Evaluating the Evaluator: Measuring LLMs’ Adherence to Task Evaluation Instructions
Bhuvanashree Murugadoss, Christian Poelitz, Ian Drosos, Vu Le, Nick McKenna, Carina Suzana Negreanu, Chris Parnin, Advait Sarkar
Abstract
LLMs-as-a-judge is a recently popularized method which replaces human judgements in task evaluation with automatic evaluation using LLMs. Due to widespread use of RLHF (Reinforcement Learning from Human Feedback), state-of-the-art LLMs like GPT4 and Llama3 are expected to have strong alignment with human preferences when prompted for a quality judgement, such as the coherence of a text. While this seems beneficial, it is not clear whether the assessments by an LLM-as-a-judge constitute only an evaluation based on the instructions in the prompts, or reflect its preference for high-quality data similar to its fine-tune data. To investigate how much influence prompting the LLMs-as-a-judge has on the alignment of AI judgements to human judgements, we analyze prompts with increasing levels of instructions about the target quality of an evaluation, for several LLMs-as-a-judge. Further, we compare to a prompt-free method using model perplexity as a quality measure instead. We aggregate a taxonomy of quality criteria commonly used across state-of-the-art evaluations with LLMs and provide this as a rigorous benchmark of models as judges. Overall, we show that the LLMs-as-a-judge benefit only little from highly detailed instructions in prompts and that perplexity can sometimes align better with human judgements than prompting, especially on textual quality.
BibTeX
@article{Murugadoss_Poelitz_Drosos_Le_McKenna_Negreanu_Parnin_Sarkar_2025, title={Evaluating the Evaluator: Measuring LLMs’ Adherence to Task Evaluation Instructions}, volume={39}, url={https://ojs.aaai.org/index.php/AAAI/article/view/34157}, DOI={10.1609/aaai.v39i18.34157}, abstractNote={LLMs-as-a-judge is a recently popularized method which replaces human judgements in task evaluation with automatic evaluation using LLMs. Due to widespread use of RLHF (Reinforcement Learning from Human Feedback), state-of-the-art LLMs like GPT4 and Llama3 are expected to have strong alignment with human preferences when prompted for a quality judgement, such as the coherence of a text. While this seems beneficial, it is not clear whether the assessments by an LLM-as-a-judge constitute only an evaluation based on the instructions in the prompts, or reflect its preference for high-quality data similar to its fine-tune data. To investigate how much influence prompting the LLMs-as-a-judge has on the alignment of AI judgements to human judgements, we analyze prompts with increasing levels of instructions about the target quality of an evaluation, for several LLMs-as-a-judge. Further, we compare to a prompt-free method using model perplexity as a quality measure instead. We aggregate a taxonomy of quality criteria commonly used across state-of-the-art evaluations with LLMs and provide this as a rigorous benchmark of models as judges.
Overall, we show that the LLMs-as-a-judge benefit only little from highly detailed instructions in prompts and that perplexity can sometimes align better with human judgements than prompting, especially on textual quality.}, number={18}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Murugadoss, Bhuvanashree and Poelitz, Christian and Drosos, Ian and Le, Vu and McKenna, Nick and Negreanu, Carina Suzana and Parnin, Chris and Sarkar, Advait}, year={2025}, month={Apr.}, pages={19589-19597} }