Reproduce “Open Challenges and Opportunities in Human-Agent Systems” 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
Reuth Mirsky, Sarath Sreedharan, Nitay Alon, Stefan Sarkadi, Munindar P. Singh, Liz Sonenberg, Peter Stone, Silvia Tulli and 2 more
The chapter proposes a research agenda for agentic systems that represent people’s goals, knowledge, and authority, and adapt as relationships change. It suggests combining foundation models’ language abilities with structured partner models, shared-autonomy rules, argumentation, and human-agent co-learning. A hypothetical personal assistant illustrates the challenges; the authors report no implementation or evidence that this approach improves outcomes.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.