M-GPT combines interaction-level graph convolution with multi-scale session attention and reports state-of-the-art next-purchase accuracy on three multi-behavior datasets.
Title resolution pending
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
-
Multi-Grained Preference Enhanced Transformer for Multi-Behavior Sequential Recommendation
M-GPT combines interaction-level graph convolution with multi-scale session attention and reports state-of-the-art next-purchase accuracy on three multi-behavior datasets.