LANTERN combines cross-attention and learned gating to fuse survey responses with supplementary context, improving multi-label answer prediction F1 from 0.73 to 0.78 on a proprietary dataset.
ISBN 9781510860964
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Modeling User Behavior from Adaptive Surveys with Supplemental Context
LANTERN combines cross-attention and learned gating to fuse survey responses with supplementary context, improving multi-label answer prediction F1 from 0.73 to 0.78 on a proprietary dataset.