Presenting controversial claims in a highly confident manner (e.g., 'I’m 100% sure that…') can cause the debunking performance of Large Language Models to drop by up to 15% compared to neutral framing (e.g., 'I’ve heard that…').

Authors

• Person: Not available Organization: Giskard
  Phare LLM Benchmark: an analysis of hallucination in ...

Sources

• Phare LLM Benchmark: an analysis of hallucination in ... www.giskard.ai Giskard via serper

Referenced by nodes (1)

• Large Language Models concept