Requirement--Evidence Alignment for Compositional E-Commerce Queries
2026-08-03 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors address the problem of e-commerce search where queries have multiple specific requirements that must all be met. They propose a new method called REAlign that links each requirement to the actual evidence in product descriptions, helping the system know which conditions are satisfied, violated, or missing. Their approach improves the ranking of products by focusing on meeting these requirements rather than just overall relevance, resulting in better and more accurate search results. Tests showed REAlign worked better than existing methods, especially for the top few search results.
e-commerce searchquery rerankingrequirement satisfactionevidence alignmentpolicy optimizationrank list utilityconstraint modelingproduct retrieval
Authors
Weihao Shen, Wei Chen, Fuwei Zhang, Meng Yuan, Yuqin Lan, Guojun Liu, Qingsong Hua, Wei Lin, Fuzhen Zhuang
Abstract
Compositional e-commerce queries express multiple requirements that must hold jointly, yet existing rerankers collapse these constraints into aggregate relevance and often promote topical near misses over feasible products. In this paper, we introduce REAlign, a novel requirement-evidence-aligned reranking framework that explicitly connects typed query requirements with visible evidence. REAlign distinguishes satisfied, violated, and unsupported conditions, constructs requirement-targeted contrasts that expose failure modes, and optimizes duplicate-free partial rankings through Requirement-Aware Group-Relative Policy Optimization. Its list utility preserves relevance while incorporating requirement satisfaction, evidence support, material violations, and output validity. Experiments on two fixed-pool e-commerce benchmarks show consistent improvements over strong supervised and policy-optimization baselines under matched training budgets, with fewer violations among top-ranked candidates and larger gains at shallow ranks. Controlled ablations confirm the complementary value of requirement modeling, evidence grounding, and decomposed optimization.