Excited to help start the *That is good data* repo led by @_dieuwke_ @adinamwilliams and @xenia_ohmer
. A place to record/share data points and general issues with our evaluation benchmark. Feel free to share your eval issues there
RESEARCH
-
New Data Repository Launched for Evaluation Benchmark Issues
By
–
-
Machine-Generated Affective Empathy: Potential Aversion Concerns
By
–
Is there any reason to believe that affective empathy would result in similar aversion when machine generated?
-
FLASK: Fine-Grained Language Model Evaluation Based on Alignment Skills
By
–
FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Sets Ye et al.: https://
arxiv.org/abs/2307.10928 #ArtificialIntelligence #DeepLearning #MachineLearning -
LLMs and Query Writing: Fundamental Mismatch Between Probabilistic and Deductive Approaches
By
–
..,LLMs and query writing are fundamentally mismatched. The former is probabilistic, creative, and inductive—it’s best used to generate ideas from short prompts. But analysts have to do the opposite when answering questions. They need to be precise, rigorous, and deductive..”
-
Cloud Architecture Optimization for AI and Data Science Efficiency
By
–
The challenge lies in fitting those pieces together for maximum reuse and efficiency. Here's to solving the cloud jigsaw! #ai #datascience #cloud #saschat #Analytics #ExploreSAS
-
Google AI and Partnership AI Collaborate on Safety Best Practices
By
–
Glad to see @GoogleAI continuing to do great work with @PartnershipAI on safety best practices and synthetic media. These kinds of industry collaborations are exactly why we started PAI waaay back in 2016. Great work!
-
Unified Data Management Strategy Prevents Multiple Truths Problem
By
–
A6: "Multiple versions of truth" issues can sprout like weeds in a garden of data. Without a unified data management strategy, the garden becomes a jungle! #ai #datascience #cloud #saschat
-
Essential Python Packages for Data Science Excellence
By
–
Want to excel in #DataScience using Python? Here are some key packages you should know: NumPy, Pandas, Matplotlib, Scikit-learn, #TensorFlow, #Keras. By @ingliguori #DataScientists #innovation #Python #MachineLearning #DataAnalytics
-
Cloud Solutions and IP Drive Agility and Organizational Resilience
By
–
Indeed, cloud solutions coupled with IP create a potent mix for tackling specific challenges. They bring agility and resilience to orgs, enabling them to rapidly test, scale, learn, and pivot when necessary. #ai #datascience #cloud #saschat #Analytics #ExploreSAS
-
Cloud AI and Data Science Insights Generation Platform
By
–
A2. Spot on! No matter the cloud destination, you're plugged in and ready to generate insights! #ai #datascience #cloud #saschat #Analytics #ExploreSAS