A new public benchmark of 24 text embedding models on six built asset datasets shows uneven performance and that general-purpose benchmarks fail to predict domain-specific results.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Benchmarking pre-trained text embedding models in aligning built asset information
A new public benchmark of 24 text embedding models on six built asset datasets shows uneven performance and that general-purpose benchmarks fail to predict domain-specific results.