Generative ranking with action-oriented sequences and lightweight position and time biases improves user metrics at comparable inference cost to a production recommender.
Sliding spectrum decomposition for diversified recommendation // Proceedings of the 27th ACM SIGKDD conference on knowledge discovery & data mining
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards Large-scale Generative Ranking
Generative ranking with action-oriented sequences and lightweight position and time biases improves user metrics at comparable inference cost to a production recommender.