Ah thanks for this! I've been looking for more direct evidence. I want to collect up a little literature of these results I tried to run an experiment on OntoNotes but it was pretty tough (rate limits etc, plus just hard to frame the right prompts for all the entity types).
AI
-
Testing ChatGPT’s Voice Recognition Accuracy with Simple Prompts
By
–
I’m most curious about the voice recognition accuracy (we already know many of chatgpt’s strengths and weaknesses), so I gave it simple prompt tests (it helps that my voice is a mess as I recover from a cold). It did a good job transcribing, and produced relevant answers.
-
V2X Technology Transforms EV Fleets Into Global Battery Network
By
–
V2X Can Transform #EV Fleets Into the World’s Biggest #Battery https://
bit.ly/3vHupbx via @BRINKNewsNow #sustainability -
Advanced AI Solutions Showcased at Embedded Vision Summit 2023
By
–
We are looking forward to the 2023 @EmbVisionSummit next week! We'll be showcasing our powerful and advanced solutions for #AI at the edge. Come see us at booth #604. https://
embeddedvisionsummit.com #EVS23 #EdgeAI -
Eleven Multilingual v1: New Speech Synthesis Model Introduced
By
–
Introducing Eleven Multilingual v1: Our New Speech Synthesis Model https://
bit.ly/3Nou3Rr #AI #MachineLearning #LLMs #deeplearning -
Simulants: AI Method Synthesizes Privacy-Preserving Clinical Trial Data
By
–
"Simulants" is a method for synthesizing clinical trial data that maintains subject privacy, facilitating data sharing and innovation in fields like drug safety and bias analysis. Submitted by: Afrah Shafquat @Medidata
-
JoinBoost: SQL-based tree model training library
By
–
"JoinBoost" is a Python library that enables the training of tree models on normalized databases using pure SQL queries, offering portability, scalability, and competitive performance. Submitted by: Zachary Huang @Columbia
-
Data-IQ Framework Stratifies Training Data by Predictive Confidence
By
–
"Data-IQ" is a versatile framework enabling the systematic stratification of training data into outcome-based subgroups using predictive confidence & aleatoric uncertainty, aiding in feature acquisition, dataset selection, & reliable model usage. By: Nabeel Seedat @Cambridge_Uni
-
Compressed NLP Models: Cost-Effective LLM Deployment Alternative
By
–
"Why smaller specialized NLP models might be the secret to easier LLM deployment" proposes the use of compressed NLP models as a cost-effective & efficient alternative to #LLMs, with similar performance but significantly reduced deployment challenges. By: Meryem Arik @titanML
-
Minimax Optimal Stability Estimation Under Distribution Shift
By
–
"Minimax Optimal Estimation of Stability Under Distribution Shift" this study proposes an estimator for gauging system stability under environmental changes, providing a method to predict performance deterioration and ensure robustness. Submitted by: Yuanzhe Ma @Columbia