GRACE combines eligibility-aware constrained decoding with wide-beam decoder kernel optimizations to bring generative ads retrieval inside real-time latency and compute budgets.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.IR 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
GRACE: Generative Recommender Acceleration Engine for Real-Time Ads Retrieval
GRACE combines eligibility-aware constrained decoding with wide-beam decoder kernel optimizations to bring generative ads retrieval inside real-time latency and compute budgets.