Alibaba's LazFormer transfers generative pre-training to industrial recommendation ranking
_reachsumit · x · 2026-09-15
Alibaba researchers propose LazFormer, a scalable Transformer recommender that autoregressively pre-generates sequential features and transfers dense parameters into ranking via residual adapters, addressing negative transfer and sparse-parameter overfitting while cutting compute and convergence time versus training a ranking model from scratch.
More from Research
- Microsoft and SJTU open-source Argus, an agent that ran 1,548 hours and solved a 20-year-old math problem — jiqizhixin · 2026-09-16
- Composing Continual Learning Mechanisms Boosts Long-Horizon Memorization in LMs — JohnsHopkins · 2026-09-16
- Gavel Elicits Native Skill Routing from a Frozen LLM's Hidden States — Tsinghua · 2026-09-16
- ScienceBuddy: Recursive-in-Recursive Self-Improvement for Scientific Agents — Shuhan Xue · 2026-09-16
- RCT: LLMs fail to significantly boost novices' wet-lab molecular biology success — shae_mcl · 2026-09-16
- Six years on, scvi-tools still widely used — outlasting foundation models and coding agents — anshulkundaje · 2026-09-16