John deVadoss

co-Founder NeuralFabric acq. by @Cisco | co-Founder @IntWorkAll | Board @GBBC_io | General Manager @Microsoft | Phd RL research @UMassAmherst

[John deVadoss @john_devadoss]

A Public AI Wealth fund, not 'basic income'. It is time for Congress to act.

https://t.co/t40f66tO4S

[Ryan Bahlous-Boldi @RyanBoldi]

Your RL post-training may be sabotaging your LLM’s test-time scaling!

Conventional RL pretends that you can collapse all reward signals *upfront* into a single *scalar reward*.
We introduce Vector Policy Optimization (VPO), which natively maximizes *vector-valued* rewards, boosting test time search performance, even on the original scalar.

[Souradip Chakraborty @SOURADIPCHAKR18]

🚨Typical RL algorithms and on-policy distillation methods are blind samplers: they use privileged info to score rollouts, but not to *find* them.

We ask: can we use privileged info to *actively sample* the rollouts RL wishes it can stumble upon with compute?

⤵️ Pedagogical RL

[IEEE Spectrum @IEEESpectrum]

Don’t Regulate AI Models. Regulate AI Use. https://t.co/9cksNuReWu

[John deVadoss @john_devadoss]

@RLanceMartin Thank you for writing this up.