In the realm of sponsored search advertising, matching advertisements with the search intent of a user's query is crucial. Query-to-bidwords(i.e. bidding keywords) rewriting, which involves transforming user queries into keywords for bidding, is a vital technique that has garnered significant attention from both industry and academia. Recently, with the prevalence of large language models (LLMs), generative retrieval methods have proven effective in producing high-relevance rewrites. However, we have identified a significant limitation in existing approaches: While fine-tuning LLMs for specific domains enhances semantic relevance, these models have no perception of the intrinsic value of their generated outputs, such as commercial value. Therefore, after supervised fine-tuning (SFT), a reinforcement learning from human feedback (RLHF) phase is often employed to address this issue. Nevertheless, traditional preference alignment methods often face challenges in aligning fine-grained values and are susceptible to overfitting, which diminishes the effectiveness and quality of the generated results. To address these challenges, we propose VALUE (Value-Aware Large language model for qUery rewriting via wEighted trie), the first framework that ensures the generation of high-value and highly relevant bidwords. Our approach utilizes weighted trie, an innovative modification of the traditional trie data structure. By modulating the LLM's output probability distribution with value information from the trie during decoding process, we constrain the generation space and guide the trajectory of text production. Our method not only addresses fine-grained value alignment but also effectively reduces the hallucination issues often encountered with LLMs. Offline experiments demonstrate the effectiveness of our method in semantic matching and preference alignment, showing a remarkable improvement in the value attribute by more than fivefold. Online A/B tests further revealed that our Revenue Per Mille (RPM) metric increased by 1.64%. VALUE has been deployed on our advertising system since October 2024 and served the Double Eleven promotions, the biggest shopping carnival in China.
-
Notifications
You must be signed in to change notification settings - Fork 1
alimama-tech/LLM4QR_VALUE
About
No description, website, or topics provided.
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published