AI & ML
Auditing LLM Pseudo-Label Distillation for Review Informativeness at E-Commerce Scale
arXiv cs.IR0 views
arXiv:2611.13797v1 Announce Type: new Abstract: We study hard-label distillation from Gemini 3.1 Flash-Lite to compact encoders on 5.34 million Turkish marketplace reviews. A strict new-SKU temporal test measures teacher fidelity; a separate, teacher-label-blind audit by one company annotator measures editorial agreement. An H3 category prefix slightly improves teacher fidelity on the full new-SKU test. On the 16,000-review chronological audit subset, it instead reduces agreement with both Gemini and the annotator, with a clearer paired difference for CORN than for cross entropy. Thus the context effect depends on the evaluation population, and teacher fidelity alone does not establish editorial alignment. Under the measured offline configurations, a single-L4 student pipeline processes reviews 50.2 times faster than the teacher pipeline.
Source：https://arxiv.org/abs/2611.13797PreviousExploiting Feature-invariant Scaling Laws for Efficient A/B Experimentation in Recommendation SystemsNextControlled Knowledge Engineering for Taxonomy Induction with Tool-Grounded LLM Agents and Runtime-Adaptive Policies
Related posts
Singapore Airlines elevates travel & work with OpenAI technology
0OpenAI Blog
How to Plan a Five-Slide Social Carousel with One Consistent Character
0dev.to
Kimi K2.5 to K3.1 Pro: 8 API Models, One Enterprise Task, and What Actually Got Done
0dev.to
Visa tested 22 shopping agents. 43% failed - and half of those bought anyway. Here's what actually stops it.
0dev.to
Building WUUUH: Skills Drift Faster Than Confidence — Keeping a Hand In During a Career Gap, What Worked, What Broke
0dev.to
FDR-Controlled Variable Selection for Generalized Linear Models and Cox Regression with Virtual Dummies
0arXiv eess.SP