Data-Model Co-Evolution: Growing Test Sets to Refine LLM Behavior

2025-10-14

AI Summary

Precisely encoding nuanced, domain-specific policies into prompt instructions for large language models (LLMs) remains a fundamental challenge. Method: This paper proposes a data–model co-evolution paradigm featuring an iterative, human-feedback-driven closed loop that simultaneously expands the test set dynamically and refines prompt instructions—integrated with structured human–AI collaboration, rationale-based behavioral attribution analysis, and iterative instruction evaluation. Contribution/Results: The approach mechanizes the concretization of ambiguous policies, systematically uncovers edge cases, and strengthens policy verifiability. A user study demonstrates that the framework significantly improves the systematicity and consistency of instruction refinement, while enhancing LLM adherence to localized policies and complex semantic rules.

Technology Category

• Natural Language Processing: Prompt Engineering / Prompting
• Humans and AI: Learning Human Values and Preferences
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

A long-standing challenge in machine learning has been the rigid separation between data work and model refinement, enforced by slow fine-tuning cycles. The rise of Large Language Models (LLMs) overcomes this historical barrier, allowing applications developers to instantly govern model behavior by editing prompt instructions. This shift enables a new paradigm: data-model co-evolution, where a living test set and a model's instructions evolve in tandem. We operationalize this paradigm in an interactive system designed to address the critical challenge of encoding subtle, domain-specific policies into prompt instructions. The system's structured workflow guides people to discover edge cases, articulate rationales for desired behavior, and iteratively evaluate instruction revisions against a growing test set. A user study shows our workflow helps participants refine instructions systematically and specify ambiguous policies more concretely. This work points toward more robust and responsible LLM applications through human-in-the-loop development aligned with local preferences and policies.

Problem

• Overcoming rigid separation between data work and model refinement
• Encoding subtle domain-specific policies into prompt instructions
• Systematically refining LLM behavior through human-in-the-loop development

Innovation

• Data-model co-evolution refines LLM behavior dynamically
• Interactive system encodes domain policies into prompts
• Growing test set evaluates instruction revisions iteratively