4 papers
Steerable Pluralism: Pluralistic Alignment via Few-Shot Comparative Regression
Jadie Adams, Brian Hu, Emily Veenhuis +5
Large language models (LLMs) are currently aligned using techniques such as reinforcement learning from human feedback (RLHF). However, these methods use scalar rewards that can on…
ALIGN: Prompt-based Attribute Alignment for Reliable, Responsible, and Personalized LLM-based Decision-Making
Bharadwaj Ravichandran, David Joy, Paul Elliott +6
Large language models (LLMs) are increasingly being used as decision aids. However, users have diverse values and preferences that can affect their decision-making, which requires…
GeoWATCH for Detecting Heavy Construction in Heterogeneous Time Series of Satellite Images
Jon Crall, Connor Greenwell, David Joy +3
Learning from multiple sensors is challenging due to spatio-temporal misalignment and differences in resolution and captured spectra. To that end, we introduce GeoWATCH, a flexible…
Language Models are Alignable Decision-Makers: Dataset and Application to the Medical Triage Domain
Brian Hu, Bill Ray, Alice Leung +4
In difficult decision-making scenarios, it is common to have conflicting opinions among expert human decision-makers as there may not be a single right answer. Such decisions may b…