iXentBench · Falsifiable reasoning
Collection
One certified 2v2 LLM match on iXentBench: stated predictions checked by a blind rule engine. Demo, data and method by iXentLabs. • 3 items • Updated
AI & ML interests
Deterministic, falsifiable evaluation of LLM agents: multi-agent reasoning, theory of mind, game theory and causal spatial reasoning (iXentBench)
Recent Activity
View all activity
antoniguasch
updated a
collection 2 days ago
antoniguasch
updated a
Space 2 days ago
antoniguasch
published a
Space 2 days ago
antoniguasch
updated a
dataset 2 days ago
antoniguasch
published a
dataset 2 days ago
antoniguasch
authored a
paper 3 days ago
antoniguasch
updated a
Space 3 days ago
antoniguasch
published a
Space 3 days ago