Papers by Dominik Karasin
Characterizing Positional Bias in Large Language Models: A Multi-Model Evaluation of Prompt Order Effects (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Large Language Models can be influenced by various forms of biases, says a new study . positional bias affects how LLMs interpret and weigh information, the authors say . |
| Approach: | a new study examines the impact of positional bias on large language models . positional biased models prioritize items based on their position rather than content or quality . |
| Outcome: | a new study shows that LLMs prioritize items based on their position rather than content or quality . the positional bias affects how LLM interpret and weigh information, the authors say . |