1. EachPod
EachPod
Mechanical Dreams - Podcast

Mechanical Dreams

 Mechanical Dirk

An automatically generated podcast about machine learning and natural language processing. The two fictional hosts talk about papers that I want to learn more about on my way to work.

It's not good, but it's useful.

Science Natural Sciences
Update frequency
every 2 days
Average duration
10 minutes
Episodes
81
Years Active
2024 - 2025
Share to:
Native Sparse Attention

Native Sparse Attention

00:11:40  |   Wed 04 Jun 2025
Critical Batch Size Revisited

Critical Batch Size Revisited

00:10:57  |   Tue 03 Jun 2025
Rope to Nope and Back Again

Rope to Nope and Back Again

00:12:10  |   Sat 17 May 2025
Base of RoPE Bounds Context Length

Base of RoPE Bounds Context Length

00:11:02  |   Sat 17 May 2025
Training Deep Learning Models with Norm-Constrained LMOs

Training Deep Learning Models with Norm-Constrained LMOs

00:11:09  |   Thu 15 May 2025
SkyLadder

SkyLadder

00:12:25  |   Fri 09 May 2025
LLMs on the Line

LLMs on the Line

00:09:30  |   Wed 07 May 2025
The Leaderboard Illusion

The Leaderboard Illusion

00:10:15  |   Wed 30 Apr 2025
Why Linearly Decaying the Learning Rate to Zero Works Best

Why Linearly Decaying the Learning Rate to Zero Works Best

00:09:01  |   Wed 16 Apr 2025
Not All Data Are Unlearned Equally

Not All Data Are Unlearned Equally

00:12:38  |   Tue 15 Apr 2025
A Multi-Power Law for Loss Curve Prediction

A Multi-Power Law for Loss Curve Prediction

00:12:31  |   Mon 14 Apr 2025
Efficient Training of Ultra-Long Context Large Language Models

Efficient Training of Ultra-Long Context Large Language Models

00:10:48  |   Fri 11 Apr 2025
Multi-Token Attention

Multi-Token Attention

00:15:04  |   Thu 03 Apr 2025
From Style to Facts

From Style to Facts

00:10:50  |   Wed 02 Apr 2025
Compute Optimal Scaling of Skills

Compute Optimal Scaling of Skills

00:09:09  |   Sat 22 Mar 2025
Predictive Data Selection

Predictive Data Selection

00:08:43  |   Sat 15 Mar 2025
Continual Pre-training of MoEs

Continual Pre-training of MoEs

00:10:41  |   Wed 12 Mar 2025
s1 - Simple test-time scaling

s1 - Simple test-time scaling

00:10:37  |   Thu 06 Mar 2025
Cognitive Behaviors that Enable Self-Improving Reasoners

Cognitive Behaviors that Enable Self-Improving Reasoners

00:07:20  |   Wed 05 Mar 2025
Phi 4 Multimodal Instruct

Phi 4 Multimodal Instruct

00:11:26  |   Tue 04 Mar 2025
Disclaimer: The podcast and artwork embedded on this page are the property of Mechanical Dirk. This content is not affiliated with or endorsed by eachpod.com.