arXiv · 2610.37973
Merge and Branch: Efficient Tree-Structured Supervised Fine-Tuning of LLMs for Recommendation
Abstract
LLM-based generative recommendation reads long, text-rich interaction histories but generates only short semantic identifiers (SIDs). This asymmetry makes supervised fine-tuning inefficient: encoding the history is expensive, yet each history usually provides supervision for only one target. We present Merge and Branch (M&B) to reduce input cost and learn more from each encoded history. Merge compresses each historical item's SID and text into a small number of learned tokens. Branch uses the compressed history to predict multiple target SIDs. To share history computation while keeping these targets independent, M&B organizes each example as a shallow, wide tree. We develop tree FlashAttention to execute this layout efficiently, processing the shared history once and propagating gradients from every branch for end-to-end training. A tree-structured key/value cache extends history sharing to constrained decoding. Experiments on public benchmarks show that M&B outperforms strong recommendation baselines while substantially improving training and decoding efficiency. Online A/B tests on a large-scale e-commerce platform suggest that M&B effectively combines item text with multi-target supervision, yielding statistically significant gains in per-user clicks, orders, and GMV.
Explore related subjects
Keep this discovery
Explore connections, maps & timelines
Yunqin Zhu, Liyang Shu, Yusi Zhang, Yao Xie, Tianhao Li, Xinhang Song, Tianyu Wen. 2026-11-05. Merge and Branch: Efficient Tree-Structured Supervised Fine-Tuning of LLMs for Recommendation. https://searcharxiv.com/abs/2610.37973
Citations retain the original authors and DOI and include this site’s record URL. If this discovery service contributed to your research, you can also cite Searcharxiv by Indiciae: BibTeX · RIS.