Text-embedding models are biased by names, and removing names before embedding improves semantic similarity judgments on two evaluation tasks.
The city’s hustle and bustle couldn’t dim the spark between them
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
-
What is in a name? Mitigating Name Bias in Text Embeddings via Anonymization
Text-embedding models are biased by names, and removing names before embedding improves semantic similarity judgments on two evaluation tasks.