Alibaba Releases Real-World Customs Agent Benchmark
机器之心 · wechat · 2026-07-19
This post introduces HSCodeComp by Alibaba's ATH-MaaS team: a deep retrieval agent benchmark for real customs scenarios, testing models' ability to determine correct customs codes (HSCode) from product info and hierarchical rules.
The task requires traversing a rule tree, retrieving rules and historical rulings, and outputting a unique 10-digit code. Human experts achieve 95% accuracy, while the best agent only reaches 49.4%, revealing a structural weakness in applying hierarchical rules
More from Research
- New paper: Absolute pose estimation from affine cues and gravity direction — ducha_aiki · 2026-09-11
- LoMa Paper Ships REALLY HardPairs Dataset, Accepted at ECCV 2026 — ducha_aiki · 2026-09-11
- Johns Hopkins Launches Full-Stack Hands-on Robot Learning Class with SO-101 Arm Kits — _krishna_murthy · 2026-09-11
- SyncWorld: In-Context Robot World Model Simulates Unseen Views and Embodiments Zero-Shot — ChongZzZhang · 2026-09-11
- A 3D Pose Dataset for Dogs Released — ducha_aiki · 2026-09-11
- Five tells that still make AI video read as AI, from physics glitches to missing operators — NewPhoneWhotiz · 2026-09-11