Probe-then-Plan: Environment-Aware Planning for Industrial E-commerce Search

TL;DR

This paper introduces Environment-Aware Search Planning (EASP), a dynamic reasoning framework for industrial e-commerce search that improves relevance and efficiency by grounding plans in real-time environmental data.

Contribution

EASP reformulates search planning as a real-time, environment-aware process using a Probe-then-Plan mechanism, combining offline synthesis, supervised fine-tuning, and adaptive online serving.

Findings

EASP significantly improves relevant recall in offline evaluations.

EASP achieves substantial lifts in UCVR and GMV in online A/B tests.

EASP is successfully deployed in JD.com's AI-Search system.

Abstract

Modern e-commerce search is evolving to resolve complex user intents. While Large Language Models (LLMs) offer strong reasoning, existing LLM-based paradigms face a fundamental blindness-latency dilemma: query rewriting is agnostic to retrieval capabilities and real-time inventory, yielding invalid plans; conversely, deep search agents rely on iterative tool calls and reflection, incurring seconds of latency incompatible with industrial sub-second budgets. To resolve this conflict, we propose Environment-Aware Search Planning (EASP), reformulating search planning as a dynamic reasoning process grounded in environmental reality. EASP introduces a Probe-then-Plan mechanism: a lightweight Retrieval Probe exposes the retrieval snapshot, enabling the Planner to diagnose execution gaps and generate grounded search plans. The methodology comprises three stages: (1) Offline Data Synthesis: A Teacher Agent synthesizes diverse, execution-validated plans by diagnosing the probed environment. (2) Planner Training and Alignment: The Planner is initialized via Supervised Fine-Tuning (SFT) to internalize diagnostic capabilities, then aligned with business outcomes (conversion rate) via Reinforcement Learning (RL). (3) Adaptive Online Serving: A complexity-aware routing mechanism selectively activates planning for complex queries, ensuring optimal resource allocation. Extensive offline evaluations and online A/B testing on JD.com demonstrate that EASP significantly improves relevant recall and achieves substantial lifts in UCVR and GMV. EASP has been successfully deployed in JD.com's AI-Search system.