Evaluation design and benchmarks · MLD-0643
Length bias
A systematic preference for particular output lengths that is not justified by task quality.
Original reference page · detailed expansion pending. Original definitions, examples, notes and citations are preserved below. New examples are illustrative, not benchmark results.
Original example
Test whether longer answers receive inflated judge scores.
Original caveat
Length can sometimes be legitimately relevant to the rubric.
Context and aliases
Level: Intermediate. Stages: Evaluation · Research.
Tags: ML · DL · LLM · SLM · Safety.
Aliases: Verbosity bias.