Institution profile

Southern University of Science and Technology

Academic institutionasia · cn
Official website
Research library1,154linked papers
Opportunities0open roles
Selected work

Representative Papers

Prompt Injection attack against LLM-integrated Applications

Jun 08, 2023arXiv.org

Prompt injection attacks pose an increasingly severe security threat to large language model (LLM) integrated applications, yet existing black-box attack methods suffer from limited practical efficacy. Method: This paper proposes HouYi—the first real-world-oriented, three-stage black-box prompt injection framework comprising pre-prompt injection, context-aware segmentation, and malicious payload delivery. HouYi uniquely enables automated triggering of high-impact consequences—including arbitrary LLM misuse and application-level prompt stealing—via black-box fuzzing, context-aware prompt engineering, and web-injection-inspired modeling. Contribution/Results: Evaluated through real-world penetration testing across 36 mainstream LLM applications, HouYi uncovered 31 critical vulnerabilities, independently confirmed by ten vendors—including Notion—with impact on millions of users. The work significantly advances LLM security practice by bridging the gap between theoretical attack models and deployable, scalable exploitation techniques.

536 citations38 influentialRead paper

Scalable Dynamic Embedding Size Search for Streaming Recommendation

Jul 22, 2024International Conference on Information and Knowledge Management

To address the unbounded embedding storage overhead caused by continuously growing users and items in streaming recommendation, this paper proposes Scalable Lightweight Embeddings (SCALL), which adaptively allocates embedding dimensions per entity under strict memory budget constraints. Its core contributions are threefold: (1) the first probabilistic distribution-based embedding size sampling mechanism, enabling precise memory control; (2) a reinforcement learning search paradigm with fixed-length state representations, supporting dynamic size generation for unseen entities; and (3) mean-pooling-based state modeling coupled with dynamic memory-aware optimization. Experiments on two public streaming recommendation datasets demonstrate that SCALL reduces storage overhead by up to 62% while improving recommendation accuracy—Recall@10 increases by 3.1–5.7%. To our knowledge, SCALL is the first method to jointly optimize memory controllability and model performance in streaming recommendation.

9 citations1 influentialRead paper

Eggly

Jun 12, 2023Proceedings of the ACM on Interactive Mobile Wearable and Ubiquitous Technologies

Clinical neurofeedback training (NFT) for children with autism spectrum disorder (ASD) faces challenges in accessibility, mobility, and engagement due to its reliance on stationary setups and lack of gamified, context-aware design. Method: We propose the first mobile augmented reality (AR) neurofeedback gaming framework tailored for ASD children, leveraging consumer-grade EEG headsets and tablets. The system integrates real-time theta/beta ratio decoding, contextual AR visual feedback, and adaptive difficulty adjustment within a lightweight Unity-based AR rendering pipeline (ARKit/ARCore) and closed-loop neurofeedback control. Contribution/Results: Two field studies—a single-session trial and a three-week multi-session intervention—demonstrated significantly improved attention and engagement; all participants completed the training. Qualitative feedback indicated high enjoyment and immersion, while quantitative analysis revealed a consistent trend toward normalized theta/beta ratios, validating the feasibility and efficacy of mobile AR-NFT for ASD intervention.

8 citationsRead paper

Instance-Conditioned Adaptation for Large-scale Generalization of Neural Combinatorial Optimization

May 03, 2024arXiv.org

Existing neural combinatorial optimization (NCO) methods exhibit poor generalization to large-scale routing problems—such as the Traveling Salesman Problem (TSP) and Capacitated Vehicle Routing Problem (CVRP)—limiting their applicability in real-world intelligent transportation systems. To address this, we propose Instance-Conditional Adaptive Mechanism (ICAM), a construction-based graph neural network model that achieves cross-scale adaptability via lightweight adapters conditioned on instance-specific embeddings. We further introduce a novel three-stage unsupervised reinforcement learning paradigm, enabling end-to-end training on instances ranging from 100 to 1,000 nodes without access to optimal solution labels. Experiments demonstrate that ICAM achieves state-of-the-art performance among construction-based NCO approaches on TSP and CVRP benchmarks, scales robustly up to 1,000 nodes, and delivers highly efficient inference—significantly outperforming existing methods.

6 citations1 influentialRead paper

Generation Enhances Understanding in Unified Multimodal Models via Multi-Representation Generation

Jan 29, 2026

This work proposes UniMRG, a unified multimodal framework that addresses the limited synergy between visual understanding and generation in existing models. By integrating auxiliary generative tasks—such as pixel reconstruction, depth estimation, and semantic segmentation—within a single architecture, UniMRG enables bidirectional enhancement between comprehension and synthesis. The method employs an architecture-agnostic post-training strategy, uniquely leveraging multitask generation to retroactively improve visual understanding capabilities. Experimental results demonstrate that UniMRG significantly advances performance in fine-grained perception, spatial relationship modeling, and hallucination suppression, while simultaneously enhancing generation quality. These findings validate the efficacy of the proposed understanding-generation co-evolution mechanism within a unified model.

2 citationsRead paper
Recent publications

Latest Papers