Nanxin Zeng — accepted AI-conference papers

2026

Certain Head, Uncertain Tail: Expert-Sample for Test-Time Scaling in Fine-Grained MoE

Yuanteng Chen, Peisong Wang, Nanxin Zeng, Yuantian Shao, Shuang Qiu, Gang Li +2

ICML 2026poster

Test-time scaling improves LLM performance by generating multiple candidate solutions, yet token-level sampling requires temperature tuning that trades off diversity against stability. Fine-grained MoE, featuring hundreds of well-trained experts per layer and multi-expert activation per token, offer…

Cited by 0 Source ↗Scholar ↗

2025

SimWorld: A Unified Benchmark for Simulator-Conditioned Scene Generation via World Model

Xinqing Li, Ruiqi Song, Qingyu Xie, Ye Wu, Nanxin Zeng, Yunfeng Ai

IROS 2025

With the rapid advancement of autonomous driving technology, a lack of data has become a major obstacle to enhancing perception model accuracy. Researchers are now exploring controllable data generation using world models to diversify datasets. However, previous work has been limited to studying ima

Cited by 4 Source ↗code ↗Scholar ↗