Blog

• Beyond Text: How We Built Multimodal Retrieval for E-Commerce Search

  March 10, 2026 research e-commerce multimodal

  Most e-commerce search systems rely purely on text — but product images carry signals that text often misses. We propose a two-stage vision-language alignment framework for retrieval, achieving +5% Recall@100 over text-only baselines on large-scale data.

  Read more →
• Multi-Channel Retrieval Fusion: From Heuristic Weights to Unified Ranking

  May 02, 2025 Search Learning-to-Rank E-Commerce Recommendation

  E-commerce search pulls candidates from text, semantic, and behavioral channels — then merges them with hand-tuned weights. We replace that with a unified LTR model trained end-to-end, deployed at Target.com with +2.85% conversion lift in A/B testing.

  Read more →