Deterministic or probabilistic? The psychology of LLMs as random number generators
This work addresses the problem of understanding and mitigating biases in LLMs for researchers and developers, but it is incremental as it builds on existing studies of model biases.
The paper investigated the performance of various LLMs in generating random numbers under different configurations, revealing that despite their stochastic architecture, they often produce deterministic responses due to biases in training data.
Large Language Models (LLMs) have transformed text generation through inherently probabilistic context-aware mechanisms, mimicking human natural language. In this paper, we systematically investigate the performance of various LLMs when generating random numbers, considering diverse configurations such as different model architectures, numerical ranges, temperature, and prompt languages. Our results reveal that, despite their stochastic transformers-based architecture, these models often exhibit deterministic responses when prompted for random numerical outputs. In particular, we find significant differences when changing the model, as well as the prompt language, attributing this phenomenon to biases deeply embedded within the training data. Models such as DeepSeek-R1 can shed some light on the internal reasoning process of LLMs, despite arriving to similar results. These biases induce predictable patterns that undermine genuine randomness, as LLMs are nothing but reproducing our own human cognitive biases.