FedLNS: レイヤー正規化シグネチャモデリングを活用した連合LLMにおける敵対的操作の緩和
FedLNS: Leverage LayerNorm Signature Modeling to Mitigate Adversarial Manipulation in Federated LLMs
連合学習において、悪意のあるクライアントによる更新をサーバー側で軽量に検出するフレームワークFedLNSを提案。正規化層のパラメータ変化をシグネチャとして利用し、履歴を考慮した参照と比較することで、追加の通信なしに敵対的更新をスクリーニングする。
詳しい要約
1. どんなもの?
2. 先行研究と比べてどこがすごい?
3. 技術・手法の肝は?
4. どうやって有効だと検証した?
5. 議論はある?
6. 次に読むべき論文は?
※ AIが要旨から生成した要約です。正確性は原文をご確認ください。
著者: Kai Li, Jong-Ik Park, Carlee Joe-Wong, Wei Ni, Falko Dressler
分類: cs.LG, cs.CR, cs.DC
原文アブストラクト
Federated training enables language models to learn from distributed private text, but the server cannot directly verify the local supervision or optimization process that produces each client update. A malicious client can therefore train on corrupted targets, introduce incorrect context-token associations, and degrade the global model through repeated aggregation. Such degradation can also increase the risk of unreliable or hallucinatory generation. We propose Federated Learning with Normalization Signatures (FedLNS), a server-side framework for lightweight malicious-update screening. FedLNS represents each client update through changes in trainable normalization-layer parameters and screens suspicious updates against a robust, history-aware cross-client reference. Because the signatures are extracted at the server from the returned local models, FedLNS requires no additional client-to-server parameter or metadata exchange compared to standard federated learning (FL) methods. After screening, the retained full-model updates can be aggregated using standard FL or another compatible aggregation rule. FedLNS requires no raw client data, trusted server dataset, labeled attack examples, or separately trained detector. Experiments on GPT-style, BERT-style, and LLaMA-style models trained from scratch with 200 clients show that, under 40% population-level target manipulation, FedLNS achieves lower test perplexity than the strongest of six baselines for all three architectures under both IID (independently and identically distributed) and non-IID data partitions.
関連論文
- 連合学習のための鍵付き来歴透かしと相補的格子ベース安全集約連合学習/セキュリティ
- 垂直連合学習におけるバックドア脆弱性の理解:研究と実践のギャップ連合学習/セキュリティ
- SCRUB-FL: バックドアの忘却による表現の浄化とクレンジング連合学習/セキュリティ