Connect with us

Hi, what are you looking for?

Tech News

Microsoft LASERs away LLM inaccuracies

Microsoft logo
Illustration: The Verge

During the January Microsoft Research Forum, Dipendra Misra, a senior researcher at Microsoft Research Lab NYC and AI Frontiers, explained how Layer-Selective Rank Reduction (or LASER) can make large language models more accurate.

With LASER, researchers can “intervene” and replace one weight matrix with an approximate smaller one. Weights are the contextual connections models make. The heavier the weight, the more the model relies on it. So, does replacing something with more correlations and contexts make the model less accurate? Based on their test results, the answer, surprisingly, is no.

“We are doing intervention using LASER on the LLM, so one would expect that the model loss should go up as we are doing more approximation,…

Continue reading…

You May Also Like

Editor's Pick

Jennifer Huddleston Keeping kids and teens safe online is a priority for many parents today. Unsurprisingly, Congress and many state legislatures have also been taking...

Editor's Pick

Nearly two weeks ago, I was discussing in our FREE EB Digest newsletter why I felt Alphabet (GOOGL) was poised for a decline after...

Editor's Pick

In this edition of StockCharts TV‘s The Final Bar, Dave speaks to weakening short-term breadth conditions as the McClellan Oscillator rotates to a bearish level this...

Editor's Pick

After topping Wall Street’s earnings and revenue expectations on Tuesday, Eli Lilly & Co (LLY) shot up 5% from the prior day’s close, only...