Pith. sign in

IHEval : Evaluating language models on following the instruction hierarchy

2 Pith papers cite this work. Polarity classification is still indexing.

2 Pith papers citing it

fields

cs.AI 1 cs.CL 1

years

2026 2

verdicts

UNVERDICTED 2

representative citing papers

Many-Tier Instruction Hierarchy in LLM Agents

cs.CL · 2026-04-10 · unverdicted · novelty 7.0

ManyIH and ManyIH-Bench address instruction conflicts in LLM agents with up to 12 privilege levels across 853 tasks, revealing frontier models achieve only ~40% accuracy.

citing papers explorer

Showing 2 of 2 citing papers.