Determine transfer of targeted supervision to other constraints

Determine whether targeted, programmatically labeled supervision for explicit budget compliance transfers to other checkable e-commerce constraints.

Background

The explicit-budget diagnostic shows that a small model trained on an automatically checkable budget rule can outperform general preference alignment and model scaling on both threshold and control slices.

The authors explicitly state that they have not tested whether this targeted-supervision strategy generalizes beyond budget compliance. The unresolved question concerns transfer to other constraints that can likewise be represented by programmatic rules.

References

On this constraint, such targeted supervision beats both scale and general alignment; we have not tested whether that transfers to other checkable constraints, and the fully general case may still be best handled by a hard budget filter at inference.

— ZooWork-ShopRanker: An Open, Preference-Aligned E-Commerce Reranker  (2609.31002 - Xue et al., 25 Sep 2026) in Appendix, Section 'Explicit budget: full analysis' (Section label app:budget)