用端到端训练提升大模型生成的用户画像推荐效果。
End-to-end Training for Recommendation with Language-based User Profiles
- 直接训练大模型生成用户画像，优化推荐目标。
- 性能超越零样本方法，媲美顶尖嵌入式推荐模型。
- 保留可解释性，适合需要透明推荐的场景。
基于自然语言的用户画像在推荐系统中日益受到关注，相比嵌入式方法更具透明性和可解释性。现有研究多采用大语言模型（LLM）进行零样本推理生成画像，但质量不足，影响推荐效果。本文提出 LangPTune，首个面向推荐任务的端到端训练框架，通过显式优化大模型以提升画像质量。在多种训练配置和基准上的实验表明，LangPTune 不仅显著优于零样本基线，还能达到当前最优嵌入式方法的性能水平。此外，通过 GPT-4 模拟和众包用户研究，验证了训练过程有效保持了画像的可解释性。代码已开源：https://github.com/ZhaolinGao/LangPTune。
原文摘要 · Abstract (English)
There is a growing interest in natural language-based user profiles for recommender systems, which aims to enhance transparency and scrutability compared with embedding-based methods. Existing studies primarily generate these profiles using zero-shot inference from large language models (LLMs), but their quality remains insufficient, leading to suboptimal recommendation performance. In this paper, we introduce LangPTune, the first end-to-end training framework to optimize LLM-generated user profiles. Our method significantly outperforms zero-shot approaches by explicitly training the LLM for the recommendation objective. Through extensive evaluations across diverse training configurations and benchmarks, we demonstrate that LangPTune not only surpasses zero-shot baselines but can also matches the performance of state-of-the-art embedding-based methods. Finally, we investigate whether the training procedure preserves the interpretability of these profiles compared to zero-shot inference through both GPT-4 simulations and crowdworker user studies. Implementation of LangPTune can be found at https://github.com/ZhaolinGao/LangPTune.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。