aiqbench analysis
This study applies JEPA-style predictive learning to derive high-fidelity embeddings from JA4 network fingerprints for protocol classification tasks like TLS and SSH identification. By matching latent predictions rather than regenerating raw inputs, the Transformer-based model achieves strong cosine similarity despite training on fragmented view families across disparate datasets. The approach demonstrates that self-supervised objectives effective in vision can be successfully transferred to compact network telemetry data without requiring complete sample overlap.
Applying JEPA-Style Predictive Learning to JA4-Derived Network Fingerprints
arXiv:2607.08465v2 Announce Type: replace Abstract: I-JEPA and V-JEPA learn by matching latent predictions to target encoder outputs rather than regenerating the original input, and this has worked well for images and video. We explore whether the same objective works for compact network fingerprints. We built JA4-JEPA, a Transformer-based model trained on JA4, JA4H, JA4S, and JA4X subfields drawn from JA4DB and CIC-IDS- 2017. The training data combines roughly 397K samples from both sources, though no single sample contains all four view families. We evaluated the learned representations with a frozen kNN probe on protocol-family classification across TLS, DNS, and SSH. On 39,416 heldout samples the model achieved a cosine similarity of 0.9899 and a kNN accuracy of 0.9220. These results indicate that JEPA-style predictive learning can produce useful embeddings from JA4-derived fingerprints, even with incomplete view overlap across sources. Keywords: JA4, network fingerprinting, JEPA, predictive representation learning, self-supervised learning
Abstract:I-JEPA and V-JEPA learn by matching latent predictions to target encoder outputs rather than regenerating the original input, and this has worked well for images and video. We explore whether the same objective works for compact network fingerprints. We built JA4-JEPA, a Transformer-based model trained on JA4, JA4H, JA4S, and JA4X subfields drawn from JA4DB and CIC-IDS- 2017. The training data combines roughly 397K samples from both sources, though no single sample contains all four view families. We evaluated the learned representations with a frozen kNN probe on protocol-family classification across TLS, DNS, and SSH. On 39,416 heldout samples the model achieved a cosine similarity of 0.9899 and a kNN accuracy of 0.9220. These results indicate that JEPA-style predictive learning can produce useful embeddings from JA4-derived fingerprints, even with incomplete view overlap across sources. Keywords: JA4, network fingerprinting, JEPA, predictive representation learning, self-supervised learning
Artificial Intelligence (cs.AI)
arXiv:2607.08465 [cs.AI]
(or arXiv:2607.08465v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2607.08465 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 236 words
Related coverage
J++ Lens: Jacobian Filtering Enables More Faithful Workspace Lenses
LessWrong · Oct 8, 2026
Dialogue with Eliezer Yudkowsky on Neural Networks
LessWrong · Oct 2, 2026
BioM-JEPA: joint-embedding prediction of graph-connected gene blocks in single cells
arXiv · Aug 7, 2026
PIRAMID: Progress and Plans
LessWrong · Jul 27, 2026