How to Train an LLM-RecSys Hybrid for Steerable Recs

Learn how to fine‑tune Qwen3‑8B to embed product IDs in its vocabulary, generate recommendations from interaction data, and steer results via conversational prompts.

Overview

I’ll be demoing how I finetuned Qwen3-8B to understand product IDs. The result is a language model that can converse in both English and item IDs, not with retrieval or other tools, but as a single, “bilingual” model where items (i.e., semantic IDs) are part of its vocabulary. Like a recommender model, it can recommend items given historical interactions. But the big surprise—and capability unlock—was when I found that I could simply chat with the model to steer its recommendations, and it could reason about its choices, offer explanations, and creatively name product bundles.

Video

Links

https://eugeneyan.com/writing/semantic-ids/
RQ-VAE generated Semantic IDs train Qwen3-8B for conversational, steerable recommendation generation.
https://github.com/eugeneyan/semantic-ids-llm
Qwen3-8B finetuning incorporates RQ-VAE Semantic IDs for unified LLM recommenders.

Tech stack