{"ID":6536322,"CreatedAt":"2026-07-14T01:21:01.169441415Z","UpdatedAt":"2026-07-14T03:10:56.521482197Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.10096","arxiv_id":"2607.10096","title":"Scaling and Stabilizing Large-Scale Embedding-Based Retrieval","abstract":"Embedding-based retrieval (EBR) is foundational to large-scale e-commerce search, yet its effectiveness is often constrained by the quality of training signals and the representational capacity of the encoder. Standard dual-encoders suffer from a training-inference gap: they are optimized on narrow candidate pools but must discriminate against hundreds of millions of items during inference. Furthermore, while transitioning to higher-capacity backbones can mitigate this gap, simply replacing a mature model can lead to inconsistent retrieval behavior and a loss of the domain-specific knowledge established in previous iterations. In this paper, we present a unified pipeline deployed at Walmart that addresses both signal quality and model evolution. Our contributions are two-fold: (1) Hybrid Hard Negative Mining: We integrate Online Cross-Batch Sampling to increase negative diversity by an order of magnitude and Hybrid Offline Mining, which combines cross-encoder predictions with metadata heuristics to identify nuanced mismatches. (2) Legacy-Aware Distillation: We transition from DistilBERT to a higher-capacity GTE-base encoder. To ensure a smooth and superior transition, we introduce a Warm-Start Distillation technique that transfers domain-specific expertise from the legacy model to the new backbone. Validated through extensive offline experiments and online A/B testing, the proposed pipeline is deployed in live production, delivering a +7.34% improvement in NDCG@5 and a +0.50% lift in gross revenue.","short_abstract":"Embedding-based retrieval (EBR) is foundational to large-scale e-commerce search, yet its effectiveness is often constrained by the quality of training signals and the representational capacity of the encoder. Standard dual-encoders suffer from a training-inference gap: they are optimized on narrow candidate pools but...","url_abs":"https://arxiv.org/abs/2607.10096","url_pdf":"https://arxiv.org/pdf/2607.10096v1","authors":"[\"Zhen Yang\",\"Juexin Lin\",\"Hongwei Shang\",\"Kaihao Li\",\"Feng Liu\",\"Satya Chembolu\",\"Xunfan Cai\",\"Xinyi Liu\",\"Cun Mu\",\"Tony Lee\",\"Ciya Liao\"]","published":"2026-07-11T03:24:26Z","proceeding":"cs.IR","tasks":"[\"cs.IR\"]","methods":"[]","has_code":false}