Lucy G Cheke

I Spy With My Model’s Eye: Visual Search as a Behavioural Test for MLLMs

Inferring Capabilities from Task Performance with Bayesian Triangulation

Leaving the barn door open for Clever Hans: Simple features predict LLM benchmark answers

100 instances is all you need: predicting LLM success by testing on a few instances

A little less conversation, a little more action, please: Investigating the physical common-sense of LLMs in a 3D embodied environment