Generate, Evaluate, Iterate: Synthetic Data for Human-in-the-Loop Refinement of LLM Judges
This addresses the need for efficient and scalable data refinement in LLM evaluation workflows, though it is incremental as it builds on existing LLM-as-a-judge paradigms.
The paper tackles the problem of limited diverse data for refining LLM-as-a-judge evaluation criteria by introducing a tool that generates synthetic test cases, finding in a user study (N=24) that 83% preferred it over manual methods and that the synthetic data was as effective as hand-crafted data.
The LLM-as-a-judge paradigm enables flexible, user-defined evaluation, but its effectiveness is often limited by the scarcity of diverse, representative data for refining criteria. We present a tool that integrates synthetic data generation into the LLM-as-a-judge workflow, empowering users to create tailored and challenging test cases with configurable domains, personas, lengths, and desired outcomes, including borderline cases. The tool also supports AI-assisted inline editing of existing test cases. To enhance transparency and interpretability, it reveals the prompts and explanations behind each generation. In a user study (N=24), 83% of participants preferred the tool over manually creating or selecting test cases, as it allowed them to rapidly generate diverse synthetic data without additional workload. The generated synthetic data proved as effective as hand-crafted data for both refining evaluation criteria and aligning with human preferences. These findings highlight synthetic data as a promising alternative, particularly in contexts where efficiency and scalability are critical.