Publikation
TEmBed-T: A Multi-Dimensional Benchmark for Table-Level Embeddings
Ayeen Poostforoushan; Liane Vogel; Carsten Binnig
In: Computing Research Repository eprint Journal (CoRR), Vol. abs/2607.24130, Pages 1-8, arXiv, 2026.
Zusammenfassung
Tabular data is the dominant structured-data modality, and learn-
ing table representations has become a core research direction.
Table-level embeddings in particular underpin a wide range of ap-
plications, including table retrieval, data lake discovery, and table
classification. Despite their importance, there is still limited under-
standing of how different embedding approaches behave across
tasks, making systematic evaluation and analysis essential. In this
work, we introduce a systematic evaluation of table-level embed-
dings that captures several complementary properties required for
downstream effectiveness. We realize this evaluation by extend-
ing TEmBed, a recently proposed testbed for tabular embeddings,
whose table-level coverage is currently limited to a single retrieval
task. An empirical study over the TEmBed model pool confirms
that no single model excels across all tasks, demonstrating that
table-level embedding quality cannot be reduced to retrieval alone.