Prompt engineering is a dark art.
More like alchemy than science
@hardmaru
-
Prompt Engineering: More Alchemy Than Science
By
–
-
Sakana AI Labs Spotlight on Google Cloud Platform
By
–
Proud to be spotlighted by @GoogleCloud “
@SakanaAILabs
, a Tokyo-based startup that leverages NVIDIA GPUs on Google Cloud, along with services like Filestore and Persistent Disk, to power its research activities and host its foundational models.” -
OpenAI’s Ongoing Financial Needs and Future Funding Requirements
By
–
Related article: “OpenAI Needs to Raise More Money Again Soon”
-
Do AI Companies Work? Market Irrationality Requirement
By
–
Do AI companies work? “The market needs to be irrational for you to stay solvent.” https://
benn.substack.com/p/do-ai-compan
ies-work
… -
Reading RL Textbook as a Candidate Screening Method
By
–
“You have read the RL textbook and done the exercises” Great way to filter for candidates http://
incompleteideas.net/book/the-book-
2nd.html
… -
Sakana AI Wins METI/NEDO GENIAC Award for Research Breakthroughs
By
–
Sakana AI received an award from 🇯🇵 Japan Gov’t’s METI/NEDO GENIAC challenge! Over the summer, we were able to execute several experiments on our Government-sponsored GPU cluster, leading to several research breakthroughs that we will be releasing soon. So proud of the team! 🏆🎊 https://t.co/kaYjKROCNH
— hardmaru (@hardmaru) 11 septembre 2024Sakana AI received an award from Japan Gov’t’s METI/NEDO GENIAC challenge! Over the summer, we were able to execute several experiments on our Government-sponsored GPU cluster, leading to several research breakthroughs that we will be releasing soon. So proud of the team!
-
LLMs Generating Novel Research Ideas: Study with 100+ NLP Researchers
By
–
Exciting related work: “Can LLMs Generate Novel Research Ideas? A Large-Scale Human Study with 100+ NLP Researchers” New paper by @ChengleiSi @tatsu_hashimoto @Diyi_Yang @stanfordnlp
, with help from 100+ NLP researchers Paper: https://
arxiv.org/abs/2409.04109 -
Community Addresses LLM Overfitting to Evaluation Benchmarks
By
–
It’s good to see the community raise the issue of extreme overfitting to evaluation benchmarks for LLMs. This is the real “Reflection” that the open-source community needs to have.
-
Japanese Media Inconsistency in Katakana Writing for Company Names
By
–
Japanese Katakana Conundrums Nikkei: Sakana AI is 「サカナAI」and NVIDIA is ‘NVIDIA’ NHK: Sakana AI is ‘Sakana AI’ and NVIDIA is 「エヌビディア」