LLM-as-a-judge validity is strongly task-dependent across physics assessment formats
As large language models (LLMs) are increasingly considered for automated assessment and feedback, understanding when LLM marking is valid is essential. We evaluate LLM-as-a-judge marking across four settings spanning three assessment formats - structured questions, written essays, and scientific plots - comparing GPT-...
Will Yeadon, Tom Hardy, P. Mackay et al.
· 2 citations