5 papers · 1 filter
Understanding Chain-of-Thought Effectiveness in Code Generation: An Empirical and Information-Theoretic Analysis
Naizhu Jin, Zhong Li, Guang Yang +2
Large language models (LLMs) achieve strong performance on code generation, but the mechanisms by which Chain-of-Thought (CoT) prompting helps remain unclear. We present a systemat…
Who is Introducing the Failure? Automatically Attributing Failures of Multi-Agent Systems via Spectrum Analysis
Yu Ge, Linna Xie, Zhong Li +2
Large Language Model Powered Multi-Agent Systems (MASs) are increasingly employed to automate complex real-world problems, such as programming and scientific discovery. Despite the…
GUARD:Dual-Agent based Backdoor Defense on Chain-of-Thought in Neural Code Generation
Naizhu Jin, Zhong Li, Tian Zhang +1
With the widespread application of large language models in code generation, recent studies demonstrate that employing additional Chain-of-Thought generation models can significant…
MSCoT: Structured Chain-of-Thought Generation for Multiple Programming Languages
Naizhu Jin, Zhong Li, Tian Zhang +1
With the rapid development of code intelligence, the application of multiple programming languages is becoming increasingly widespread. However, most existing code generation model…
SABER: Model-agnostic Backdoor Attack on Chain-of-Thought in Neural Code Generation
Naizhu Jin, Zhong Li, Yinggang Guo +3
Recent studies have proposed integrating Chain-of-Thought (CoT) reasoning to further enhance the reliability of Code Language Models (CLMs) in generating code, a step-by-step appro…