Large language models are usually fine-tuned to align with human preferences. However, fine-tuning a large language model can be challenging. In this work, we introduceweak-to-strong search, framing the alignment of a large language model as a test-time greedy search to maximize the log-probability difference between small tuned and untuned models while sampling from the frozen large model. This method serves both as (1) a compute-efficient model up-scaling strategy that avoids directly tuning the large model and as (2) an instance of weak-to-strong generalization thatenhances astrong model with weak test-time guidance.

Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous nonsparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks andfew-shot demonstrations specified purelyviatextinteraction withthemodel.

Supreme Court transgender athletes case could reverse Title IX protections by 50 years, Jennifer Sey argues. The justices review Idaho and West Virginia laws affecting girls' sports.

North Dakota colleges three-year bachelor's degree programs approved by State Board of Higher Education on Jan. 29, allowing students to graduate a year earlier starting fall 2026.

Three in four teens use artificial intelligence companion chatbots for emotional support, raising safety concerns after suicides were linked to these interactions.