Rewrite-to-Rank: Optimizing Ad Visibility via Retrieval-Aware Text Rewriting

TL;DR

This paper proposes a retrieval-aware text rewriting method for advertisements to improve their visibility in retrieval systems and LLM responses, using supervised fine-tuning and reinforcement learning, with significant experimental improvements.

Contribution

It introduces a novel fine-tuning framework with custom loss and reinforcement learning to optimize ad phrasing for better retrieval ranking and inclusion in LLM outputs.

Findings

Models trained with PPO outperform prompt engineering and supervised fine-tuning.

Achieved up to 2.79 increase in ad inclusion frequency at top 5 results.

Significant improvements in ranking metrics demonstrate effectiveness.

Abstract

Search algorithms and user query relevance have given LLMs the ability to return relevant information, but the effect of content phrasing on ad visibility remains underexplored. We investigate how LLM-based rewriting of advertisements can improve their ranking in retrieval systems and inclusion in generated LLM responses, without modifying the retrieval model itself. We introduce a supervised fine-tuning framework with a custom loss balancing semantic relevance and content fidelity. To evaluate effectiveness, we propose two metrics: DeltaMRR@K (ranking improvement) and DeltaDIR@K (inclusion frequency improvement). Our approach presents a scalable method to optimize ad phrasing, enhancing visibility in retrieval-based LLM workflows. Experiments across both instruction-based and few-shot prompting demonstrate that PPO trained models outperform both prompt engineering and supervised fine-tuning in most cases, achieving up to a 2.79 DeltaDIR@5 and 0.0073 DeltaMRR@5 in instruction-based prompting. These results highlight the importance of how the ad is written before retrieval and prompt format and reinforcement learning in effective ad rewriting for LLM integrated retrieval systems.