Category: achievement

Our papers “Differentially Private Cross-Silo Recommendation from Implicit Feedback”, “Unlearning Isn’t Deletion: Investigating Reversibility of Machine Unlearning in LLMs”, and “On the Adversarial Robustness of Large Vision-Language Models under Visual Token Compression”, are accepted by International Conference on Machine Learning (ICML 2026). Congratulations to Xun, Xiaoyu and Xinwei! Our paper “LURE: Latent Space Unblocking for Multi-Concept Reawakening in Diffusion Models” is accepted by IJCAI’2026. Congratulations […]

Our paper “Hitchhiking on Inherent Randomness: Diffusion-based Graph Generation under Differential Privacy” is accepted by ACM CCS 2026. Congratulations to Renxuan!

Our papers “From Domains to Instances: Dual-Granularity Data Synthesis for LLM Unlearning” and “AudioStealer: Extracting Audio Prompts via Shapley Value-Guided Query Search” are accepted by Findings of the Association for Computational Linguistics (ACL). Congratulations to Xiaoyu and Yingbin!

Our paper “Answering Federated Range Queries with Local Differential Privacy” is accepted by ICDE 2026. Congratulations to Yuemin!

Our paper “SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning” is accepted by CVPR 2026. Congratulations to Tairan!

Our papers “How Green Is Your Login? A Cross-Protocol Benchmark of Authentication Energy & Latency” and “Decoding Web Memorization: A Semantic Membership Inference Attack on LLMs” are accepted by The Web Conference (WWW’26). Congratulations to Weizheng and Zi!

Our paper “BeeKeeper: Securing Cross-Technology Communication via Channel-Aware Dual-Binding” is accepted by INFOCOM 2026. Congratulations to Weizheng!

Our papers “United We Defend: Collaborative Membership Inference Defenses in Federated Learning” and “The Prompt Stealing Fallacy: Rethinking Metrics, Attacks, and Defenses” are accepted by 35th USENIX Security Symposium (USENIX Sec), 2026. Congratulations to Li and Zehang!

Our paper “WiFinger: Fingerprinting Noisy IoT Event Traffic Using Packet-level Sequence Matching” is accepted by NDSS 2026. Congratulations to Ronghua!

Our papers “‘Adversarial Signed Graph Learning with Differential Privacy” and “Communication-efficient Federated Graph Classification via Generative Diffusion Modeling” are accepted by SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2026. Congratulations to Haobin and Xiuling!