A5010138604- Advanced Malware Detection Techniques
- Topic Modeling
- IPv6, Mobility, Handover, Networks, Security
- Complex Network Analysis Techniques
- Advanced Graph Neural Networks
- Software Testing and Debugging Techniques
- Internet Traffic Analysis and Secure E-voting
- RFID technology advancements
- Advanced Authentication Protocols Security
- Wireless Body Area Networks
- Security and Verification in Computing
- Caching and Content Delivery
- Distributed systems and fault tolerance
- User Authentication and Security Systems
- Advanced MIMO Systems Optimization
- Digital Rights Management and Security
- Graph Theory and Algorithms
- Green IT and Sustainability
- Bluetooth and Wireless Communication Technologies
- Software Engineering Research
- Software Reliability and Analysis Research
- Blockchain Technology Applications and Security
- Vehicular Ad Hoc Networks (VANETs)
- Cryptographic Implementations and Security
- Natural Language Processing Techniques
Korea Advanced Institute of Science and Technology
2015-2022
Institute of Electrical and Electronics Engineers
2021
Regional Municipality of Niagara
2021
IEEE Computer Society
2021
One approach to assess the security of embedded IoT devices is applying dynamic analysis such as fuzz testing their firmware in scale. To this end, existing approaches aim provide an emulation environment that mimics behavior real hardware/peripherals. Nonetheless, practice, can emulate only a small fraction images. For example, Firmadyne, state-of-the-art tool, run 183 (16.28%) 1,124 wireless router/IP-camera images we collected from top eight manufacturers. Such low success rate caused by...
Binary code similarity analysis (BCSA) is widely used for diverse security applications such as plagiarism detection, software license violation and vulnerability discovery. Despite the surging research interest in BCSA, it significantly challenging to perform new this field several reasons. First, most existing approaches focus only on end results, namely, increasing success rate of BCSA by adopting uninterpretable machine learning. Moreover, they utilize their own benchmark sharing neither...
Long Term Evolution (LTE) is becoming the dominant cellular networking technology, shifting network away from its circuit-switched legacy towards a packet-switched that resembles Internet. To support voice calls over LTE network, operators have introduced Voice-over-LTE (VoLTE), which dramatically changes how are handled, both user equipment and infrastructure perspectives. We find this dramatic shift opens up number of new attack surfaces not been previously explored. call attention to...
An overwhelming number of true and false news stories are posted shared in social networks, users diffuse the based on multiple factors. Diffusion from one user to another depends not only stories' content genuineness but also alignment topical interests between users. In this paper, we propose a novel Bayesian nonparametric model that incorporates homogeneity as key component regulates similarity posting sharing users' interests. Our extends hierarchical Dirichlet process topics Gaussian...
Cellular basebands play a crucial role in mobile communication.However, it is significantly challenging to assess their security for several reasons.Manual analysis inevitable because of the obscurity and complexity baseband firmware; however, such requires repetitive efforts cover diverse models or versions.Automating also non-trivial firmware large contains numerous functions associated with complex cellular protocols.Therefore, existing approaches on are limited only couple versions...
A cellular network is a closed system, and each operator has built unique “walled garden” for their by combining different operation policies, configurations, implementation optimizations. Unfortunately, some of these combinations can induce performance degradation due to misconfiguration or unnecessary procedures. To detect such degradation, thorough understanding even the minor details standards operator-specific implementations important. However, it difficult problems, as control plane...
While humans naturally develop theory of mind (ToM), the capability to understand other people's mental states and beliefs, state-of-the-art large language models (LLMs) underperform on simple ToM benchmarks. We posit that we can extend our understanding LLMs' abilities by evaluating key human precursors -- perception inference perception-to-belief in LLMs. introduce two datasets, Percept-ToMi Percept-FANToM, evaluate these precursory inferences for LLMs annotating characters' perceptions...
Subgraphs are rich substructures in graphs, and their nodes edges can be partially observed real-world tasks. Under partial observation, existing node- or subgraph-level message-passing produces suboptimal representations. In this paper, we formulate a novel task of learning representations subgraphs. To solve problem, propose Partial Subgraph InfoMax (PSI) framework generalize models, including DGI, InfoGraph, MVGRL, GraphCL, into our framework. These models maximize the mutual information...
Additive compositionality of word embedding models has been studied from empirical and theoretical perspectives. Existing research on justifying additive existing requires a rather strong assumption uniform distribution. In this paper, we relax that propose more realistic conditions for proving compositionality, develop novel sub-word model satisfies under those conditions. We then empirically show our model’s improved semantic representation performance similarity noisy sentence similarity.
each year to highlight selected papers from a conference.The in this issue cover broad spectrum of applied