Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To! Qi et al.: https://
arxiv.org/abs/2310.03693 #ArtificialIntelligence #DeepLearning #MachineLearning
LLMS
-
Fine-tuning Aligned Language Models Compromises Safety Study
By
–
-
10 ChatGPT Prompts to Make Money in 2024
By
–
ChatGPT-4 made countless millionaires. But so many are STUCK making no money. Here's 10 ChatGPT Prompts to make money in 2024: Bookmark for later. 1. Etsy Prompt: “Find 3 to 5 niches with low competition but high demand and create Spreadsheet Templates that will make
-
MathVista: Evaluating Math Reasoning in Visual Contexts
By
–
MathVista: Evaluating Math Reasoning in Visual Contexts with GPT-4V, Bard, and Other Large Multimodal Models Lu et al.: https://
arxiv.org/abs/2310.02255 #Artificialintelligence #DeepLearning #MachineLearning -
Debugging RAG Apps: From Prototype to Production
By
–
It's easy to get a prototype of a RAG app up and running… but going to production is a lot harder Is the retrieval step messing up? Does the prompt need to be improved? Debugging RAG apps can be complicated. @curious_kaylynn just published a great video on how to do it more
-
LLM Inference Breakthroughs Driving AI Performance at GTC24
By
–
Learn the game-changing power of inference for #LLMs at #GTC24.Hear from @ControlExpert, @databricks
, @togethercompute
, & more on how inference for LLMs is driving breakthrough performance for AI-enabled applications and services. Register today > https://
nvda.ws/3vFDm8L -
Personalization and Memory Features in Copilot and ChatGPT
By
–
I would also add personalisation and ability to remember past conversations. It was part of the Copilot Pro announcement and a similar feature is in development for ChatGPT.
-
Defining Jailbreaks vs. Prompt Injections in AI Interactions
By
–
Maybe one way to define the distinction: Jailbreaks violate the fixed policies of the vendor, whereas prompt injections violate the dynamic policies of a "superuser" who controls the outer prompt.
-
Improving AI Model Instruction Following Capabilities
By
–
Please send over examples of the model doing this, we are actively working to improve the models capabilities and instruction following.
-
Competition Between AI Chatbot Platforms and Upcoming Features
By
–
As the timeline is still unclear, it would still boil down to the competition with Bard and the release of their upcoming Bots feature
-
Stable Code 3B Outperforms Larger Models on Edge Devices
By
–
Stable Code 3B outperforms code models of a similar size and matches CodeLLaMA 7b performance despite being 40% of the size. This makes it ideal for running on edge devices to ensure privacy and drive better dev experiences