About this paper
LCM-Promo, a finetuned large language model trained on transactional commerce data, predicts promotion response through monotonic curves and single-token conversion-threshold probabilities, improving conversion and reducing promotion cost in a production A/B test.
We present LCM-Promo, an LLM-based uplift model for promotion allocation. LCM-Promo is a task-specific Large Commerce Model (LCM): a finetuned LLM trained on transactional commerce data to predict user intent and promotion response. It learns monotonic promotion-response curves from production logs by interpreting single-token outputs as conversion-threshold probabilities, while conditioning on session context such as day and delivery shift. In a production A/B test on churn reactivation, LCM-Promo increased conversion by 1% and reduced promotion cost by 6% relative to the Segment-EMA production baseline.
LLM-based uplift modelLarge Commerce Modelfinetuned LLMtransactional commerce datauser intentpromotion responsemonotonic promotion-response curvessingle-token outputsconversion-threshold probabilitiessession context
- Upvotes
- 3
- Authors
- 7
- Published
- Oct 27, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
LCM-Promo: An LLM-Based Uplift Model for Promotion Allocation was first seen on Hugging Face · Daily papers on Nov 10, 2026. Since then it has appeared in 18 checks on 3 different days. Its best position was #22 of 30, reached on Nov 11, 17:28 UTC. It dropped off the list after Nov 12, 03:28 UTC.