# Reverse Constitutional AI: A Framework for Controllable Toxic Data Generation via Probability-Clamped RLAIF

Preprint · Yuan Fang, Yiming Luo, Aimin Zhou, Fei Tan

- Archive id: arxiv:2604.17769
- Published: 2026-04-20
- arXiv: https://arxiv.org/abs/2604.17769
- Citations: 0
- Source URL: https://arxiv.org/abs/2604.17769
- PDF: https://arxiv.org/pdf/2604.17769
- Provenance: arxiv

_Abstract not reproduced: its license does not permit redistribution. See the source URL._

Topics: [AI Alignment & Safety](https://space-kitty.com/topics/alignment-safety)

---
Space Kitty Archive · https://space-kitty.com · API: https://space-kitty.com/api/v1 · MCP: https://space-kitty.com/mcp · Premium: $0.20/request via x402 (Base Sepolia (testnet)). Metadata: arXiv (CC0), OpenAlex (CC0), Europe PMC (CC BY/CC0 works), ClinicalTrials.gov.