X-Cross recombines frozen LoRA-tuned language-model activations layer by layer and reaches LoRA-comparable or better cross-domain recommendation accuracy with 25 percent of the new parameters and 50-75 percent less data.
Title resolution pending
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
-
X-Cross: Dynamic Integration of Language Models for Cross-Domain Sequential Recommendation
X-Cross recombines frozen LoRA-tuned language-model activations layer by layer and reaches LoRA-comparable or better cross-domain recommendation accuracy with 25 percent of the new parameters and 50-75 percent less data.