Benchmark RadarAI BENCHMARK PROFILE
TREAT
TREAT evaluates LLMs' ability to recognize theorem identities from equivalence-preserving formula transformations, with 737 identities and 29,480 transformed rows.
- Released
- 2026-07-29
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It targets representation-robust access to formal knowledge, critical for AI tools interacting with formal systems.
Motivation
AI systems increasingly operate between flexible input representations and formal objects used by downstream tools.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.