Research program
Memory and continual learning
The question
How can a system keep learning from experience safely, without retraining from scratch?
Most models stop learning once training ends. A general system needs to remember people, projects and lessons, and to improve over time, while keeping private data private and letting users delete what it knows about them.
Our predictions
Our view, with how sure we are. Each one links to the evidence it rests on.
Progress around the world
Public results from other teams that shape our view.
· ARC Prize Foundation (testing OpenAI's model)
GPT-6 Astra scores 99.9% on ARC-AGI-3, with a caveat
ARC Prize verified 62.7% with its standard harness and 99.9% with a harness that keeps the model's reasoning state between steps; in one setting it used fewer actions than the median person on 96% of levels. ARC Prize states it is not claiming this is AGI.
What it means for AGI: the gap on learning new tasks closed within six months of the test's launch. The jump from 62.7% to 99.9% came from how the system keeps memory, not from a new model, which supports the view that AGI is a system, not one model. The run cost about $18,800, far more than a person would.
· Stanford University and Google researchers (arXiv)
Generative agents simulate a small society
Twenty-five agents with memory, reflection and planning lived in a simulated town and showed believable individual and group behaviour.
What it means for AGI: memory and reflection give agents continuity, and groups of agents produce behaviour no single prompt produces.
Source: arxiv.org(opens external site)
Our projects
No projects of ours here yet.
Publications
No publications yet.
Collaborations
Open to collaboration.