ZooWork-ShopRanker: LLM-judge-aligned open e-commerce rerankers from 0.6B to 8B

Siqiao Xue · hf · 2026-09-28

ZooWork-ShopRanker releases open e-commerce rerankers (0.6B/4B/8B) trained on pairwise labels from a panel of reasoning LLMs acting as a preference oracle, plus ShopRank-Bench (10k private-traffic preference pairs). The 8B and 4B beat the strongest open reranker baseline, gains hold across formats and MTEB.

Related event: Alibaba-Linked Team Open-Sources ZooWork-ShopRanker E-commerce Rerankers(2 posts)→

Original post →

More from Research

Research channel →