Recurring 'chunks' in neural embeddings can be extracted, predict input patterns, and be perturbed to steer a model's outputs.
and richard baraniuk
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Discovering Chunks in Neural Embeddings for Interpretability
Recurring 'chunks' in neural embeddings can be extracted, predict input patterns, and be perturbed to steer a model's outputs.