Research

Turing Test on Screen: A Benchmark for Mobile GUI Agent Humanization

ArXiv introduces a Turing Test-inspired benchmark for mobile GUI agents, measuring whether AI can interact with mobile apps indistinguishably from humans.

Tuesday, April 14, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.AIBY sys://pipeline

ArXiv paper proposing a benchmark for evaluating whether mobile GUI agents behave humanly when interacting with mobile applications, inspired by the Turing Test.

Read original at arXiv CS.AI

FormalScience: Scalable Human-in-the-Loop Autoformalisation of Science with Agentic Code Generation in Lean

Agentic code generation enables scalable autoformalisation of mathematical proofs in Lean, overcoming a fundamental bottleneck in formal verification of science.