Signin

Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters (Paper)

Yannic Kilcher • October 30, 2024

Video Thumbnail

You May Also Like

Video Thumbnail

Learning at test time in LLMs

Machine Learning Street Talk

Video Thumbnail

[GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Yannic Kilcher

Video Thumbnail

TokenFormer: Rethinking Transformer Scaling with Tokenized Model Parameters (Paper Explained)

Yannic Kilcher

Video Thumbnail

GSM-Symbolic: Understanding the Limitations of Mathematical Reasoning in Large Language Models

Yannic Kilcher

Video Thumbnail

Charlie Snell, UC Berkeley. Title: Scaling LLM Test-Time Compute

AI4All

Video Thumbnail

AI can't cross this line and we don't know why.

Welch Labs

Video Thumbnail

ORPO: Monolithic Preference Optimization without Reference Model (Paper Explained)

Yannic Kilcher

Video Thumbnail

Speculations on Test-Time Scaling (o1)

Sasha Rush

Video Thumbnail

Were RNNs All We Needed? (Paper Explained)

Yannic Kilcher

Video Thumbnail

Mamba: Linear-Time Sequence Modeling with Selective State Spaces (Paper Explained)

Yannic Kilcher

Video Thumbnail

50 Year Old Venus Spacecraft Is About To Land On EARTH?????

Scott Manley

Video Thumbnail

It's Not About Scale, It's About Abstraction

Machine Learning Street Talk

Video Thumbnail

Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention

Yannic Kilcher

Video Thumbnail

Safety Alignment Should be Made More Than Just a Few Tokens Deep (Paper Explained)

Yannic Kilcher

Video Thumbnail

Why Does Diffusion Work Better than Auto-Regression?

Algorithmic Simplicity

Video Thumbnail

Flow Matching for Generative Modeling (Paper Explained)

Yannic Kilcher

Video Thumbnail

ROCKET that Burns WATER as FUEL?! Homemade Hydrogen Engine No One Talks About

Skill Make

Video Thumbnail

The Dark Matter of AI [Mechanistic Interpretability]

Welch Labs

Video Thumbnail

On the Biology of a Large Language Model (Part 1)

Yannic Kilcher

Video Thumbnail

How DeepSeek Rewrote the Transformer [MLA]

Welch Labs

Yannic Kilcher

About

No channel description available.

Latest Posts

Video Thumbnail

All about AI Accelerators: GPU, TPU, Dataflow, Near-Memory, Optical, Neuromorphic & more (w/ Author)

Yannic Kilcher

Video Thumbnail

xLSTM: Extended Long Short-Term Memory

Yannic Kilcher

Video Thumbnail

Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention

Yannic Kilcher

Video Thumbnail

[GRPO Explained] DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models

Yannic Kilcher