A per-user news recommender with farthest-item negative sampling in embedding space matches the offline accuracy of larger models and enables on-device training.
In: Proceedings of the 1st International Work- shop on Data Semantics - DataSem ’10, p
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Rethinking negative sampling in content-based news recommendation
A per-user news recommender with farthest-item negative sampling in embedding space matches the offline accuracy of larger models and enables on-device training.