ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Jiangyuan, Xiao, Kejun, Sun, Qi, Zhao, Huaipeng, Luo, Tao, Zhang, Jian Dong, Zeng, Xiaoyi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917136327245824
author Wang, Jiangyuan
Xiao, Kejun
Sun, Qi
Zhao, Huaipeng
Luo, Tao
Zhang, Jian Dong
Zeng, Xiaoyi
author_facet Wang, Jiangyuan
Xiao, Kejun
Sun, Qi
Zhao, Huaipeng
Luo, Tao
Zhang, Jian Dong
Zeng, Xiaoyi
contents Existing benchmarks in e-commerce primarily focus on basic user intents, such as finding or purchasing products. However, real-world users often pursue more complex goals, such as applying vouchers, managing budgets, and finding multi-products seller. To bridge this gap, we propose ShoppingBench, a novel end-to-end shopping benchmark designed to encompass increasingly challenging levels of grounded intent. Specifically, we propose a scalable framework to simulate user instructions based on various intents derived from sampled real-world products. To facilitate consistent and reliable evaluations, we provide a large-scale shopping sandbox that serves as an interactive simulated environment, incorporating over 2.5 million real-world products. Experimental results demonstrate that even state-of-the-art language agents (such as GPT-4.1) achieve absolute success rates under 50% on our benchmark tasks, highlighting the significant challenges posed by our ShoppingBench. In addition, we propose a trajectory distillation strategy and leverage supervised fine-tuning, along with reinforcement learning on synthetic trajectories, to distill the capabilities of a large language agent into a smaller one. As a result, our trained agent achieves competitive performance compared to GPT-4.1.
format Preprint
id arxiv_https___arxiv_org_abs_2508_04266
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents
Wang, Jiangyuan
Xiao, Kejun
Sun, Qi
Zhao, Huaipeng
Luo, Tao
Zhang, Jian Dong
Zeng, Xiaoyi
Computation and Language
Existing benchmarks in e-commerce primarily focus on basic user intents, such as finding or purchasing products. However, real-world users often pursue more complex goals, such as applying vouchers, managing budgets, and finding multi-products seller. To bridge this gap, we propose ShoppingBench, a novel end-to-end shopping benchmark designed to encompass increasingly challenging levels of grounded intent. Specifically, we propose a scalable framework to simulate user instructions based on various intents derived from sampled real-world products. To facilitate consistent and reliable evaluations, we provide a large-scale shopping sandbox that serves as an interactive simulated environment, incorporating over 2.5 million real-world products. Experimental results demonstrate that even state-of-the-art language agents (such as GPT-4.1) achieve absolute success rates under 50% on our benchmark tasks, highlighting the significant challenges posed by our ShoppingBench. In addition, we propose a trajectory distillation strategy and leverage supervised fine-tuning, along with reinforcement learning on synthetic trajectories, to distill the capabilities of a large language agent into a smaller one. As a result, our trained agent achieves competitive performance compared to GPT-4.1.
title ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents
topic Computation and Language
url https://arxiv.org/abs/2508.04266