AdaRec: Smarter recommendations from fewer clicks
Meet AdaRec: smarter recommendations from fewer clicks
AdaRec is a new way to personalize recommendations using large language models. Instead of hand-crafted features, it turns your past interactions into short, readable stories—“narrative profiles”—and lets an LLM reason over them with just a few examples.
It thinks in two channels: horizontal behavioral alignment (spotting patterns from similar users) and vertical causal attribution (identifying which factors truly drive your choices).
- Up to 8% better than traditional ML and LLM baselines when only a few examples are available.
- In zero-shot setups, up to 19% better than expert-crafted profiles—great for long‑tail items and cold starts.
- Lightweight fine-tuning on AdaRec’s own synthetic data can match full fine-tuning, saving time and compute.
Bottom line: clearer profiles, faster adaptation, and stronger recommendations with less data—for e-commerce and beyond.
Paper: http://arxiv.org/abs/2511.07166v1
Register: https://www.AiFeta.com
AI RecommenderSystems LLM Personalization Ecommerce NLP MachineLearning
