Tool Priority First up, an example of how to create some sense of priority amongst tools Basically just tell the LLM (with language) when to use which tools Thank you @Yoooongtae for creating this excellent guide! https://
langchain.readthedocs.io/en/latest/modu
les/agents/examples/custom_tools.html#defining-the-priorities-among-tools
…
CODE
-
Tool Priority: Guide to LLM Tool Selection
By
–
-
LangChain Direct Return Flag for Tool Responses
By
–
Direct Return Sometimes, if a specific tool is chosen, you may just want to return the response of that directly to the user There is now a flag to easily allow for exactly that Thank you @andrewgleave @brunotorious for the suggestion! https://
langchain.readthedocs.io/en/latest/modu
les/agents/examples/custom_tools.html#using-tools-to-return-directly
… -
OpenAI API Update: Logprobs and Agent Tool Improvements
By
–
v0.0.58 Add generation reason and logprobs to @OpenAI response – @diwanksingh Add @OpenAI logit bias param – @benderville bug fixes (h/t @shivanshug11 , @brain_exe_ai ) And a bunch of Agent & Tool improvements! So many we can't list them all in this tweet ->
-
YOLOv5 Object Detection Models: Performance and Optimized Inference Speed
By
–
YOLOv5 object detection models are well known for their excellent performance and optimized inference speed.🚀
— Satya Mallick (@LearnOpenCV) 6 janvier 2023
▶️ In case you missed it, check out the full read on YOLOv5 Instance Segmentation – https://t.co/XPIV4ntDjZ #yolov5 #segmentation #computervision #deeplearning #ai pic.twitter.com/yQfGitP3D5YOLOv5 object detection models are well known for their excellent performance and optimized inference speed. In case you missed it, check out the full read on YOLOv5 Instance Segmentation – https://
learnopencv.com/yolov5-instanc
e-segmentation/
… #yolov5 #segmentation #computervision #deeplearning #ai -
Securing AI Applications Against Prompt Injection Attacks
By
–
It’s is a burgeoning area of interest on how yo make your application full proof to prompt leaks or any kind of prompt injection attacks. Some remedies are making sure that people can’t easily get to your source prompt (secret sauce) by tweaking with the input.
-
YouCode: AI-Powered Search Engine Generates Code On-the-Fly
By
–
2. YouCode For the first-time ever a search engine generates code on-the-fly using AI along with the search results for a given query.
-
Simplifying Embeddings Creation and Module Shipping
By
–
make it easier to make and ship embeddings as modules?
-
Databricks Transpiler Converts Splunk SPL Queries to PySpark
By
–
Say "hello" to transpiler, a Databricks Labs open-source project that automates the translation of Splunk SPL queries into scalable PySpark data frame operations. Learn more https://
dbricks.co/3V8DpRp -
94.5% Accuracy Fashion MNIST Achievement in Five Epochs
By
–
94.5% on Fashion MNIST in 5 epochs is pretty amazing!
-
Bypassing AI Detection: Natural Text and Watermarking Constraints
By
–
I'm not sure about that. E.g. tricking the ZeroGPT algo means actually writing more natural text (so it'll be better, not worse), and tricking the watermarking means not being limited by the constraints of the watermarking algo.