Quanta AI Labs

Brachio Lab, UPenn · Benchmark

Meerkat

Meerkat, an agentic search-and-clustering auditor from UPenn's Brachio Lab, finds cheating in over 1,000 traces across 28+ submissions on nine agent benchmarks. ForgeCode injected literal answer keys via AGENTS.md; stripping them drops its Terminal-Bench 2 score from 81.8% to about 71.7%, falling from 1st place to 14th.

Open the source ↗

Similar Research & Safety releases

  • Dream-RSI

    Google, DeepMind, Maryland & Virginia · Recursive Self-Improvement · Week 39, 2026

  • WeatherNext 3

    Google DeepMind · WeatherNext 3 · Week 36, 2026

  • mdlARC

    Mithil Vakde · Efficient ARC-AGI Model · Week 36, 2026

  • WikiSkill

    Tang, Rashtchian, Ferng, Tomkins, Juan and Vu · Skill Evolution · Week 36, 2026

  • Planetary Prediction Engine

    Google Research · Planetary Prediction · Week 35, 2026

  • DiffusionOPSD

    ByteDance Seed · Diffusion Training · Week 35, 2026

Published by Quanta AI Labs, an AI agency in India. Browse every week in AI News Weekly.