Benchmark Almanac

SAA Lab

We track 2 benchmarks that SAA Lab made or helped make.

0 in use 0 with public code 0 with a leaderboard here

Benchmarks

Name Kind Released Best score Status
LitBench Judges predict human preferences across 2,480 held-out story pairs. The hub port instead samples the training corpus and adds self-critique. Taste and style Jul 2025 Not in cards
SLPHelm UltraSuitePlus Models analyze recordings of children’s speech for speech sound disorders. The suite also includes transcription and symptom classification tasks. Speech recognition and translation Sep 2025 Not in cards