Unpacking the bias of large language models | MIT News
... framework that can be used to diagnose and correct position bias in future model designs, leading to more accurate, reliable AI agents.
Read the original at news.mit.edu ↗
- Topics
- Bias & FairnessChatbots & LLMsGovernance & Risk ManagementHealthcareAI Agents & Automation
- Mentions
- MIT
- Tracked via
- “AI Bias”