Reproduce “PARAssist: A Framework for Personalized and Adaptive Robotic Assistance from Ambiguous User Requests” 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
Pourya Aliasghari, Goldie Nejat
PARAssist ranks assistance tasks for ambiguous requests using observations of a person’s behavior and requests to infer their assistance pattern. It combines this preference model with the current context to suggest tasks, while leaving the final choice to the user. In scripted tests, suggestions reflected different histories; in one laundry scenario, 9 of 10 trials met a researcher-defined criterion, not a measure of actual intent.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.