Prompting Techniques for Reducing Social Bias in LLMs through System 1 and System 2 Cognitive Processes
This addresses bias mitigation in AI for fairness applications, though it is incremental as it builds on existing prompting methods.
The study tackled reducing social bias in large language models by exploring prompting techniques based on dual process theory, finding that combinations like human personas and chain-of-thought prompting can reduce stereotypical judgments by up to 33% across various bias categories.
Dual process theory posits that human cognition arises via two systems. System 1, which is a quick, emotional, and intuitive process, which is subject to cognitive biases, and System 2, is a slow, onerous, and deliberate process. Prior research in LLMs found that using chain-of-thought (CoT) prompting in LLMs, which has been often compared to System 2 reasoning, can lead to reduced gender bias. Along these lines, we investigate the relationship between bias, CoT prompting, a direct debiasing, and dual process theory modeling in LLMs. We compare zero-shot CoT, debiasing, and dual process theory-based prompting strategies on two bias datasets spanning nine different social bias categories. We incorporate human and machine personas to determine whether LLM modeling of the effects of dual process theory exist independent of explicit persona models or are tied to the LLM's modeling of human-like generation. We find that a human persona, debiasing, System 2, and CoT prompting all tend to reduce social biases in LLMs, though the best combination of features depends on the exact model and bias category -- resulting in up to a 33 percent drop in stereotypical judgments by an LLM.