Metraipolis › archive › dc_4bfe134bc7bd

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation — arXiv:2609.01046

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.01046
License (as archived)
CC0-1.0
Archived
2026-10-17T06:33:54.385Z
SHA-256 of the archived text
55926f9919557dff70cd5f533a1950fd5e0678450cc4b49782f98c7356deae89
Stable URL of this page
https://metraipolis.com/documento/dc_4bfe134bc7bd

How to cite

HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation — arXiv:2609.01046 — arXiv (en), https://arxiv.org/abs/2609.01046 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4bfe134bc7bd (Metraipolis archive dc_4bfe134bc7bd, sha256 55926f9919557dff70cd5f533a1950fd5e0678450cc4b49782f98c7356deae89)
BibTeX
@misc{dc4bfe134bc7bd,
  title = {HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation — arXiv:2609.01046},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.01046},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4bfe134bc7bd; sha256 55926f9919557dff70cd5f533a1950fd5e0678450cc4b49782f98c7356deae89},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.01046. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Production LLMs must handle inputs that attempt to override system instructions, bypass safety policies or elicit harmful responses. A common mitigation is a separate guardrail model. Existing reports, however, provide little evidence on Russian prompt injection or Russian surface obfuscation. We present HiveTraceGuard-Pro, a 0.6B generative guardrail LoRA-tuned from Qwen3-0.6B. It is trained on Russian and English and uses one binary scoring rule (safe/unsafe) for the final target turn. Its training corpus pairs harmful examples, where a counterpart exists, with benign examples from the same domain and applies eight obfuscation transforms to both labels. In one harness, we compare HiveTraceGuard-Pro with thirty-four other guards on nineteen benchmark groups, sixteen of which are public. Its aggregate key is 0.7432, behind 0.7641 and 0.7552 for the two higher-scoring guards. Over the sixteen public groups alone, its key is 0.7153 and four of the thirty-four other suite guards score higher. In a fifteen-model comparison, HiveTraceGuard-Pro has the highest clean Russian robustness combined-F1 (0.88) and Russian prompt-injection recall (0.999). Both results use Russian sets assembled by our team, and at least 27.1% of the prompt-injection set overlaps the training corpus. Its 14.3 ms median latency is the lowest among those fifteen models in that run. Across the suite, FPR is 0.268 and FNR is 0.156. All reported response results use a legacy standalone-reply serialization rather than the natural assistant-role path of the shipped chat template. We release the merged weights on Hugging Face under Apache-2.0. The corpus, evaluation sets and evaluation code remain internal.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-02 in cs.AI).
Categorie arXiv: Cryptography and Security (cs.CR) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.01046 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-17T06:15:12.855Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).