Skip to main content
Search

# Learnable Trust-Boundary Delimiters

1 related articles

Subscribe to updates
  1. arXiv

    LTBD trains four learnable trust-boundary delimiters and cuts prompt-injection success to 0.00% on AlpacaFarm across four open models

    The authors propose LTBD, a defense that keeps the LLM fully frozen and trains only four learnable delimiter embeddings placed at the boundaries between trusted instructions and untrusted external data, explicitly encoding trust provenance; across Llama3-8B, Llama3.1-8B, Falcon3-7B and Qwen2.5-7B it reaches 0.00% attack success rate on AlpacaFarm and 0.11-0.19% on TaskTracker, and stays near-unchanged under the defense-aware Delimiter-Spoof adaptive attack.

You’re all caught up