Position, Role or Label? A Controlled Study of Instruction Conflicts in LLM Agents
LLM agents take instructions from the operator, repository configuration, user, tool results and saved memory, which often conflict. Instruction-hierarchy benchmarks ask whether the higher-ranked source wins, but a source sets position, role and label at once. First, we build WhoWins, 1,080 code-scored scenarios in fiv...
Anish Laddha, Nitesh Pradhan, Gaurav Srivastava
· 2 citations