Damian Yerrick on Mythryl: “"LSTM" stands for "long short-term memory", the neural network…”

"LSTM" stands for "long short-term memory", the neural network architecture that predates the more parallel "transformer" architecture of current language models. Yet out of frustration with LLM output that has filled web search results, I first read it as "looks slop to me."

Sung Kim@sungkim.bsky.social

Understanding LSTM Networks Long Short Term Memory networks – called “LSTMs” – are a special kind of RNN, capable of learning long-term dependencies. They were introduced by Hochreiter & Schmidhuber, and were refined and popularized by many people in following work. colah.github.io/posts/2015-0...

26/9/26, 9:38 pm0 reposts0 likes

Post