Stefanos Koffas

Orcid: 0000-0001-6543-4801

According to our database1, Stefanos Koffas authored at least 30 papers between 2022 and 2026.

Collaborative distances:
  • Dijkstra number2 of four.
  • Erdős number3 of four.

Timeline

Legend:

Book  In proceedings  Article  PhD thesis  Dataset  Other 

Links

On csauthors.net:

Bibliography

2026
Backdoor Attacks on Decentralised Post-Training.
CoRR, April, 2026

Removing the Trigger, Not the Backdoor: Alternative Triggers and Latent Backdoors.
CoRR, March, 2026

Let's focus: Focused backdoor attack against federated transfer learning.
Neurocomputing, 2026

CatBack: Universal Backdoor Attacks on Tabular Data via Categorical Encoding.
Proceedings of the 33rd Annual Network and Distributed System Security Symposium, 2026

Dr. Jekyll and Mr. Hyde: Two Faces of LLMs.
Proceedings of the Joint National Conference on Cybersecurity (ITASEC & SERICS 2026), 2026

2025
SoK: The Last Line of Defense: On Backdoor Defense Evaluation.
CoRR, November, 2025

L<sup>2</sup> · M = C<sup>2</sup> Large Language Models Are Covert Channels.
Proceedings of the 2025 IEEE International Conference on Acoustics, 2025

Backdoor Attacks on Transformers for Tabular Data: An Empirical Study.
Proceedings of the Computer Security. ESORICS 2025 International Workshops, 2025

Towards Backdoor Stealthiness in Model Parameter Space.
Proceedings of the 2025 ACM SIGSAC Conference on Computer and Communications Security, 2025

2024
Toward Stealthy Backdoor Attacks Against Speech Recognition via Elements of Sound.
IEEE Trans. Inf. Forensics Secur., 2024

Unveiling the Threat: Investigating Distributed and Centralized Backdoor Attacks in Federated Graph Neural Networks.
DTRAP, 2024

L<sup>2</sup> · M = C<sup>2</sup> Large Language Models as Covert Channels... a Systematic Analysis.
CoRR, 2024

The SpongeNet Attack: Sponge Weight Poisoning of Deep Neural Networks.
CoRR, 2024

Beyond PhantomSponges: Enhancing Sponge Attack on Object Detection Models.
Proceedings of the 2024 ACM Workshop on Wireless Security and Machine Learning, 2024

Backdoors on Manifold Learning.
Proceedings of the 2024 ACM Workshop on Wireless Security and Machine Learning, 2024

BAN: Detecting Backdoors Activated by Adversarial Neuron Noise.
Proceedings of the Advances in Neural Information Processing Systems 37: Annual Conference on Neural Information Processing Systems 2024, 2024

ELMs Under Siege: A Study on Backdoor Attacks on Extreme Learning Machines.
Proceedings of the 2024 Workshop on Artificial Intelligence and Security, 2024

EmoBack: Backdoor Attacks Against Speaker Identification Using Emotional Prosody.
Proceedings of the 2024 Workshop on Artificial Intelligence and Security, 2024

2023
Backdoor Pony: Evaluating backdoor attacks and defenses in different domains.
SoftwareX, May, 2023

Tabdoor: Backdoor Vulnerabilities in Transformer-based Neural Networks for Tabular Data.
CoRR, 2023

Invisible Threats: Backdoor Attack in OCR Systems.
CoRR, 2023

BlindSage: Label Inference Attacks against Node-level Vertical Federated Graph Neural Networks.
CoRR, 2023

Towards Stealthy Backdoor Attacks against Speech Recognition via Elements of Sound.
CoRR, 2023

A Systematic Evaluation of Backdoor Trigger Characteristics in Image Classification.
CoRR, 2023

Going in Style: Audio Backdoors Through Stylistic Transformations.
Proceedings of the IEEE International Conference on Acoustics, 2023

Watermarking Graph Neural Networks based on Backdoor Attacks.
Proceedings of the 8th IEEE European Symposium on Security and Privacy, 2023

2022
Can You Hear It?: Backdoor Attacks via Ultrasonic Triggers.
Proceedings of the WiseML@WiSec 2022: Proceedings of the 2022 ACM Workshop on Wireless Security and Machine Learning, 2022

Dynamic Backdoors with Global Average Pooling.
Proceedings of the 4th IEEE International Conference on Artificial Intelligence Circuits and Systems, 2022

More is Better (Mostly): On the Backdoor Attacks in Federated Graph Neural Networks.
Proceedings of the Annual Computer Security Applications Conference, 2022

On the Effect of Clock Frequency on Voltage and Electromagnetic Fault Injection.
Proceedings of the Applied Cryptography and Network Security Workshops, 2022


  Loading...