large Language Models