Persona-Conditioned Adversarial Prompting (PCAP): Multi-Identity Red-Teaming for Enhanced Adversarial Prompt Discovery
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866916006452002816 |
|---|---|
| author | Morasso, Cristian Halimi, Anisa Hameed, Muhammad Zaid Leith, Douglas |
| author_facet | Morasso, Cristian Halimi, Anisa Hameed, Muhammad Zaid Leith, Douglas |
| contents | Existing automated red-teaming pipelines often miss attacks that depend on attacker identity, framing, or multi-turn tactics. This under-coverage underestimates real-world risk. We introduce Persona-Conditioned Adversarial Prompting (PCAP), which conditions adversarial search on attacker personas and strategy cards and runs parallel persona-conditioned beam searches to discover diverse, transferable jailbreaks. PCAP is orthogonal to the underlying search algorithm and substantially increases attack success rate (ASR) and prompt diversity (e.g., ASR on GPT-OSS~120B from $\approx58\% \rightarrow \approx97\%$), improving attack strategy coverage and diversity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_12565 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Persona-Conditioned Adversarial Prompting (PCAP): Multi-Identity Red-Teaming for Enhanced Adversarial Prompt Discovery Morasso, Cristian Halimi, Anisa Hameed, Muhammad Zaid Leith, Douglas Cryptography and Security Existing automated red-teaming pipelines often miss attacks that depend on attacker identity, framing, or multi-turn tactics. This under-coverage underestimates real-world risk. We introduce Persona-Conditioned Adversarial Prompting (PCAP), which conditions adversarial search on attacker personas and strategy cards and runs parallel persona-conditioned beam searches to discover diverse, transferable jailbreaks. PCAP is orthogonal to the underlying search algorithm and substantially increases attack success rate (ASR) and prompt diversity (e.g., ASR on GPT-OSS~120B from $\approx58\% \rightarrow \approx97\%$), improving attack strategy coverage and diversity. |
| title | Persona-Conditioned Adversarial Prompting (PCAP): Multi-Identity Red-Teaming for Enhanced Adversarial Prompt Discovery |
| topic | Cryptography and Security |
| url | https://arxiv.org/abs/2605.12565 |