BERT-as-a-Judge: A Robust Alternative to Lexical Methods for Efficient Reference-Based LLM Evaluation

TL;DR

This paper introduces BERT-as-a-Judge, a lightweight, encoder-based evaluation method for LLM outputs that outperforms lexical methods and matches larger LLM judges in accuracy, offering a scalable and reliable alternative.

Contribution

The paper presents BERT-as-a-Judge, a novel, efficient evaluation approach that improves semantic correctness assessment over lexical methods and rivals larger models.

Findings

Lexical evaluation correlates poorly with human judgments.

BERT-as-a-Judge outperforms lexical baselines in accuracy.

It matches larger LLM judges while being more computationally efficient.

Abstract

Accurate evaluation is central to the large language model (LLM) ecosystem, guiding model selection and downstream adoption across diverse use cases. In practice, however, evaluating generative outputs typically relies on rigid lexical methods to extract and assess answers, which can conflate a model's true problem-solving ability with its compliance with predefined formatting guidelines. While recent LLM-as-a-Judge approaches mitigate this issue by assessing semantic correctness rather than strict structural conformity, they also introduce substantial computational overhead, making evaluation costly. In this work, we first systematically investigate the limitations of lexical evaluation through a large-scale empirical study spanning 36 models and 15 downstream tasks, demonstrating that such methods correlate poorly with human judgments. To address this limitation, we introduce BERT-as-a-Judge, an encoder-driven approach for assessing answer correctness in reference-based generative settings, robust to variations in output phrasing, and requiring only lightweight training on synthetically annotated question-candidate-reference triplets. We show that it consistently outperforms the lexical baseline while matching the performance of much larger LLM judges, providing a compelling tradeoff between the two and enabling reliable, scalable evaluation. Finally, through extensive experimentation, we provide detailed insights into BERT-as-a-Judge's performance to offer practical guidance for practitioners, and release all project artifacts to foster downstream adoption.