Benchmark RadarAI BENCHMARK PROFILE
GroupToM-Bench
GroupToM-Bench is a multimodal benchmark for evaluating group-level theory of mind through a seven-level cognitive audit framework.
- Released
- 2026-06-02
- Readiness
- Paper only
- Primary field
- General AI
Why it matters
It probes a gap in social cognition capabilities of multimodal LLMs, with implications for understanding emergent group behavior.
Motivation
True general intelligence requires not only a model of the physical world but also a social world model: the capacity to infer how individual mental states interact and crystallize into group-level outcomes.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.