The victor in the Civil War wasn’t the ‘North’. It was the United States. We beat a violent, traitorous faction.

But that faction lived on through Jim Crow, the ‘Southern Strategy’, now MAGA…

The GOP, Fox & more are infected by the faction. The insurrection is ongoing.

The key tool being used by the faction now is the Big Lie.

They seek to limit access to the ballot box and institutionalize the ability of the faction’s current party - the GOP - to overturn election results.

They seek to defeat the US from the inside, but it’s the same war.
The goal is to rule from the minority. Their coalition is strong & disciplined and includes:

Fox, OAN, Sinclair, Daily Wire, Breitbart, White Evangelical churches, paramilitary groups, Federalist Society, Judicial Watch, Council for National Policy & the Republican Party & more.
The faction is expert at uniting a fervent base by enflaming White resentment.

It’s the same strategy they’ve always had, though the political and social groups that they’ve infected have shifted over time.

White supremacy is always at the core of their ideology.
It’s important to know who the real enemy is so that you can stop fighting battles and realize there’s a full-scale war being waged by a determined, seditious enemy.

The entire faction must be defeated, not just the GOP.

Happy #4thofJuly. May the United States prevail.

More from All

How can we use language supervision to learn better visual representations for robotics?

Introducing Voltron: Language-Driven Representation Learning for Robotics!

Paper: https://t.co/gIsRPtSjKz
Models: https://t.co/NOB3cpATYG
Evaluation: https://t.co/aOzQu95J8z

🧵👇(1 / 12)


Videos of humans performing everyday tasks (Something-Something-v2, Ego4D) offer a rich and diverse resource for learning representations for robotic manipulation.

Yet, an underused part of these datasets are the rich, natural language annotations accompanying each video. (2/12)

The Voltron framework offers a simple way to use language supervision to shape representation learning, building off of prior work in representations for robotics like MVP (
https://t.co/Pb0mk9hb4i) and R3M (https://t.co/o2Fkc3fP0e).

The secret is *balance* (3/12)

Starting with a masked autoencoder over frames from these video clips, make a choice:

1) Condition on language and improve our ability to reconstruct the scene.

2) Generate language given the visual representation and improve our ability to describe what's happening. (4/12)

By trading off *conditioning* and *generation* we show that we can learn 1) better representations than prior methods, and 2) explicitly shape the balance of low and high-level features captured.

Why is the ability to shape this balance important? (5/12)

You May Also Like