XL
Xixiang Lyu
cs.CRcs.LGcs.AIcs.CLcs.ITcs.SDcs.SImath.IT
On Valency
published · living versionsW_8gdt4cht·v1 · currentpublished
LoRA-Patch: Defending Against Backdoored LoRA Adapters via Directional Steering
with Shengyan Huo, Peihai Jiang, Yige Li
1 version
Preprints & journals
8 papers in the corpus · 2020–2026Natural Backdoor Attacks on Speech Recognition Models2607.15724v1 · Jinwen Xin, Xixiang Lyu, Jing Ma2026 · 19 citationsMachine Learning for Cyber Security, Lecture Notes in Computer Science, vol. 13655, pp. 597-610 (2023)
Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models2501.03272v1 · Peihai Jiang, Xixiang Lyu, Yige Li et al.2025 · 5 citationsarXiv
Reconstructive Neuron Pruning for Backdoor Defense2305.14876v2 · Yige Li, Xixiang Lyu, Xingjun Ma et al.2023 · 21 citationsarXiv
Anti-Backdoor Learning: Training Clean Models on Poisoned Data2110.11571v3 · Yige Li, Xixiang Lyu, Nodens Koren et al.2021 · 150 citationsarXiv
A Novel Variable K-Pseudonym Scheme Applied to 5G Anonymous Access Authentication2106.07158v1 · Dong Ma, Xixiang Lyu, Renpeng Zou2021 · 2 citationsarXiv
Privacy-preserving Federated Learning based on Multi-key Homomorphic Encryption2104.06824v1 · Jing Ma, Si-Ahmed Naas, Stephan Sigg et al.2021 · 440 citationsarXiv
Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks2101.05930v2 · Yige Li, Xixiang Lyu, Nodens Koren et al.2021 · 221 citationsarXiv
FLEAM: A Federated Learning Empowered Architecture to Mitigate DDoS in Industrial IoT2012.06150v2 · J. Li, L. Lyu, X. Liu et al.2020 · 209 citationsarXiv
Career total: 12 works. 8 are in this corpus.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.