Author

Gerhard Wunder

3 papers indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Conference Jul 2026

Diffu Vader: Validity-Constrained Discrete Diffusion for Flow-Level NIDS Red Teaming

Machine-learning-based network intrusion detection systems (NIDS) are increasingly evaluated with synthetic adversarial traffic, yet reported evasion rates often do not distinguish between structurally infeasible flow records and feasible network behavior. In generative NIDS research, synthetic data is more often used...

Hamed Fard, Ilya Komarov, Gerhard Wunder · International Conference on... · 0 citations

#natural language process... Jun 2026

Robust Text Watermarking for Large Language Models via Dual Semantic Embeddings

Experimental results show that dual-embedding watermarking can offer state-of-the-art robustness, particularly against translation, while incurring relatively low computational overhead compared with other semantic schemes.

Jonas Schäfer, Cezary Pilaszewicz, Gerhard Wunder · arXiv.org · 0 citations

Jul 2026

Prompt Compression via Activation Aggregation

structure in the activation space of LLMs is revealed: (i) mid-layer representations transfer meaningfully to early layers, suggesting a degree of cross-layer compatibility in how information is encoded; (ii) a single activation vector encodes a quantifiable and recoverable amount of semantic information; (iii) a weig...

Thibaud Ardoin, Semira Einsele, Evis Bregu et al. · arXiv.org · 0 citations