EARN inserts learnable register tokens at both ends of a user prompt, prunes prompt tokens after early layers, and achieves up to 3.79x faster LLM-based recommendation inference with comparable or better accuracy.
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
-
EARN: Efficient Inference Acceleration for LLM-based Generative Recommendation by Register Tokens
EARN inserts learnable register tokens at both ends of a user prompt, prunes prompt tokens after early layers, and achieves up to 3.79x faster LLM-based recommendation inference with comparable or better accuracy.