Collected sources and patterns will appear here. Add from search, explore, or the patterns library.
CandidateAnswer -> SelectedOption
Fall back to an LLM-based parser to extract chosen options from conversational outputs when exact pattern matching fails.
Problem it solves
Rule-based parsing fails on conversational or complex model answers, degrading benchmark evaluation accuracy.
Consumes
Emits
The real projects this mechanism was found in. Attribution is the point — this is how the best teams actually do it.