Shopify study: popularity, not modeling, drives generative retrieval's edge in e-commerce
_reachsumit · x · 2026-10-15
A new arXiv paper (2610.19065) revisits generative retrieval (GR) vs. dense retrieval (DR) for e-commerce candidate generation with counterintuitive findings:
- Across two Amazon Toys datasets, GR leads on best relevant item value per user, while DR wins on exact-target recall and relevant value across the ranked list.
- GR's value lead is largely driven by popularity; adding a static item value to DR in one reranking step reverses the lead while keeping recall above GR.
- On an industrial-scale internal dataset with LLM-judged relevance, a 14B SID-LLM still underperforms a dense encoder 47x smaller.
More from Research
- kalomaze argues ensembled synthetic data can cover data-scarce domains — kalomaze · 2026-10-15
- Axiom claims proof of generalized quasi-Riemann hypothesis for all Dedekind zeta functions — basedjensen · 2026-10-15
- TorchLean expands into a verified tensor programming framework with einsum and GPU execution in Lean — burny_tech · 2026-10-15
- StreamErase hits 26 FPS real-time video object erasing on a single GPU, 200x faster than prior methods — Zhuobai Dong · 2026-10-15
- Huawei's SwarmRSI: evolving agent swarms achieve recursive self-improvement in science — Qianru Meng · 2026-10-15
- Glance Forcing cuts video distillation trajectory cost from thousands of GPU-hours to one — Zhuobai Dong · 2026-10-15