Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

Improving Item Discoverability in e-Commerce Search via Related Intent Generation

Source

arXiv

arxiv.org/abs/2607.27172v1

Read the source

Source

  • Source name: arXiv
  • Source kind: paper
  • Authors / organization: Ji Xin, Xiao Xiao, Ishan Bhatt, Vinesh Gudla, Trace Levinson, Raochuan Fan, Shishir Kumar Prasad, Prakash Putta, Tejaswi Tenneti
  • Venue: arXiv: cs.IR
  • Published: 2026-07-29
  • Collected: 2026-07-30
  • URL: http://arxiv.org/abs/2607.27172v1
  • PDF: https://arxiv.org/pdf/2607.27172v1
  • DOI: UNKNOWN

Plain-Language Brief

Researchers propose a new system for improving item discoverability in e-commerce searches. The system uses intent-conditioned recall expansion to generate implicit user intents, which helps to expand candidate recall while maintaining relevance. This system is designed to address the cost-quality tradeoff of generative retrieval through a two-stage hybrid architecture. The first stage uses large language models (LLMs) for head queries, and the second stage introduces a finetuned small language model (SLM) for tail queries.

What the Document Says

  • The traditional search systems prioritize precision over recall, which is limiting in e-commerce marketplaces, especially grocery.
  • The authors propose a system for discovery-augmented search using intent-conditioned recall expansion.
  • The system generates implicit user intents to expand candidate recall while maintaining relevance.
  • The system consists of a two-stage hybrid architecture: first using large language models (LLMs) for head queries, and then introducing a finetuned small language model (SLM) for tail queries.
  • The authors evaluate the system using a rigorous dual framework: (a) LLM-as-a-judge metrics validated against human preferences for semantic quality, and (b) end-to-end session-level purchase analysis.
  • Results show that the approach improves both intent generation quality and downstream retrieval effectiveness.
  • The discovery-augmented search may serve as a marketplace-balancing mechanism, giving long-tail and emerging supply an opportunity for query-conditioned exposure.

Key Concepts Mentioned

  • e-commerce search
  • intent-conditioned recall expansion
  • large language models (LLMs)
  • small language model (SLM)
  • discovery-augmented search
  • head queries
  • tail queries
  • long-tail supply

Claims Made by the Document

  • Traditional search systems are limited in e-commerce marketplaces.
  • The proposed system improves both intent generation quality and downstream retrieval effectiveness.
  • Discovery-augmented search may serve as a marketplace-balancing mechanism.

Evidence Used

  • In e-commerce marketplaces and particularly grocery, the paradigm of traditional search systems is limiting.
  • Results demonstrate that our approach improves both intent generation quality and downstream retrieval effectiveness.
  • Beyond relevance gains, discovery-augmented search may serve as a marketplace-balancing mechanism.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Traditional search systems are optimized to retrieve items that strictly match a query, often prioritizing precision over recall. In e-commerce marketplaces and particularly grocery, this paradigm is limiting, as user satisfaction and commercial outcomes depend heavily on the discoverability of substitute, complementary, and thematically related items. In this paper, we present a scalable system for discovery-augmented search that leverages intent-conditioned recall expansion. Our approach generates implicit user intents to expand candidate recall while maintaining relevance. The system addresses the cost-quality tradeoff of generative retrieval through a two-stage hybrid architecture. First, we leverage closed-weight large language models (LLMs) to maximize discoverability for head queries. To extend these benefits to tail queries, we then introduce a finetuned small language model (SLM), trained via LoRA adapters and teacher-student distillation. We evaluate the system using a rigorous dual framework: (a) LLM-as-a-judge metrics validated against human preferences for semantic quality, and (b) end-to-end session-level purchase analysis. Results demonstrate that our approach improves both intent generation quality and downstream retrieval effectiveness, extending discovery coverage from approximately 60% to 80% of query traffic at roughly 30% of the teacher model's inference cost, offering a viable path for deployment in large-scale marketplaces. Beyond relevance gains, discovery-augmented search may serve as a marketplace-balancing mechanism, giving long-tail and emerging supply an opportunity for query-conditioned exposure.