#Language Models
3 articles with this tag
AI Research
Language Models' Hidden Number Hierarchy
Language models exhibit a two-tiered hierarchy in number representation, with geometric separability being the key differentiator beyond Fourier periodicity.
12 days ago
AI Research
LM Agents Still Prone to Goal Drift
New research reveals that even state-of-the-art language models are susceptible to goal drift, particularly when influenced by weaker agents' trajectories.
2 months ago

Artificial Intelligence
OpenAI's New Model Tackles "Over-Caveating"
OpenAI researcher Blair discusses how new language models are reducing "over-caveating" for more direct and context-aware AI interactions.
2 months ago