How Fine-Grained Should a RAG Benchmark Be? A Hierarchical Framework for Synthetic Question Generation.

@inproceedings{fensore-2026-finegrained, author = {Chase M. Fensore and Kaustubh D. Dhole and Jason Fan and Eugene Agichtein and Joyce C. Ho}, editor = {Hannah Bast and Alistair Moffat and Falk Scholer and Marc Najork and Min Zhang and Hannah Bast and Marc Najork and Min Zhang}, title = {{How Fine-Grained Should a RAG Benchmark Be? A Hierarchical Framework for Synthetic Question Generation}}, booktitle = {{Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval, SIGIR 2026, MelbourneVICAustralia, July 20-24, 2026}}, pages = {3715--3720}, publisher = {ACM}, year = {2026}, url = {https://doi.org/10.1145/3805712.3809925}, doi = {https://doi.org/10.1145/3805712.3809925} }