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

Abstract

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 tra

Categories

Framework mappings

Suggested from the entry's categories.

@misc{oblakov2026hivetraceguardpro,
  title = {{HiveTraceGuard-Pro: A Compact Generative Guardrail for Prompt Injection, Jailbreaks, and Adversarial Obfuscation}},
  author = {Nikita Oblakov and Sabrina Sadiekh and Evgeniy Kokuykin},
  year = {2026},
  month = sep,
  eprint = {2609.01046},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.01046}
}