A new simulated benchmark for personalized household planning shows that an explicit preference-prediction stage, rather than end-to-end learning, substantially improves LLM-based planning.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Learning to Plan with Personalized Preferences
A new simulated benchmark for personalized household planning shows that an explicit preference-prediction stage, rather than end-to-end learning, substantially improves LLM-based planning.