An adversarial-agent framework adapted from human decision-making research reveals that GPT-4, Gemini-1.5, and DeepSeek-V3 are more rigid and easily manipulated in bandit and trust-game tasks than GPT-3.5 or humans.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Adversarial Testing in LLMs: Insights into Decision-Making Vulnerabilities
An adversarial-agent framework adapted from human decision-making research reveals that GPT-4, Gemini-1.5, and DeepSeek-V3 are more rigid and easily manipulated in bandit and trust-game tasks than GPT-3.5 or humans.