verbosity-bias

Here is 1 public repository matching this topic...

tkarim45 / llm-judge-bias

Measure position/verbosity/assertiveness bias in an LLM-as-judge (Claude) by judging pairs in both orders. Finding: no position bias, but 75% verbosity bias and 100% assertiveness bias — judge scores gameable by length + tone.

evaluation bedrock bias claude llm-as-judge position-bias verbosity-bias

Updated Jul 25, 2026
Python

Improve this page

Add a description, image, and links to the verbosity-bias topic page so that developers can more easily learn about it.

Curate this topic

Add this topic to your repo

To associate your repository with the verbosity-bias topic, visit your repo's landing page and select "manage topics."

Learn more