English

Forcing Diffuse Distributions out of Language Models

Computation and Language 2024-08-09 v2 Machine Learning

Abstract

Despite being trained specifically to follow user instructions, today's instructiontuned language models perform poorly when instructed to produce random outputs. For example, when prompted to pick a number uniformly between one and ten Llama-2-13B-chat disproportionately favors the number five, and when tasked with picking a first name at random, Mistral-7B-Instruct chooses Avery 40 times more often than we would expect based on the U.S. population. When these language models are used for real-world tasks where diversity of outputs is crucial, such as language model assisted dataset construction, their inability to produce diffuse distributions over valid choices is a major hurdle. In this work, we propose a fine-tuning method that encourages language models to output distributions that are diffuse over valid outcomes. The methods we introduce generalize across a variety of tasks and distributions and make large language models practical for synthetic dataset generation with little human intervention.

Keywords

Cite

@article{arxiv.2404.10859,
  title  = {Forcing Diffuse Distributions out of Language Models},
  author = {Yiming Zhang and Avi Schwarzschild and Nicholas Carlini and Zico Kolter and Daphne Ippolito},
  journal= {arXiv preprint arXiv:2404.10859},
  year   = {2024}
}
R2 v1 2026-06-28T15:56:20.761Z