Defend LLMs Through Self-Consciousness
Fuente:
arXiv
Salvato in:
| Autori principali: | , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909819949023232 |
|---|---|
| author | Huang, Boshi de Paula, Fabio Nonato |
| author_facet | Huang, Boshi de Paula, Fabio Nonato |
| contents | This paper introduces a novel self-consciousness defense mechanism for Large Language Models (LLMs) to combat prompt injection attacks. Unlike traditional approaches that rely on external classifiers, our method leverages the LLM's inherent reasoning capabilities to perform self-protection. We propose a framework that incorporates Meta-Cognitive and Arbitration Modules, enabling LLMs to evaluate and regulate their own outputs autonomously. Our approach is evaluated on seven state-of-the-art LLMs using two datasets: AdvBench and Prompt-Injection-Mixed-Techniques-2024. Experiment results demonstrate significant improvements in defense success rates across models and datasets, with some achieving perfect and near-perfect defense in Enhanced Mode. We also analyze the trade-off between defense success rate improvement and computational overhead. This self-consciousness method offers a lightweight, cost-effective solution for enhancing LLM ethics, particularly beneficial for GenAI use cases across various platforms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_02961 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Defend LLMs Through Self-Consciousness Huang, Boshi de Paula, Fabio Nonato Artificial Intelligence Computation and Language Cryptography and Security This paper introduces a novel self-consciousness defense mechanism for Large Language Models (LLMs) to combat prompt injection attacks. Unlike traditional approaches that rely on external classifiers, our method leverages the LLM's inherent reasoning capabilities to perform self-protection. We propose a framework that incorporates Meta-Cognitive and Arbitration Modules, enabling LLMs to evaluate and regulate their own outputs autonomously. Our approach is evaluated on seven state-of-the-art LLMs using two datasets: AdvBench and Prompt-Injection-Mixed-Techniques-2024. Experiment results demonstrate significant improvements in defense success rates across models and datasets, with some achieving perfect and near-perfect defense in Enhanced Mode. We also analyze the trade-off between defense success rate improvement and computational overhead. This self-consciousness method offers a lightweight, cost-effective solution for enhancing LLM ethics, particularly beneficial for GenAI use cases across various platforms. |
| title | Defend LLMs Through Self-Consciousness |
| topic | Artificial Intelligence Computation and Language Cryptography and Security |
| url | https://arxiv.org/abs/2508.02961 |