33 citations · 34 across the 7 of their papers we have counts for
7 papers
Verifier-Backed Hard Problem Generation for Mathematical Reasoning
Yuhang Lai, Jiazhan Feng, Yee Whye Teh +1
Large Language Models (LLMs) demonstrate strong capabilities for solving scientific and mathematical problems, yet they struggle to produce valid, challenging, and novel problems -…
Step-Level Sparse Autoencoder for Reasoning Process Interpretation
Xuan Yang, Jiayu Liu, Yuhang Lai +3
Large Language Models (LLMs) have achieved strong complex reasoning capabilities through Chain-of-Thought (CoT) reasoning. However, their reasoning patterns remain too complicated…
How Jailbreak Defenses Work and Ensemble? A Mechanistic Investigation
Zhuohang Long, Siyuan Wang, Shujun Liu +3
Jailbreak attacks, where harmful prompts bypass generative models' built-in safety, raise serious concerns about model vulnerability. While many defense methods have been proposed,…
HAF-RM: A Hybrid Alignment Framework for Reward Model Training
Shujun Liu, Xiaoyu Shen, Yuhang Lai +5
The reward model has become increasingly important in alignment, assessment, and data construction for large language models (LLMs). Most existing researchers focus on enhancing re…
ALaRM: Align Language Models via Hierarchical Rewards Modeling
Yuhang Lai, Siyuan Wang, Shujun Liu +2
We introduce ALaRM, the first framework modeling hierarchical rewards in reinforcement learning from human feedback (RLHF), which is designed to enhance the alignment of large lang…
EVOR: Evolving Retrieval for Code Generation
Hongjin Su, Shuyang Jiang, Yuhang Lai +5
Recently the retrieval-augmented generation (RAG) has been successfully applied in code generation. However, existing pipelines for retrieval-augmented code generation (RACG) emplo…
DS-1000: A Natural and Reliable Benchmark for Data Science Code Generation
Yuhang Lai, Chengxi Li, Yiming Wang +7
We introduce DS-1000, a code generation benchmark with a thousand data science problems spanning seven Python libraries, such as NumPy and Pandas. Compared to prior works, DS-1000…