"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."
#lstm […]
#lstm […]
Original post on peoplemaking.games
peoplemaking.games
September 26, 2026 at 9:39 PM
"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."
#lstm […]
#lstm […]
LSTM means "looks shitty to me"
September 23, 2026 at 4:22 PM
LSTM means "looks shitty to me"
LSTM even. Buttons; the eternal struggle of my fat, dyspraxic fingers.
February 25, 2026 at 12:11 PM
LSTM even. Buttons; the eternal struggle of my fat, dyspraxic fingers.
LSTM develops the world's first rapid test that detects Crimean-Congo hemorrhagic fever #CCHF
www.lstmed.ac.uk/news-events/...
www.lstmed.ac.uk/news-events/...
LSTM develops the world's first rapid test that detects Crimean-Congo haemorrhagic fever
The new lateral flow test enables healthcare workers to identify CCHF cases quickly, potentially saving lives by speeding up treatment and reducing the spread of outbreaks, which can have a mortality
www.lstmed.ac.uk
January 18, 2025 at 12:22 PM
LSTM develops the world's first rapid test that detects Crimean-Congo hemorrhagic fever #CCHF
www.lstmed.ac.uk/news-events/...
www.lstmed.ac.uk/news-events/...
Erste Version einer stochastischen DIY Strompreis Prognose fertig. Das ganze wird in mein Energie Optimierungssystem eingebaut & basiert auf LSTM + MC + DA + Attention. Wie das funktioniert? Bald als kostenloses Webinar auf meinem YouTube Kanal 😊 wird mehrteilig:
November 16, 2024 at 2:18 PM
Erste Version einer stochastischen DIY Strompreis Prognose fertig. Das ganze wird in mein Energie Optimierungssystem eingebaut & basiert auf LSTM + MC + DA + Attention. Wie das funktioniert? Bald als kostenloses Webinar auf meinem YouTube Kanal 😊 wird mehrteilig:
i trained an LSTM to generate nearly-English gibberish for part of an art installation in 2016. i finetuned a GPT-2 model on a mixture of tweets to curate its output for weird broken autopoetry and the rare moments when it output something interesting. i was here before all these bastards
April 3, 2026 at 12:20 AM
i trained an LSTM to generate nearly-English gibberish for part of an art installation in 2016. i finetuned a GPT-2 model on a mixture of tweets to curate its output for weird broken autopoetry and the rare moments when it output something interesting. i was here before all these bastards
Scientific Reports by Springer Nature seems to have gone full rogue
pubpeer.com/publications...
The editor who accepted this and allowed for a correction should be fired. It's clear no credible peer reviewer has seen it
And can anyone make sense of this Figure?
www.nature.com/articles/s41...
pubpeer.com/publications...
The editor who accepted this and allowed for a correction should be fired. It's clear no credible peer reviewer has seen it
And can anyone make sense of this Figure?
www.nature.com/articles/s41...
PubPeer - Author Correction: Hybrid CNN-LSTM model with efficient hype...
There are comments on PubPeer for publication: Author Correction: Hybrid CNN-LSTM model with efficient hyperparameter tuning for prediction of Parkinson’s disease (2024)
pubpeer.com
October 9, 2024 at 10:47 AM
Scientific Reports by Springer Nature seems to have gone full rogue
pubpeer.com/publications...
The editor who accepted this and allowed for a correction should be fired. It's clear no credible peer reviewer has seen it
And can anyone make sense of this Figure?
www.nature.com/articles/s41...
pubpeer.com/publications...
The editor who accepted this and allowed for a correction should be fired. It's clear no credible peer reviewer has seen it
And can anyone make sense of this Figure?
www.nature.com/articles/s41...
March 15, 2025 at 3:27 AM
Data/Feature Engineering
Modellauswahl (Gaussprozess vs. LSTM, vielleicht auch sarimax)
Feintuning des Modells
Verwendete Begriffe:
MC = Monte Carlo dropout
DA = Data Augmentation
Modellauswahl (Gaussprozess vs. LSTM, vielleicht auch sarimax)
Feintuning des Modells
Verwendete Begriffe:
MC = Monte Carlo dropout
DA = Data Augmentation
November 16, 2024 at 2:18 PM
Data/Feature Engineering
Modellauswahl (Gaussprozess vs. LSTM, vielleicht auch sarimax)
Feintuning des Modells
Verwendete Begriffe:
MC = Monte Carlo dropout
DA = Data Augmentation
Modellauswahl (Gaussprozess vs. LSTM, vielleicht auch sarimax)
Feintuning des Modells
Verwendete Begriffe:
MC = Monte Carlo dropout
DA = Data Augmentation
On my way to give a talk at CNN’s NYC headquarters. Taking the opportunity to wear a niche AI humor t-shirt that probably only made sense in 2016
April 15, 2026 at 5:00 PM
On my way to give a talk at CNN’s NYC headquarters. Taking the opportunity to wear a niche AI humor t-shirt that probably only made sense in 2016
SOTA-level in less than 2.6k parameters by combining attention, LSTM, evolution and classic computer vision. A follow-up of Yujin Tang and David Ha's work: attentionagent.github.io
November 11, 2024 at 3:38 PM
SOTA-level in less than 2.6k parameters by combining attention, LSTM, evolution and classic computer vision. A follow-up of Yujin Tang and David Ha's work: attentionagent.github.io
halsey canta lstm amanhã pf
September 10, 2024 at 8:31 PM
halsey canta lstm amanhã pf
Lots of MosSI Lab representation at the LSTM 125th Scientific Symposium this week!
