TAHB Benchmark for Text-Attributed Hypergraph Learning
August 18, 2026
TAHB introduces the first public benchmark for integrating large language models with hypergraph structures. It includes 10 real-world datasets across e-commerce, academia, movies, and politics to evaluate text-aware hypergraph representation learning.
HOW THIS AFFECTS YOU
●
builderThis provides a standardized way to test graph-based recommendation or social network systems.
●
researcherYou can use this to evaluate how well your models handle higher-order groupwise relationships with textual data.