Frontier models still show greediness and frequency bias, researchers say
m_wulfmeier · x · 2026-07-21
The author says frontier models still show two decision biases: greediness, where the model collapses onto the highest-prior answer, and frequency bias, where salient context outweighs evidence.
They add that some training recipes can reduce these biases over the long term, including RL fine-tuning on self-generated rationales, as shown in their greedy agents paper.
More from Research
- FrontierML benchmark shows open models still far behind closed ones on end-to-end ML work — scaling01 · 2026-10-27
- KANLab: an interactive browser playground for Kolmogorov-Arnold Networks — Lukas Niehaus · 2026-10-27
- CMU's LLM2Cert synthesizes stability certificates verified via SMT and Lean — CarnegieMellonU · 2026-10-27
- NVIDIA open-sources FLORA, a robotics data engine ~125x cheaper than GPT-6 Astra — nvidia · 2026-10-27
- RSNA 2026 Deep Learning Labs return: eight hands-on AI classes over four days in Chicago — PRouzrokh · 2026-10-27
- Steering two orthogonal concept neurons makes a model compose "gnome" — Sauers_ · 2026-10-27