NeurIPS, the Conference and Workshop on Neural Information Processing Systems, is an international conference on machine learning and computational neuroscience. Held every December, it is a premier international conference in the field of machine learning. The 2025 NeurIPS conference will be held from November 30 to December 7 in San Diego, California, USA. Eight papers with faculty and students of the School of Artificial Intelligence, Jilin University as main authors have been accepted by the conference.
1. LLM Meeting Decision Trees on Tabular Data
First author: Hangting Ye (PhD student, 2022 cohort)
Corresponding authors: Dandan Guo, Yi Chang
2. Analytic Energy-Guided Policy Optimization for Offline Reinforcement Learning
First author: Jifeng Hu
Corresponding authors: Hechang Chen, Yi Chang
3. Tackling Continual Offline RL through Selective Weights Activation on Aligned Spaces
First author: Jifeng Hu
Corresponding authors: Hechang Chen, Yi Chang
4. InstructFlow: Adaptive Symbolic Constraint-Guided Code Generation for Long-Horizon Planning
First author: Haotian Chi (PhD student, 2021 cohort)
Corresponding authors: Hechang Chen, Yin Haiyan, Yi Chang
5. Accelerated Vertical Federated Adversarial Learning through Decoupling Layer-Wise Dependencies
First authors: Tianxing Man, Yu Bai (Master's student, 2023 cohort)
Corresponding authors: Bin Gu, Yi Chang
6. StruDiCO: Structured Denoising Diffusion with Gradient-free Inference-stage Boosting for Memory and Time Efficient Combinatorial Optimization
First author: Yu Wang (Dingxin Scholar)
Corresponding authors: Junchi Yan, Yi Chang
7. PointMAC: Meta-Learned Adaptation for Robust Test-Time Point Cloud Completion
First author: Linlian Jiang (Master's graduate, 2024)
Authors: Linlian Jiang, Rui Ma, Li Gu, Ziqiang Wang, Xinxin Zuo, Yang Wang
8. Data Efficient Adaptation in Large Language Models via continuous Low-Rank Fine-Tuning
Authors: Xiao Han, Zimo Zhao, Wanyu Wang, Maolin Wang, Zitao Liu, Yi Chang, Xiangyu Zhao
