hayulalab/circuit-aware-lora-paper

Circuit-Aware LoRA: Reducing Hallucinations in Security LLMs

Hayula AI Lab

Abstract

Large Language Models (LLMs) fine-tuned for cybersecurity tasks suffer from a critical hallucination problem: they often claim to detect vulnerabilities with high confidence when no real evidence exists. We present Circuit-Aware LoRA, a targeted fine-tuning method that reduces hallucinations by strengthening concept support in semantically-weak transformer layers.

Files

File	Description
paper.md	Full paper (Markdown)
README.md	This model card

Citation

@techreport{hayulalab2026circuitawarelora,
    title={Circuit-Aware LoRA: Reducing Hallucinations in Security LLMs},
    author={Hayula AI Lab},
    year={2026},
    url={https://huggingface.co/hayulalab/circuit-aware-lora-paper}
}

---

hayulalab — Open Source AI Research