SR-Agent: An Experience-Driven Agentic Framework for Post-Ranking Strategies Refinement in E-Commerce Recommendation

2026-07-20

AI Summary

This work addresses the limitations of static post-ranking strategies in e-commerce recommendation systems, which struggle to adapt to dynamic environments and suffer from inefficient, non-reusable manual tuning. The authors propose SR-Agent, the first end-to-end, self-evolving post-ranking framework deployed in an industrial-scale e-commerce platform. SR-Agent employs a multi-agent architecture to close the loop of problem detection, root-cause diagnosis, and safe policy updates, integrating a user-simulation agent, a structured diagnostic analysis module, and a constrained optimization engine with a four-stage reward scheme and rollback mechanism. In A/B tests on Kuaishou E-commerce, SR-Agent achieved a 0.71% increase in order volume, along with 0.34% and 0.48% improvements in browsing depth and click-category diversity, respectively, while substantially reducing iteration cycles and operational costs.

Abstract

User experience is a first-class objective in industrial e-commerce recommender systems (RS). Post-ranking strategies, which govern diversity, similarity, and exposure over a ranked list, are widely deployed in industrial RS for their simplicity and low serving cost. However, as the online recommendation environment evolves continuously, these statically configured strategies gradually become stale, degrading the user experience. Refining them typically relies on manual inspection, diagnosis, and updates, a process that is slow, costly, and hard to reuse. Although recent LLM-based agents (e.g., RecUserSim, SimUSER, and Self-EvolveRec) offer promising directions, none of them close the full loop of automated, self-evolving strategy refinement. To bridge this gap, we introduce SR-Agent, a Strategy Refinement agentic framework that, to the best of our knowledge, is the first deployed for refining post-ranking strategies in industrial RS. SR-Agent unifies three components: (i) a UserSim agent that applies staged inspection skills to surface user-perceived bad cases; (ii) an Analysis agent that consolidates recurring bad cases into structured, reusable diagnoses; and (iii) a constrained Strategy Refinement Harness that maps diagnoses to typed and bounded actions, gated by a four-stage reward pipeline with reversible rollback. Deployed on the Kuaishou e-commerce platform, SR-Agent continuously runs this refinement loop and, in a one-month online A/B test, increases order volume by 0.71%, browsing depth by 0.34%, and clicked-category diversity by 0.48%, while markedly shortening the refinement cycle and lowering operational cost.

Problem

• post-ranking strategies
• user experience
• strategy refinement
• e-commerce recommendation
• automated optimization

Innovation

• SR-Agent
• post-ranking refinement
• LLM-based agent
• user experience simulation
• automated strategy evolution