Sticky tokens, defined as vocabulary tokens that pull sentence similarity toward the model's mean, are detected across 40 text embedding models and can degrade retrieval and clustering by up to 50%.
howtoimprove mental health?
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sticking to the Mean: Detecting Sticky Tokens in Text Embedding Models
Sticky tokens, defined as vocabulary tokens that pull sentence similarity toward the model's mean, are detected across 40 text embedding models and can degrade retrieval and clustering by up to 50%.