Is this you? Claim this profile to correct it, add a bio and choose the work people see first.
Claim this profileWorks426 from public data
- Privacy Risks of General-Purpose Language Models311
This study presents the first systematic study on the privacy risks of 8 state-of-the-art language models with 4 diverse case studies and demonstrates the aforementioned privacy risks do exist and can impose practical threats to the application of general-purpose language models on sensitive data covering identity, genome, healthcare and location.
- Dual Encoding for Zero-Example Video Retrieval305
This paper takes a concept-free approach, proposing a dual deep encoding network that encodes videos and queries into powerful dense representations of their own and establishes a new state-of-the-art for zero-example video retrieval.
- 297
- Model-Reuse Attacks on Deep Learning Systems205
It is demonstrated that malicious primitive models pose immense threats to the security of ML systems, and analytical justification for the effectiveness of model-reuse attacks is provided, which points to the unprecedented complexity of today's primitive models.
- Smart Contract Vulnerability Detection: From Pure Neural Network to Interpretable Graph Feature and Expert Pattern Fusion178
This paper develops automatic tools to extract expert patterns from the source code and casts the code into a semantic graph to extract deep graph features and significantly outperforms state-of-the-art methods.
- Trojaning Language Models for Fun and Profit175
This work presents TrojanLM, a new class of trojaning attacks in which maliciously crafted LMs trigger host NLP systems to malfunction in a highly predictable manner, and provides analytical justification for the practicality of TrojanLM.
- SirenAttack: Generating Adversarial Audio for End-to-End Acoustic Systems159
SirenAttack is evaluated on a set of state-of-the-art deep learning-based acoustic systems (including speech command recognition, speaker recognition and sound event classification), with results showing the versatility, effectiveness, and stealthiness of SirenAttack.
- Backdoor Pre-trained Models Can Transfer to All157
A new approach to map the inputs containing triggers directly to a predefined output representation of the pre-trained NLP models, e.g., a preddefined output representation for the classification token in BERT, instead of a target label, which can introduce backdoor to a wide range of downstream tasks without any prior knowledge.
- 155
- VulSniper: Focus Your Attention to Shoot Fine-Grained Vulnerabilities147
This paper proposes VulSniper, which is designed to detect fine-grained vulnerabilities more effectively and achieves F1-scores of 80.6% and 73.3% on the two benchmark datasets respectively, which are significantly higher than those of the state-of-the-art methods.
- 144
- Deep Dual Consecutive Network for Human Pose Estimation143
A novel multi-frame human pose estimation framework, leveraging abundant temporal cues between video frames to facilitate keypoint detection is proposed, which ranks No.1 in the Multi-frame Person Pose Estimation Challenge on the large-scale benchmark datasets PoseTrack2017 and PoseTrack2018.
- Adversarial attacks against Windows PE malware detection: A survey of the state-of-the-art132
This paper conducts a comprehensive and systematic review to categorize the state-of-the-art adversarial attacks againstPE malware detection, as well as corresponding defenses to increase the robustness of Windows PE malware detection.
- 132
- SecGraph: a uniform and open-source evaluation system for graph data anonymization and de-anonymization117
This paper proposes and develops SecGraph, a uniform and open-source Secure Graph data sharing/publishing system that enables data owners to anonymize data by state-of-the-art anonymization techniques, measure the data's utility, and evaluate the data' vulnerability against modern De-Anonymization (DA) attacks.
- Structural Data De-anonymization112
This work quantitatively shows the conditions for perfect and (1-ε)-perfect structural data DA of 26 real world structural datasets, including Social Networks, Collaborations Networks, Communication Networks, Autonomous Systems, and Peer-to-Peer networks and designs a practical and novel single-phase cold start Optimization based DA (ODA) algorithm.
- Deep Graph Matching and Searching for Semantic Code Retrieval111
An end-to-end deep graph matching and searching (DGMS) model based on graph neural networks for the task of semantic code retrieval that significantly outperforms state-of-the-art baseline models by a large margin on both datasets.
- Copy, Right? A Testing Framework for Copyright Protection of Deep Learning Models107
A novel testing framework for deep learning copyright protection: DEEPJUDGE quantitatively tests the similarities between two deep learning models: a victim model and a suspect model, which leverages a diverse set of testing metrics and efficient test case generation algorithms to produce a chain of supporting evidence to help determine whether a suspects model is a copy of the victim model.
- Multilevel Graph Matching Networks for Deep Graph Similarity Learning97
A multilevel graph matching network (MGMN) framework for computing the graph similarity between any pair of graph-structured objects in an end-to-end fashion that consistently outperforms state-of-the-art baseline models on both the graph–graph classification and graph– graph regression tasks.
- Investigating Pose Representations and Motion Contexts Modeling for 3D Motion Prediction96
A novel RNN architecture termed AHMR (Attentive Hierarchical Motion Recurrent network) for motion prediction which simultaneously models local motion contexts and a global context is proposed and achieves much enhanced long-term prediction proficiency, such as retaining natural human-like motions over 50 seconds predictions.
- On Your Social Network De-anonymizablity: Quantification and Large Scale Evaluation with Seed Knowledge96
The overall structural information based de-anonymization attack is much more powerful than the seed knowledge-only based deanonymized attack, and even without any seed information, a social network can be perfectly or partially de-Anonymized.
- VeriFi: Towards Verifiable Federated Unlearning92
This work prompts the concept of verifiable federated unlearning and proposes VeriFi, a unified framework that allows systematic analysis of federated unlearning and quantification of its effect, with different combinations of various unlearning and verification methods.
- 88
- NeuronFair81
NeuronFair is a new DNN fairness testing framework that differs from previous work in several key aspects: inter-pretable - it quantitatively interprets DNNs' fairness violations for the biased decision; effective - it uses the interpretation results to guide the generation of more diverse instances in less time; generic - it can handle both structured and unstructured data.
- 77
Publication data from OpenAlex, with missing venues and authors filled in from Crossref; citation counts are the higher of OpenAlex and Semantic Scholar, last synced 2026-10-13. One-sentence summaries under some papers are written by Semantic Scholar’s model. Citation counts may be lower than on Google Scholar, which indexes more sources.
Report an error
Wrong papers, two people merged into one, or a profile that should not be here? Tell us and we will fix or hide it.