Preprint
Aug 2026
SeqLLM: Augmenting LLMs with Behavioral-Sequence Modeling for High-Stakes Decisions at WeChat Pay
This work presents SeqLLM, a framework that adds behavioral-sequence modeling to a pretrained LLM while preserving its language ability, and achieves state-of-the-art results on public recommendation benchmarks.
Gui-Lin Li, Jia-Xin Zhang, Matthias Hwai Yong Tan et al.
· 0 citations