@mosquitomicrobe.bsky.social spoke about our Anopheles Wolbachia work.
@mosquitomicrobe.bsky.social spoke about our Anopheles Wolbachia work.
November 27, 2024 at 3:54 PM
Lots of MosSI Lab representation at the LSTM 125th Scientific Symposium this week!
@mosquitomicrobe.bsky.social spoke about our Anopheles Wolbachia work.
@mosquitomicrobe.bsky.social spoke about our Anopheles Wolbachia work.
My news is that I will be joining the Liverpool School of Tropical Medicine! I'm very excited to join an organization that shares my passion for improving global health, at the same time I'm sad to leave all of the wonderful colleagues here at SCRI who I've enjoyed working with over the past ~6y 🦠🧪
LSTM welcomes three leading global health therapeutics experts with support from UKRI
Dr Jeremy Burrows, Professor Tanya Parish, and Dr Ami Patel will bring unique expertise in medicinal chemistry, drug discovery, vaccinology and antibody-based therapeutics into LSTM, enhancing its
www.lstmed.ac.uk
September 17, 2025 at 11:07 PM
My news is that I will be joining the Liverpool School of Tropical Medicine! I'm very excited to join an organization that shares my passion for improving global health, at the same time I'm sad to leave all of the wonderful colleagues here at SCRI who I've enjoyed working with over the past ~6y 🦠🧪
September 17, 2025 at 12:19 PM
検証結果：GPT(LoRA) >> Markov連鎖 >>>> LSTM(Pretrained) >>>>>>> LSTM(from Scratch)
ロボてりーの下位互換的なやつ作って探究の成果物として提出するか
August 15, 2026 at 11:05 AM
検証結果：GPT(LoRA) >> Markov連鎖 >>>> LSTM(Pretrained) >>>>>>> LSTM(from Scratch)
It's days like this that make me point out to my data science students that you can't just train an LSTM on Tesla's stock price and expect to get an accurate prediction of future returns.
June 5, 2025 at 7:45 PM
It's days like this that make me point out to my data science students that you can't just train an LSTM on Tesla's stock price and expect to get an accurate prediction of future returns.
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...
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...
January 17, 2025 at 12:54 PM
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...
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...
I'm not sure about this...
- Logit for churn prediction? Maybe as a baseline. I used boosted trees.
- Random forest for fraud detection? I used boosted trees.
- LSTM for sentiment analysis? I used BERT or spacy + VADER.
- Logit for churn prediction? Maybe as a baseline. I used boosted trees.
- Random forest for fraud detection? I used boosted trees.
- LSTM for sentiment analysis? I used BERT or spacy + VADER.
October 4, 2025 at 2:32 AM
I'm not sure about this...
- Logit for churn prediction? Maybe as a baseline. I used boosted trees.
- Random forest for fraud detection? I used boosted trees.
- LSTM for sentiment analysis? I used BERT or spacy + VADER.
- Logit for churn prediction? Maybe as a baseline. I used boosted trees.
- Random forest for fraud detection? I used boosted trees.
- LSTM for sentiment analysis? I used BERT or spacy + VADER.
NLP Pipeline: text → normalize → tokenize → vocabulary → IDs → embeddings → padded sequences → LSTM → hidden states → logits → probabilities → decoding → text
September 12, 2026 at 4:56 AM
NLP Pipeline: text → normalize → tokenize → vocabulary → IDs → embeddings → padded sequences → LSTM → hidden states → logits → probabilities → decoding → text
Post-Doctoral Research Associate - LSTM Decolonial History Project
Liverpool School of Tropical Medicine #skystorians 🗃️www.jobs.ac.uk/job/DOK716/p...
Liverpool School of Tropical Medicine #skystorians 🗃️www.jobs.ac.uk/job/DOK716/p...
Post-Doctoral Research Associate - LSTM Decolonial History Project at Liverpool School of Tropical Medicine
Apply for the Post-Doctoral Research Associate - LSTM Decolonial History Project role on jobs.ac.uk, the top job board for academic positions in higher education. View details and apply now.
www.jobs.ac.uk
September 3, 2025 at 5:59 AM
Post-Doctoral Research Associate - LSTM Decolonial History Project
Liverpool School of Tropical Medicine #skystorians 🗃️www.jobs.ac.uk/job/DOK716/p...
Liverpool School of Tropical Medicine #skystorians 🗃️www.jobs.ac.uk/job/DOK716/p...
SaLSa is a #MATLAB package for classifying animal behavior. It combines semi-automatic labeling with LSTM neural networks to analyze data from pose estimation to identify behavioral syllables. Read about SaLSa in this week's post on OpenBehavior:
edspace.american.edu/openbehavior...
edspace.american.edu/openbehavior...
December 13, 2024 at 6:36 PM
SaLSa is a #MATLAB package for classifying animal behavior. It combines semi-automatic labeling with LSTM neural networks to analyze data from pose estimation to identify behavioral syllables. Read about SaLSa in this week's post on OpenBehavior:
edspace.american.edu/openbehavior...
edspace.american.edu/openbehavior...
Day 33: Clearly not working. Still, I made progress with understanding LSTM. #100DaysOfCode
February 27, 2025 at 4:49 AM
Day 33: Clearly not working. Still, I made progress with understanding LSTM. #100DaysOfCode
暴れん坊MVPのLSTM(from Scratch)くんはこちら
August 15, 2026 at 1:25 PM
暴れん坊MVPのLSTM(from Scratch)くんはこちら