Same chore, different bathroom. A different toilet shape, a tighter space, or a new cleaning tool changes how the robot needs to move. Training data and evaluations should reflect the variety robots will encounter in real homes.
1
39
Human guidance is part of the experimental setup. For agent research, recording when experts redirected the search and what they verified makes a result easier to interpret, reproduce, and turn into the next evaluation task.
We’re sharing new research demonstrating how AI agents can accelerate the science of stress-testing encryption.
Guided by humans, our models uncovered two major weaknesses in legacy 1024-bit RSA cryptosystems.
Our VP Science @Pushmeet explains more ↓ xitter.cf/pushmeet/status/211875…
2
35
Before scaling a robot dataset, test the task boundaries.
Can the policy still pick up the cup when its pose changes, the handle is partly hidden, or objects shift inside?
A pilot should map these conditions so the next collection batch has a clear purpose.
4
64
Image editing needs two scores: did the requested change happen, and did everything else stay intact?
Across repeated edits, track identity drift and unintended changes separately from visual appeal. A single preference score can hide the tradeoff.
Images, upgraded—introducing ChatGPT Images 2.6.
Now with GIFs and reference images.
Plus, two new models are coming to the API with 50% faster image generation.
4
47
Great to see a small team deliver a release like Ripple. For teams working with limited compute, task-level evals can help pinpoint gaps in data coverage. Knowing which examples to collect or improve next helps make each training run count.
Love this scrappy project. Hoping more US labs follow suit, showing they can also make solid models with small focused teams. This also intellectually follows on the olmo lineage, so super cool to see what our baby looks like scaled up.
4
110