Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
veganism.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
Veganism Social is a welcoming space on the internet for vegans to connect and engage with the broader decentralized social media community.

Administered by:

Server stats:

297
active users

veganism.social: AboutStatusPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.0-alpha.5+vegan

#nlp

10 posts9 participants2 posts today
Public
Loredana Latis-Expert Educație

„Îmi era greu să spun NU, mă simțeam prinsă între așteptările altora și propriile mele nevoi… până când am descoperit puterea de a-mi clarifica valorile și a-mi stabili granițele!”

Mariana Diaconu a reușit, prin NLP, să-și redefinească prioritățile, să-și crească încrederea în sine și să preia controlul asupra vieții ei.

Dacă și tu simți că ai nevoie de mai multă claritate și echilibru, vezi cum a reușit Mariana! Link în comentarii.

#Claritate#StimaDeSine#NLP
Public
Holle Meding

Tracing Chemical Knowledge Over Centuries with #LLMs 🧪

Diego Alves, Sergei Bagdasarov & Badr M. Abdullah prompted models to generate structured metadata for 47k+ texts from the #RoyalSociety Corpus (1665–1996), enabling large-scale comparison of #Chemistry and #Biology over time.

They tracked how chemical substances migrated between disciplines revealing a "chemicalization" of biology in the 19th century and a long-term trend toward standardization. #OpenScience #DiachronicAnalysis #NLP

Distribution of Scientific Disciplines in Royal Society Articles (1665–1996) - Diego Alves, Sergei Bagdasarov & Badr M. Abdullah
ALT
Public
Loredana Latis-Expert Educație

„Mă simțeam prinsă într-o ceață continuă, fără direcție… până când am descoperit cum să-mi clarific mintea și să-mi găsesc drumul!”

Mihaela Vișan a trecut printr-o transformare profundă cu ajutorul NLP, lăsând în urmă confuzia și dobândind claritatea de care avea nevoie pentru a merge înainte cu încredere.

Dacă și tu simți că ai nevoie de mai multă claritate în viața ta, vezi cum a reușit Mihaela! Link în comentarii.

#Claritate#NLP#Transformare
Public
Holle Meding

🕰️ Small but time-aware!

At the " #LLM for for the History, Philosophy, and Sociology of Science workshop" @tuberlin Jochen Büttner presented a compact small language model (39M parameters) that integrates a temporal dimension for historical text analysis.

A proof of concept using historical UK weather reports shows the model successfully reflects both seasonal patterns and linguistic shifts

Wanna try it?
👉 github.com/j-buettner/time_transformer

Jochen Büttner's presentation on a time-aware language model
ALT
Jochen Büttner's presentation on a time-aware language model
ALT
#DigitalHumanities#Transformer#NLP
Public
Hacker News

Multi-Token Attention

arxiv.org/abs/2504.00927

#HackerNews #Multi-Token #Attention #research #AI #advancements #attentionmechanism #deeplearning #NLP

arXiv logo
arXiv.orgMulti-Token AttentionSoft attention is a critical mechanism powering LLMs to locate relevant parts within a given context. However, individual attention weights are determined by the similarity of only a single query and key token vector. This "single token attention" bottlenecks the amount of information used in distinguishing a relevant part from the rest of the context. To address this issue, we propose a new attention method, Multi-Token Attention (MTA), which allows LLMs to condition their attention weights on multiple query and key vectors simultaneously. This is achieved by applying convolution operations over queries, keys and heads, allowing nearby queries and keys to affect each other's attention weights for more precise attention. As a result, our method can locate relevant context using richer, more nuanced information that can exceed a single vector's capacity. Through extensive evaluations, we demonstrate that MTA achieves enhanced performance on a range of popular benchmarks. Notably, it outperforms Transformer baseline models on standard language modeling tasks, and on tasks that require searching for information within long contexts, where our method's ability to leverage richer information proves particularly beneficial.
Public
Holle Meding

📰 Classifying Genre in Historical Medical Periodicals

Next in line: Vera Danilova presents her work on genre classification in digitized periodicals from European patient organizations (1951–1990) using #LLMs as part of the #ActDisease project.

🔹 XLM-RoBERTa (UDM) led Q&A tasks with 32% more correct answers than mBERT/hmBERT.
🔹 hmBERT (UDM) topped Administrative classification (+16%)
🔹 CORE-based models excelled in legal genre prediction.

#DigitalHumanities @tuberlin #classification #NLP

Public
Loredana Latis-Expert Educație

„Îmi era greu să spun NU, mă simțeam prinsă între așteptările altora și propriile mele nevoi… până când am descoperit puterea de a-mi clarifica valorile și a-mi stabili granițele!”

Mariana Diaconu a reușit, prin NLP, să-și redefinească prioritățile, să-și crească încrederea în sine și să preia controlul asupra vieții ei.

Dacă și tu simți că ai nevoie de mai multă claritate și echilibru, vezi cum a reușit Mariana! Link în comentarii.

#Claritate#StimaDeSine#NLP
Public
Charlie Welch

On Tuesday, I gave a lecture on harms of AI and NLP. Turns out, the ChatGPT update the same day started a viral wave of Ghibli style image gen.

Worth seeing Miyazaki's response if you haven't. It is absolutely brutal: youtu.be/ngZ0K3lWKRc

We discussed: Can generative models be used to make art? What is the role of intent? Why are we using gen AI models and should we?

techcrunch.com/2025/03/26/open
404media.co/email/5d418149-ed1

Dami Lee: youtu.be/FRnKY-5TqOM

Image of slide from my NLP class stating that "Art has intent and reflects human experience" and "AI (think of the octopus experiment) is not grounded in human experience" It shows thumbnails from two youtube videos linked in the post, one from Dami Lee and the other, a short clip of Hayao Miyazaki responding to generative AI stating "I strongly feel that this is an insult to life itself".
ALT
#nlp#genai#chatgpt
Public
洪 民憙 (Hong Minhee)

Browser-Native Translation and Language Detection APIs Coming Soon

hackers.pub/@hongminhee/2025/t

Hackers' Pub · Browser-Native Translation and Language Detection APIs Coming SoonJust reviewed the W3C draft for the Translator and Language Detector APIs. This is genuinely exciting development for web developers. The proposal would add native browser support for:Text translation between languagesLanguage detection of arbitrary textBoth with streaming capabilities<No more relying on third-party translation services or embedding external APIs for basic language operations. All processing happens locally in the browser. The API design is clean and straightforward:// Translation exampleconst translator = await Translator.create({ sourceLanguage: "en", targetLanguage: "fr"});const translatedText = await translator.translate("Hello world");// Language detection exampleconst detector = await LanguageDetector.create();const results = await detector.detect("Hello world");// Returns array of detected languages with confidence scores<This will be a game-changer for multilingual sites and applications. The browser handles downloading appropriate language models and manages usage quotas. The spec is still in draft form but shows promising progress toward standardizing these capabilities across browsers. Looking forward to seeing this implemented.
#javascript#translation#language
VegansExploreLive feeds
About