Code-Oriented LM Embeddings (COLE) extracted from raw PyTorch class text with frozen language models improve surrogate predictors in NAS, cutting evaluation budget by 34% in BANANAS on NAS-Bench-201 for CIFAR-100.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Towards Code-Oriented LM Embeddings for Surrogate-Assisted Neural Architecture Search
Code-Oriented LM Embeddings (COLE) extracted from raw PyTorch class text with frozen language models improve surrogate predictors in NAS, cutting evaluation budget by 34% in BANANAS on NAS-Bench-201 for CIFAR-100.