Adversarial examples have been hard to solve in vision mainly because it was imperceptible to humans and added adv noise is hard to remove. Looking at the demo, it seems relatively easy to build a filter to remove adversarial suffix before feeding the query into LLM…
PROMPT ENGINEERING
-
SDXL Model Guide: Image Generation, Inpainting, and Refinement Techniques
By
–
This guide covers the basics, but also shows how you can use the SDXL model to: • make an image from another image (img2img)
• inpaint using a mask
• use a refiner to add fine-detail to your images -
GenAI Models Creating Opinionated Markup Outputs
By
–
would be fun if a genai model could create opinionated markups like this
-
Llama 2 and RAG webinar with industry experts
By
–
going live for a webinar with some awesome folks talking about llama-2 and using it for RAG Come join! https://
crowdcast.io/c/vk4palyeh9t0 -
Query Generation for LangChain and LangSmith Tools
By
–
is basically generating different queries eg if you have a higher level question "what is the difference between langchain and langsmith" the queries generated may be: "what is langchain"
"what is langsmith"
"difference between langchain and langsmith" -
Llama 2 First Class Support with Optimized Prompts
By
–
Llama 2 support One really cool thing about this project – we made support for Llama 2 a first class citizen This includes prompts specifically tuned for the Llama 2 model We'll be talking about this heavily in our webinar later today! https://
crowdcast.io/c/vk4palyeh9t0 -
GPT-4 Chatbot Building Course Now Open for Subscribers
By
–
Enrollment is open for my How to Build a GPT-4 Chatbot course! It's available right now for @Every paid subscribers for a significant discount. Last time we signed up all 80 seats in the early-bird so if you want to learn to build in AI, check it out now!
-
Game-Changing ChatGPT Custom Instructions: User Experiences
By
–
Who has ChatGPT custom instructions that have been game changing for them?
-
Supercharge Llama 2 with RAG Webinar Tomorrow
By
–
Supercharge Llama 2 with RAG Beyond excited to be doing a webinar exploring how to supercharge Llama 2 with two of the best in the biz: @jamescalam from @pinecone @RLanceMartin from @langchain This last-minute webinar will be TMRW @ 3PM PST
-
ChatGPT’s reliance on two years of human labor for instruction-following
By
–
2/ Without these 2+ years of labor, ChatGPT wouldn't be able to chat. It wouldn't know how to answer questions. Ask “explain the moon landing to a 6-year-old,” and it will uselessly reply with word associations. OpenAI talks about this here. https://
openai.com/research/instr
uction-following
…