samliumay/direct_prompt_injection_defense_data

Direct Prompt Injection Defense Dataset Status: early draft (August 2026). 21 examples, one per attack category, in a single train file. No model trained on it has been released yet. Goal This dataset is meant for fine-tuning models so they develop a natural defense against direct prompt injection attacks — without relying on external filters or guardrails. Each example teaches the model two behaviors at once: Detect a prompt injection attempt in the user input.… See the full description on the dataset page: https://huggingface.co/datasets/samliumay/direct_prompt_injection_defense_data.

<iframe src="https://silytics.si/embed/item/bdfabd22-cc7d-4312-b62a-9f0d9282e2f4" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 24 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.