# ComboShoppingBench: Evaluating LLM Agents for Budget-Constrained Basket Shopping with Coupons

arXiv: https://arxiv.org/abs/2608.09282 · HTML: https://mlsanity.com/paper/2608.09282/

## Key facts

- Cited by 2 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 18 of its 28 references that are in the map.
- First posted on arXiv 10 Aug 2026.

## Placed in

- LLM agents and tool use / [Social simulation and multi-agent debate](https://mlsanity.com/agents/social-simulation-and-multi-agent-debate/)

## Builds on (most cited first, 18 of 18)

- [WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents](https://mlsanity.com/paper/2207.01206/) (2022)
- [Mind2Web: Towards a Generalist Agent for the Web](https://mlsanity.com/paper/2306.06070/) (2023)
- [Inferring Networks of Substitutable and Complementary Products](https://mlsanity.com/paper/1506.08839/) (2015)
- [DeepShop: A Benchmark for Deep Research Shopping Agents](https://mlsanity.com/paper/2506.02839/) (2025)
- [CrossCBR: Cross-view Contrastive Learning for Bundle Recommendation](https://mlsanity.com/paper/2206.00242/) (2022)
- [Bundle Recommendation with Graph Convolutional Networks](https://mlsanity.com/paper/2005.03475/) (2020)
- [Multi-view Intent Disentangle Graph Networks for Bundle Recommendation](https://mlsanity.com/paper/2202.11425/) (2022)
- [WebMall -- A Multi-Shop Benchmark for Evaluating Web Agents](https://mlsanity.com/paper/2508.13024/) (2025)
- [ShopSimulator: Evaluating and Exploring RL-Driven LLM Agent for Shopping Assistants](https://mlsanity.com/paper/2601.18225/) (2026)
- [ShoppingComp: Are LLMs Really Ready for Your Shopping Cart?](https://mlsanity.com/paper/2511.22978/) (2025)
- [Shopping Companion: Benchmarking and Training LLM Agents for Long-Horizon Preference-Grounded E-Commerce Tasks](https://mlsanity.com/paper/2603.14864/) (2026)
- [ShopGym: An Integrated Framework for Realistic Simulation and Scalable Benchmarking of E-Commerce Web Agents](https://mlsanity.com/paper/2605.16116/) (2026)
- [EComAgentBench: Benchmarking Shopping Agents on Long-Horizon Tasks with Distributed Hidden Intent](https://mlsanity.com/paper/2606.17698/) (2026)
- [CASE: Cadence-Aware Set Encoding for Large-Scale Next Basket Repurchase Recommendation](https://mlsanity.com/paper/2604.06718/) (2026)
- [BRIDGE: Bundle Recommendation via Instruction-Driven Generation](https://mlsanity.com/paper/2412.18092/) (2024)
- [Modeling Item-Level Dynamic Variability with Residual Diffusion for Bundle Recommendation](https://mlsanity.com/paper/2507.03280/) (2025)
- [EpicCBR: Item-Relation-Enhanced Dual-Scenario Contrastive Learning for Cold-Start Bundle Recommendation](https://mlsanity.com/paper/2602.11680/) (2026)
- [Time-Interval-Aware Disentangled Expert Modeling for Next-Basket Recommendation](https://mlsanity.com/paper/2605.00499/) (2026)

## Built on by (most cited first, 2 of 2)

- [Coding Agent Memory Post-training: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via Reinforcement Learning](https://mlsanity.com/paper/2609.34422/) (2026)
- [RealWorldShop: Benchmarking and Improving Conversational Shopping Agents in Real-World E-commerce](https://mlsanity.com/paper/2609.38974/) (2026)

## Abstract (arXiv)

Real-world shopping often requires constructing a basket of complementary items rather than retrieving a single product. Such combo-shopping tasks arise in device setup, meal preparation, event planning, and group takeout ordering, requiring joint reasoning about item compatibility, availability, store-level requirements, delivery fees, coupons, and budgets. Evaluation is challenging because multiple baskets may satisfy the same request, making exact-match metrics unsuitable, whereas semantic evaluation alone cannot detect infeasible orders, invalid coupon combinations, or incorrect payments. We introduce ComboShoppingBench, an agentic shopping benchmark for open-ended yet verifiable basket construction in a simulated commerce and takeout environment. During task synthesis, an exploration agent constructs a feasible and semantically coherent basket of purchasable products; this witness guides the generation of coupons, budget constraints, user queries, and aligned evaluation rubrics. During evaluation, LLM judges assess semantic satisfaction, response quality, and claim faithfulness, while deterministic validation checks product-ID validity, budget compliance, and coupon optimality. Experiments with diverse LLM agents demonstrate that even strong agents struggle on ComboShoppingBench, highlighting substantial room for improvement in reliable, constraint-aware combo shopping.