Qualitative analysis of 306 rejected AI-generated PRs reveals 14 rejection reasons in four categories, highlighting needs for better guidance on implementation, validation, and prioritization.
arXiv:cs.HC/2405.16310 https://arxiv.org/abs/2405.16310
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
CodeQ aggregates token rationales into code categories to enable global interpretability of LLMs, claiming over 50% entropy reduction and revealing model preference for syntactic cues plus human misalignment in a 37-person study.
The paper consolidates risks of overreliance on LLMs, identifies gaps in current measurement approaches, and proposes mitigation strategies to keep AI as a human-compatible thought partner.
citing papers explorer
-
Understanding the Rejection of Fixes Generated by Agentic Pull Requests -- Insights from the AIDev Dataset
Qualitative analysis of 306 rejected AI-generated PRs reveals 14 rejection reasons in four categories, highlighting needs for better guidance on implementation, validation, and prioritization.
-
Enabling Global, Human-Centered Explanations for LLMs:From Tokens to Interpretable Code and Test Generation
CodeQ aggregates token rationales into code categories to enable global interpretability of LLMs, claiming over 50% entropy reduction and revealing model preference for syntactic cues plus human misalignment in a 37-person study.
-
Measuring and mitigating overreliance to build human-compatible AI
The paper consolidates risks of overreliance on LLMs, identifies gaps in current measurement approaches, and proposes mitigation strategies to keep AI as a human-compatible thought partner.