Xixiang Lyu

4 papers · 2021–2025 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (11) 🐣 Hot Topic Early Bird 🏆 Grand Slam

Venues

AAAI (1) ICLR (1) ICML (1) NeurIPS (1)

Top co-authors

Yige Li (4) Xingjun Ma (3) Bo Li (3) Nodens Koren (3) Lingjuan Lyu (3) Yu-Gang Jiang (1) Jing Ma (1) Peihai Jiang (1)

Research topics

Artificial Intelligence (4) Core AI (4) Deep Learning (3) Machine Learning (3) Learning Types · Machine Learning (3) Application Areas (2) Learning Types · Deep Learning (2) Security & Privacy (1) Optimization & Theory (1) Privacy (1) Techniques (1)

Keywords

backdoor attack (4) deep neural network (2) adversarial learning (2) backdoor defense (2) machine unlearning (1) model fine-tuning (1) pre-trained language model (1) gradient ascent (1) trigger pattern (1) embedding analysis (1) neuron pruning (1) poisoned data (1) attention distillation (1) clean model (1) token unlearning (1) model compression (1) anti-backdoor learning (1) model security (1) neural network security (1) model unlearning (1)

Papers

Backdoor Token Unlearning: Exposing and Defending Backdoors in Pretrained Language Models AAAI 2025 Reconstructive Neuron Pruning for Backdoor Defense ICML 2023 Anti-Backdoor Learning: Training Clean Models on Poisoned Data NeurIPS 2021 Neural Attention Distillation: Erasing Backdoor Triggers from Deep Neural Networks ICLR 2021