Thread by @bentossell

There’s an AI conference going on today called NeurIPS2022.

I’ve seen a few threads of what’s going on and the papers presented:

Interesting things from the conference from Meta.

https://twitter.com/1034844617261248512/status/1597712822087000064

Amazon Science presentation

https://twitter.com/AmazonScience/status/1598041125171904512

All the papers Amazon Science is presenting:

amazon.science/blog/a-quick-g…

The 40 papers and 10 demos @DeepMind is presenting at the conference (link in the attached tweet).

https://twitter.com/4783690002/status/1597251493752868864

@DeepMind Lol

https://twitter.com/MichaelTrazzi/status/1597735498247974914

@DeepMind List of the 150 Transformer papers at NeurIPS Conference:

https://twitter.com/329683616/status/1597237870028533761

@DeepMind Thread about “conditional diffusion models” from @du_yilun:

https://twitter.com/851865313/status/1597618021342023680

@DeepMind @du_yilun How neural nets can generate weights (what variables most impact prediction) even unsupervised:

https://twitter.com/1151388211752636418/status/1597625474934796291

@DeepMind @du_yilun 7 Papers from @james_y_zou

https://twitter.com/762359343656361984/status/1597238984656138240

@DeepMind @du_yilun @james_y_zou Overview of a few papers and posters:

https://twitter.com/lyeskhalil/status/1597630998518280192

@DeepMind @du_yilun @james_y_zou Forging a link between path-integration and existence of hexagonal grids:

https://twitter.com/RylanSchaeffer/status/1597606723308879872

@DeepMind @du_yilun @james_y_zou Video of Test-time prompt tuning framework from a NeurIPS paper:

https://twitter.com/1312627190983348224/status/1597458719801237504

@DeepMind @du_yilun @james_y_zou Gradient estimator for discrete distributions:

https://twitter.com/702089336842375169/status/1595217877132451840

@DeepMind @du_yilun @james_y_zou @davidchalmers42 talk about AI Challenges, LLM+ Consciousness.

From a listener of the speech “10% probability that LLMs boast some level of consciousness”

https://twitter.com/davidchalmers42/status/1597487979383394306

@DeepMind @du_yilun @james_y_zou @davidchalmers42 @SaxeLab paper describing how prior knowledge impacts learning dynamics.

https://twitter.com/SaxeLab/status/1597179383412051968

@DeepMind @du_yilun @james_y_zou @davidchalmers42 @SaxeLab @ema_marconato paper describing how neural networks can explain their predictions.

https://twitter.com/1483799104949993476/status/1597054655921430530

@DeepMind @du_yilun @james_y_zou @davidchalmers42 @SaxeLab @ema_marconato A professor(?) highlighting some of his students key results:

https://twitter.com/2733642475/status/1596547879430144003

@DeepMind @du_yilun @james_y_zou @davidchalmers42 @SaxeLab @ema_marconato “Analagous score function for discrete data”.

https://twitter.com/1234195291202473985/status/1597459801826799616

@DeepMind @du_yilun @james_y_zou @davidchalmers42 @SaxeLab @ema_marconato Paper on how to create a massive database of text (TBs) to train language models:

https://twitter.com/1382962150075957248/status/1597262246278680576

Lmk if there are other interesting ones you want me to include!