2/ For one evaluation, MMLU (
https://
arxiv.org/abs/2009.03300), the community was surprised that the leaderboard numbers for the top model, LLaMA 65B, were significantly lower than the numbers in the published LLaMa paper: a 30% difference! We dived in a rabbit hole to understand
RESEARCH
-
LLaMA 65B MMLU Benchmark Discrepancy Analysis
By
–
-
Open LLM Leaderboard evaluation accuracy analysis findings
By
–
A on "what was going on with the Open LLM Leaderboard?" its numbers didn't match the ones reported in LLaMA paper so we dived in it and wrote a blog post of learnings! Here's the thread version for those of you who didn't want to read a blog post
-
ACL Initiative Broadens NLP Participation for Underrepresented Countries
By
–
Deadline approaching! This excellent initiative is meant to broaden participation in ACL in countries where our registration fees are way out of reach. Please spread the word & apply by June 28. #NLProc
-
LLMs Demonstrate Powerful Investigation and Analysis Capabilities
By
–
There is something so powerful about large language models’ abilities here. I could have commissioned an entire team of investigators to attempt to piece something like this together about me, given them months of time and thousands of dollars, and not gotten anything close to
-
Gemini: DeepMind Combines Language Models with Reinforcement Learning
By
–
The next @deepmind algorithm, Gemini, may combine a powerful language model with cutting-edge reinforcement learning, perhaps helping with planning and memory. That's one thing I learned from a very interesting chat with @demishassabis recently.
-
Generative AI Technical Evolution: Key Innovations 2012-2022
By
–
Fair enough but Generative AI only really took off since 2020 with much of the technical innovation occurring from 2022 onwards (Chain of Thought 22, LoRA 21 but really adopted 22 and RLHF really took off in a big way 2022). The approaches in 2012 would not get us to LLMs. RNNs
-
Accelerating PyTorch Training with Mixed-Precision and Fully Sharded Data Parallelism
By
–
"Accelerating PyTorch Model Training Using Mixed-Precision and Fully Sharded Data Parallelism" https://
magazine.sebastianraschka.com/p/accelerating
-pytorch-model-training
… A little write-up of a talk I gave last week! -
DragGAN Source Code and Demo Now Available Online
By
–
DragGAN source code and demo available now! 💥
— abhishek (@abhi1thakur) 26 juin 2023
Try it here: https://t.co/WJgFgc617J
Code: https://t.co/BFWpUS9HaB pic.twitter.com/Xiw6R80dWpDragGAN source code and demo available now! Try it here: https://
huggingface.co/spaces/radames
/DragGan
…
Code: https://
github.com/XingangPan/Dra
gGAN
… -
Top AI Models and Datasets Weekly Trends Highlighted
By
–
Trending models, datasets, spaces and PAPERS (
https://
huggingface.co/Papers!!) of the week on http://
hf.co. Congrats to @TIIuae (falcon) @MosaicML (mpt) @StabilityAI (SDXl) @MetaAI (MusicGen) @MSFTResearch (Textbooks AAYN) @UMBaltimore (through your eyes) and many others! -
Generative AI Terms and Their Definitions Guide
By
–
Generative AI Terms and Their Definitions
#AI #AIio #BigData #ML #NLU #Futureofwork @Kevin_Jackson @sallyeaves @schmarzo @kdnuggets @akwyz @BrettKing @TechNative @MiaD http://
ow.ly/wuxl30svStY