GPT models asked to make random binary choices show large, version-specific biases; only GPT-4o-Mini came close to a 50/50 split in one-shot tests.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Evaluating Binary Decision Biases in Large Language Models: Implications for Fair Agent-Based Financial Simulations
GPT models asked to make random binary choices show large, version-specific biases; only GPT-4o-Mini came close to a 50/50 split in one-shot tests.