AI Researchers have been working on Symbolic Logic fusions with LLM Transformers One example from @GoogleDeepMind
: https://
technologyreview.com/2024/01/17/108
6722/google-deepmind-alphageometry/
…
MACHINE LEARNING
-
Symbolic Logic Fusions with LLM Transformers by Google DeepMind
By
–
-
Model Fine-Tuning for Human Preferences
By
–
Your model, after some fine-tuning for human preferences.pic.twitter.com/lzq5WAqP1b
— hardmaru (@hardmaru) 15 avril 2024Your model, after some fine-tuning for human preferences.
-
AI Inbreeding Risk Threatens Quality and Diversity in Generative Models
By
–
#GenerativeAI risks "inbreeding," where AI-generated content used for training #future #AIs leads to a decline in quality and #diversity, akin to the effects of genetic inbreeding. This could degrade AI's ability to simulate human language and #creativity
-
Keras MLX Backend Development: Community Contributors Wanted
By
–
Keras has a MLX backend in progress. I'd like to work with a few folks who are part of both the MLX community & the Keras community to get it over the finish line. If you're interested — please send me an email!
-
AI-Generated Data in Pre-training: Quality Impact Analysis
By
–
I think synthetic data for fine-tuning is a seoarate issue from whether accidental AI-generated text in pre-training data has an impact on model quality
-
AI Generated Training Data Quality vs Human Data Impact
By
–
No idea! A lot of people seem to believe that accidental AI generated training data is uniquely harmful to models compared to low quality human data – I'm trying to figure out what the latest thinking on that is
-
Causal Inference Expert Mixtape Resource for AI Research
By
–
One of the authors is @causalinf
, who was is an expert in thinking about causality. https://
mixtape.scunning.com -
Human Baselines Outperform GPT-4 on Non-Square Grid Tasks
By
–
We also compare LLMs with human baselines. Although human responses are not perfect, they outperform GPT-4 (0314) by a substantial margin. Furthermore, like GPT-4 (0314), non-expert humans struggle with non-square grid shapes.
6/n -
LLM Performance Variability Across Different Spatial Structures
By
–
We examine various spatial structures (squares, triangles, hexagons, rings, and trees), which reveal substantial variability in LLM performance across different structures. (GPT-4 here is 0314)
5/n