arXiv (All)
open access
IHDec: Divergence-Steered Contrastive Decoding for Securing Multi-Turn Instruction Hierarchies
Large Language Models (LLMs) often fail to maintain instruction hierarchies (IH) when processing multi-source inputs with varying role-level priorities, paradoxically adhering to lower-priority directives during conflicts. While existing defenses mitigate this issue, they are largely restricted to s
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
- Epistemic Policy Divergence in Multi-Turn LLM Contamination: A Protocol-Gradient Investigation#1129900
- Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference#1062509
- Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence in LLM Inference#1043092
- Admissibility Failure in Generative Representation Systems: A Practical Illustration from Human–AI Interaction#84076
- The Stack - Layered Constraint Architectures, Validator Diversity, and the Prevention of Recursive Drift#693520
Record · ID 1013604
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