Yuhang Lai

11 papers A* 1Journal 6Unranked 4
YearRankTypeTitle / Venue / Authors
2026 J jnl
Step-Level Sparse Autoencoder for Reasoning Process Interpretation.
CoRR
Xuan Yang, Jiayu Liu, Yuhang Lai, Hao Xu, Zhenya Huang, Ning Miao
2025 — conf
HAF-RM: A Hybrid Alignment Framework for Reward Model Training.
ACL (1)
Shujun Liu, Xiaoyu Shen, Yuhang Lai, Siyuan Wang, Shengbin Yue, Zengfeng Huang, Xuanjing Huang, Zhongyu Wei
2025 — conf
How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation.
EMNLP (Findings)
Zhuohan Long, Siyuan Wang, Shujun Liu, Yuhang Lai
2025 J jnl
How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation.
CoRR
Zhuohan Long, Siyuan Wang, Shujun Liu, Yuhang Lai, Xuanjing Huang, Zhongyu Wei
2024 — conf
ALaRM: Align Language Models via Hierarchical Rewards Modeling.
ACL (Findings)
Yuhang Lai, Siyuan Wang, Shujun Liu, Xuanjing Huang, Zhongyu Wei
2024 J jnl
ALaRM: Align Language Models via Hierarchical Rewards Modeling.
CoRR
Yuhang Lai, Siyuan Wang, Shujun Liu, Xuanjing Huang, Zhongyu Wei
2024 J jnl
ARKS: Active Retrieval in Knowledge Soup for Code Generation.
CoRR
Hongjin Su, Shuyang Jiang, Yuhang Lai, Haoyuan Wu, Boao Shi, Che Liu, Qian Liu, Tao Yu
2024 — conf
EvoR: Evolving Retrieval for Code Generation.
EMNLP (Findings)
Hongjin Su, Shuyang Jiang, Yuhang Lai, Haoyuan Wu, Boao Shi, Che Liu, Qian Liu, Tao Yu
2024 J jnl
HAF-RM: A Hybrid Alignment Framework for Reward Model Training.
CoRR
Shujun Liu, Xiaoyu Shen, Yuhang Lai, Siyuan Wang, Shengbin Yue, Zengfeng Huang, Xuanjing Huang, Zhongyu Wei
2023 A* conf
DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation.
ICML
Yuhang Lai, Chengxi Li, Yiming Wang, Tianyi Zhang, Ruiqi Zhong, Luke Zettlemoyer, Wen-tau Yih, Daniel Fried, Sida I. Wang, Tao Yu
2022 J jnl
DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation.
CoRR
Yuhang Lai, Chengxi Li, Yiming Wang, Tianyi Zhang, Ruiqi Zhong, Luke Zettlemoyer, Scott Wen-tau Yih, Daniel Fried, Sida I. Wang, Tao Yu