Offscript: Automated Auditing of Instruction Adherence in LLMs
This addresses the need for reliable instruction adherence in LLMs used for information seeking, offering an incremental improvement in auditing mechanisms.
The paper tackled the problem of evaluating whether LLMs follow user-provided behavioral instructions, presenting Offscript, an automated auditing tool that detected potential instruction deviations in 86.4% of conversations in a pilot study.
Large Language Models (LLMs) and generative search systems are increasingly used for information seeking by diverse populations with varying preferences for knowledge sourcing and presentation. While users can customize LLM behavior through custom instructions and behavioral prompts, no mechanism exists to evaluate whether these instructions are being followed effectively. We present Offscript, an automated auditing tool that efficiently identifies potential instruction following failures in LLMs. In a pilot study analyzing custom instructions sourced from Reddit, Offscript detected potential deviations from instructed behavior in 86.4% of conversations, 22.2% of which were confirmed as material violations through human review. Our findings suggest that automated auditing serves as a viable approach for evaluating compliance to behavioral instructions related to information seeking.