Good morning fellow AI enthusiast! This is a special iteration for a new paper we just published with @learnprompting
, @towards_AI
, and fantastic people along with the winners of the HackAPrompt competition… and we made a video about it! Learn more https://
open.substack.com/pub/louisbouch
ard/p/ignore-this-title-and-hackaprompt?r=25qlky&utm_campaign=post&utm_medium=web
…
LLMS
-
New AI Paper Published with HackAPrompt Competition Winners
By
–
-
OpenAI models to learn from user chat interactions
By
–
Source and more info https://
reddit.com/r/OpenAI/comme
nts/17wlhbv/your_gpt_will_soon_learn_from_your_chats/
… -
ChatGPT introduces memory feature for personalized user interactions
By
–
ChatGPT set to remember past interactions for more personalized conversations h/t @yupiop12
-
Google Bard Adds Math Assistance and Data Visualization Features
By
–
For math equations, you can simply upload a picture to get a detailed explanation Link https://
testingcatalog.com/google-bard-ro
lls-out-math-equation-assistance-and-data-visualization-features/
… -
Google Bard Adds Math Assistance and Data Visualization Features
By
–
Google Bard rolls out math equation assistance and data visualization features
-
Hugging Face Launches Deep Reinforcement Learning Course
By
–
Welcome to the Deep Reinforcement Learning Course – Hugging Face Deep RL Course https://
bit.ly/3PMaByU #AI #MachineLearning #DeepLearning #LLMs #DataScience -
LeanDojo Open-Source LLM for Certified Mathematical Theorem Proving
By
–
We have precisely done that. Our Leandojo is the first open-source LLM for theorem proving: so 100% accurate and certified proofs of math theorems. We will soon be launching the co-pilot as well at Neurips. Stay tuned. https://
leandojo.org -
Playful response about working on LLMs: “poast training”
By
–
them: oh, you work on LLMs? pre-training or post? me: poast training
-
Chain-of-Note Prompting Technique for RAG Applications
By
–
Chain-of-Note Template Chain-of-Note is a new prompting technique by @wyu_nd et al for RAG applications that helps improve performance when the retrieved information might be noisy. We implemented a Chain-of-Note app as a LangChain template. Given a question, query Wikipedia
-
GPT-4 Could Help Review Denied Scientific Studies
By
–
Many scientific studies are denied without peer review. HAI faculty affiliate @james_y_zou asks, what if GPT-4 could help? https://
stanford.io/3QK6gey