arXiv (All)
open access
DANGERLENS: Probing Dangerous Representations Hidden by Safety Alignment
Safety alignment teaches language models to refuse harmful requests, yet the internal accessibility of hazardous concepts can retain traces of earlier training. Can checkpoints with equally safe behavior differ in what remains accessible behind their refusals? In this work, we investigate this quest
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- ASCII Attack: Recontextualising Harmful Requests as Artistic Critique in Large Language Models#822607
- Bounded Reachability & Jailbreak Detection via Contraction-Constrained State Space Models#1187167
- SafeEvo: Deciphering the Safety Alignment Mechanism and Evolution in Language Models#1233968
- Language models judge war differently when tested for alignment#971968
- Beyond Unsafe Detection: Counterfactually Anchored Evidence Attribution for Multi-Turn LLM Safety Failures#1064317
Record · ID 1345188
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive