# Shopping Companion: Benchmarking and Training LLM Agents for Long-Horizon Preference-Grounded E-Commerce Tasks

arXiv: https://arxiv.org/abs/2603.14864 · HTML: https://mlsanity.com/paper/2603.14864/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use.
- Builds on 14 of its 31 references that are in the map.
- First posted on arXiv 16 Mar 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)
- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 14 of 14)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [REALM: Retrieval-Augmented Language Model Pre-Training](https://mlsanity.com/paper/2002.08909/) (2020)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents](https://mlsanity.com/paper/2207.01206/) (2022)
- [A-MEM: Agentic Memory for LLM Agents](https://mlsanity.com/paper/2502.12110/) (2025)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Zep: A Temporal Knowledge Graph Architecture for Agent Memory](https://mlsanity.com/paper/2501.13956/) (2025)
- [Pyserini: An Easy-to-Use Python Toolkit to Support Replicable IR Research with Sparse and Dense Representations](https://mlsanity.com/paper/2102.10073/) (2021)
- [ShoppingBench: A Real-World Intent-Grounded Shopping Benchmark for LLM-based Agents](https://mlsanity.com/paper/2508.04266/) (2025)
- [Conversational Product Search Based on Negative Feedback](https://mlsanity.com/paper/1909.02071/) (2019)
- [Wizard of Shopping: Target-Oriented E-commerce Dialogue Generation with Decision Tree Branching](https://mlsanity.com/paper/2502.00969/) (2025)
- [Learning to Ask: Conversational Product Search via Representation Learning](https://mlsanity.com/paper/2411.14466/) (2024)

## Built on by (most cited first, 5 of 5)

- [ShopGym: An Integrated Framework for Realistic Simulation and Scalable Benchmarking of E-Commerce Web Agents](https://mlsanity.com/paper/2605.16116/) (2026)
- [ComboShoppingBench: Evaluating LLM Agents for Budget-Constrained Basket Shopping with Coupons](https://mlsanity.com/paper/2608.09282/) (2026)
- [Paying to Know: Micro-Transaction Markets for Verified Product Information in Agentic E-Commerce](https://mlsanity.com/paper/2606.24783/) (2026)
- [MMShopBench: A Real-Log Benchmark for Multimodal, Multi-Turn Shopping Agents](https://mlsanity.com/paper/2607.29002/) (2026)
- [PDEU-Bench: Benchmarking the Personalized Planning Lifecycle of Tool-Calling LLM Agents](https://mlsanity.com/paper/2609.34930/) (2026)

## Abstract (arXiv)

In e-commerce, LLM agents show promise for shopping tasks such as recommendations, budget management, and bundle deals, where accurately capturing user preferences from long-horizon conversations is critical. However, progress is limited by two key challenges: (1) the absence of benchmarks for evaluating long-term preference-aware shopping tasks, and (2) the lack of fine-grained supervision for shopping agent training. To fill the benchmark gap, we introduce Shopping Companion Bench, a novel benchmark comprising two shopping tasks that require cross-session preference memory, grounded in a product pool of over 1.2 million real-world items. Our analysis further identifies two major sources of failure on this benchmark: cascading errors caused by preference hallucination, and insufficient verification of product attributes against user requirements. To address these failure modes, we design annotation-free, tool-wise rewards that provide process supervision for each tool call, alleviating reward sparsity in long-horizon tasks. Experimental results demonstrate that even state-of-the-art models such as GPT-5 achieve success rates below 70%, highlighting the difficulty of our benchmark. Notably, our fine-tuned lightweight 4B model consistently outperforms strong baselines in both preference capture and task performance, suggesting the effectiveness of our reward design.