Reproduce “Behavioral Indicators of Overreliance During Interaction with Conversational Language Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Chang Liu, Qinyi Zhou, Xinjie Shen, Xingyu Bruce Liu, Tongshuang Wu, Xiang 'Anthony' Chen
The authors examine whether mouse and keyboard actions during language-model-assisted tasks correlate with accepting incorrect AI recommendations. Across quiz solving, summarization, and trip planning, they encode and cluster interaction sequences and compare them with outcomes. In trip-planning windows, higher-overreliance clusters show 5–6 copy/paste actions per 60 seconds, versus 1–3 in lower-overreliance clusters; this association is not causal or diagnostic.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.