8 papers
PolicyPad: Collaborative Prototyping of LLM Policies
K. J. Kevin Feng, Tzu-Sheng Kuo, Quan Ze Chen +3
As LLMs gain adoption in high-stakes domains like mental health, domain experts are increasingly consulted to provide input into policies governing their behavior. From an observat…
Cocoa: Co-Planning and Co-Execution with AI Agents
K. J. Kevin Feng, Kevin Pu, Matt Latzke +6
As AI agents take on increasingly long-running tasks involving sophisticated planning and execution, there is a corresponding need for novel interaction designs that enable deeper…
Levels of Autonomy for AI Agents
K. J. Kevin Feng, David W. McDonald, Amy X. Zhang
Autonomy is a double-edged sword for AI agents, simultaneously unlocking transformative possibilities and serious risks. How can agent developers calibrate the appropriate levels o…
Policy Prototyping for LLMs: Pluralistic Alignment via Interactive and Collaborative Policymaking
K. J. Kevin Feng, Inyoung Cheong, Quan Ze Chen +1
Emerging efforts in AI alignment seek to broaden participation in shaping model behavior by eliciting and integrating collective input into a policy for model finetuning. While plu…
Canvil: Designerly Adaptation for LLM-Powered User Experiences
K. J. Kevin Feng, Q. Vera Liao, Ziang Xiao +3
Advancements in large language models (LLMs) are sparking a proliferation of LLM-powered user experiences (UX). In product teams, designers often craft UX to meet user needs, but i…
SPICA: Retrieving Scenarios for Pluralistic In-Context Alignment
Quan Ze Chen, K. J. Kevin Feng, Chan Young Park +1
When different groups' values differ, one approach to model alignment is to steer models at inference time towards each group's preferences. However, techniques like in-context lea…