LUNAR is a synthetic benchmark anchored in real-world logs that shows current LLMs cannot reliably integrate cross-domain behavioral evidence, and that stronger personalization often comes with weaker privacy protection.
Learning to reason for multi-step retrieval of personal context in personalized question answering
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LUNAR: Benchmarking Personalized Large Language Models on UNiversal User BehAvioR Logs
LUNAR is a synthetic benchmark anchored in real-world logs that shows current LLMs cannot reliably integrate cross-domain behavioral evidence, and that stronger personalization often comes with weaker privacy protection.