Know Your Needs Better: Towards Structured Understanding of Marketer Demands with Analogical Reasoning Augmented LLMs

TL;DR

This work tackles translating non-expert marketers' natural-language demands into structured targeting criteria by introducing SELL, a simple yet expressive language built from Keys, Values, Operators, and AND/OR composition. The proposed ARALLM framework combines Analogical Reasoning based Prompting with a Reasoning-Augmented Multi-Task Distillation pipeline to produce accurate SELL expressions using a compact reasoning library and small deployed models. Empirical results on GPT-3.5 and smaller LLMs show clear gains in structure accuracy and overall alignment, and online deployment demonstrates fast inference and improved campaign performance. The approach yields a practical NL2SELL pipeline and a scalable pathway to industry-ready, privacy-preserving targeting systems, with future work aimed at automating reasoning-library construction and extending to other structured-logic translation tasks.

Abstract

In this paper, we explore a new way for user targeting, where non-expert marketers could select their target users solely given demands in natural language form. The key to this issue is how to transform natural languages into practical structured logical languages, i.e., the structured understanding of marketer demands. In practical scenarios, the demands of non-expert marketers are often abstract and diverse. Considering the impressive natural language processing ability of large language models (LLMs), we try to leverage LLMs to solve this issue. To stimulate the LLMs' reasoning ability, the chain-of-thought (CoT) prompting method is widely used, but existing methods still have some limitations in our scenario: (1) Previous methods either use simple "Let's think step by step" spells or provide fixed examples in demonstrations without considering compatibility between prompts and concrete questions, making LLMs ineffective when the marketers' demands are abstract and diverse. (2) Previous methods are often implemented in closed-source models or excessively large models, which is not suitable in industrial practical scenarios. Based on these, we propose ARALLM (i.e., Analogical Reasoning Augmented Large Language Models) consisting of two modules: Analogical Reasoning based Prompting and Reasoning-Augmented Multi-Task Model Distillation. Part of our data and code can be found at https://github.com/alipay/Analogic-Reasoning-Augmented-Large-Language-Model.

Figures (8)

• Figure 1: The comparison of user targeting approach.
• Figure 2: Overview of analogical reasoning based prompting methods.
• Figure 3: The impact of the size of parameter settings.
• Figure 4: The application of ARALLM in User Targeting.
• Figure 5: Reasoning steps generation prompt.