My intuition is two things:
1. Carefully curated pre-training data & finding the right data mix 2. Domain/ task specific evals for downstream use-cases In the end training data is still king! – finding which combination to go for is where the real moneys at. In addition- Post
MACHINE LEARNING
-
Training Data Mix and Domain-Specific Evals Drive AI Model Performance
By
–
-
Data Science Learning Path: Mathematics and ML Bootcamp Courses
By
–
For All who want to get Into Data Science
Mathematics-Basics to Advanced for Data Science And GenAI : https://
udemy.com/course/mathema
tics-basics-to-advanced-for-data-science-and-ml/?couponCode=HAPPYDIWALI
… Complete Data Science,Machine Learning,DL,NLP Bootcamp: https://
udemy.com/course/complet
e-machine-learning-nlp-bootcamp-mlops-deployment/?couponCode=HAPPYDIWALI
… -
Unitree robots learn humanlike walking in two days
By
–
In 2 days, Unitree's robots have learned to walk more humanely. Wow.pic.twitter.com/LDSz7aNi7M
— Chubby♨️ (@kimmonismus) 1 novembre 2024In 2 days, Unitree's robots have learned to walk more humanely. Wow.
-
Contrastive RL Enables Emergent Skills Without Rewards
By
–
A Single Goal is All You Need: Skills and Exploration Emerge from Contrastive RL without Rewards, Demonstrations, or Subgoals https://
bit.ly/4gMQUlI
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
Tool use mastered by a Minecraft-generating LLM
By
–
If I had known that "Tool use" would be first mastered by a Minecraft-generating LLM… (cf the video at 0:32)
-
The Batch: Five AI Fears Explored This Halloween
By
–
Happy Halloween! On this spooky day, The Batch continues its annual tradition of exploring fears related to AI. This special edition has 5 articles: * AI Burns All the Energy * Innovation Dies
* No Work for Coders
* Benchmarks Are Meaningless
* Synthetic Data Distorts -
SmolLM2: Faster, Better, Cheaper Language Model
By
–
SmolLM2 – faster, better and cheaper! Intelligence is definitely too cheap to meter.
-
LLMs Enhanced by Emotional Stimuli Understanding
By
–
Large Language Models Understand and Can be Enhanced by Emotional Stimuli https://
bit.ly/3N8wt5f
#AI #MachineLearning #DeepLearning #LLMs #DataScience -
SmolLM2 1.7B Beats Larger Models with Apache 2.0 License
By
–
Fuck it – it’s raining smol LMs – SmolLM2 1.7B – beats Qwen 2.5 1.5B & Llama 3.21B, Apache 2.0 licensed, trained on 11 Trillion tokens > 135M, 360M, 1.7B parameter model
> Trained on FineWeb-Edu, DCLM, The Stack, along w/ new mathematics and coding datasets
> Specialises in