rizz.dev
homeaboutblogcontactGitHub
  1. Home
  2. /
  3. Glossary
  4. /
  5. BBH
Glossary · term19 of 202

BBH

Big-Bench Hard, a set of difficult reasoning benchmarks where compressed prompts lose little because the answer is short and reconstructable from intent.

also written as big-bench hard

Related terms
GSM8KA benchmark of grade-school math word problems commonly used to test reasoning accuracy under prompt compression.view term->LLMLinguaA prompt-compression method that reports roughly 20x token reduction at about a 1.5-point accuracy cost, measured on short-answer reasoning benchmarks rather than agent tasks.view term->
At a glance
cited by
1 post
categories
1
first used
jul 2026
Appears in
  • Token-Compression Skills Do Not Survive MeasurementToken compression skills report savings the invoice never shows. Cache reads carry 87 percent of an agent bill, and cutting tokens can push the cost up.meta-analysis1 min
previousautonomy evaluationnextbilled iteration