Concatenating k-nearest-neighbor context from domain and in-domain text during continual pretraining improves German process-industry semantic search at roughly one quarter of the GPU cost of standard DAPT.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
Efficient Domain-adaptive Continual Pretraining for the Process Industry in the German Language
Concatenating k-nearest-neighbor context from domain and in-domain text during continual pretraining improves German process-industry semantic search at roughly one quarter of the GPU cost of standard DAPT.