libgen_scihub_100_bot
nexus_search_3_bot
bafybeigpp6mtsmjngaqscfkjwzivbptt4ui5yb7uih6qe43obofxujdoxa
Last one is still experiencing stability issues, first query is slow and may require page refresh several times if not opened during 2-3 minutes
Biggest library ever with embedded fast search engine, entirely in IPFS.
Z-Library + LibGen + 128M scientific papers (400 000 in IPFS) + media files (2M).
I believe you have created IPFS exactly for this. Just pin!
@IPFS
@Filecoin
@juanbenet
#IPFS
#zlib
Since Z-Library has published IPFS CIDs, we have started to ingest difference between LibGen and Z-Library.
All new Z-Library books are becoming available in all Nexus bots (and later in STC)
#zlib
#IPFS
Z-Lib books have been deduplicated and converted to IPFS with its search metadata.
Database: bafykbzacedcipiz7xsyunrsz6iwfi5l2m4se4xxgbfe2f6ncpa3kxssn7lqwg
Web-viewer:
bafykbzacedzm5r7k4ru7c6bu3qett5jsc2hk64xqrw4rck6otogskx34pw55q
Nexus allows to search semantically over largest shadow libraries such as LibGen and Sci-Hub.
Results are fantastic, especially given we have only a single mid-tier server to launch it all. Image if this tool could be legal?
#libgen
#scihub
#ai
Thanks to the independent developer, now we have browser extension Nexus Now! for Firefox, Chrome and Edge allowing you to jump from publishers' pages to STC and download the paper of interest. Links to extensions can be found at GitHub
A Telegram community SciJournal Club of 6,500 individuals faced disruption. Elsevier has just taken actions against community participants who disseminated scholarly papers among those who cannot afford subscriptions.
#KnowledgeFreedom
#Elsevier
#OpenAccess
#PublishingMafia
Copyright isn't just a form of oppression. Its barriers prevent emerging AI from legally accessing the most valuable part of our cultural heritage.
#Copyright
#AI
#RegulatoryBarriers
We've just freed up access to another German publisher. Now, 8,000 high-quality academic texts are available for those who can't afford these educational resources.
#OpenAccess
#EducationForAll
Internet granted a unique opportunity to equalize education and make it continuous. Instead, we are spending 90% of computational capacities to porn and TicToc, but still restricting access to life-saving pieces of information
Updates for STC incoming:
Registered a domain for STC: ipns://stdtc.eth that may be opened through IPFS Desktop or Brave browser.
No more these long ugly names! All updates for STC will be available there.
🏴☠️ AI Dataset STC230929
The largest text corpus to date, surpassing all including books3.
Size: 170GB, Books: 250K, Papers: 1.5M.
Recognition: GROBID + EPUB.
Use:
Launch IPFS
pip3 install stc-geck && geck - documents
Seed with IPFS: /ipns/standard-template-construct.org
#AI
🌐 Free access to high-quality scholarly publications is essential for
#AI
development and innovation. We must challenge the status quo and call for the end of copyright restrictions on these resources for AI researchers. 📚✊
#OpenAccess
#CopyrightRevolution
Recent releases of Nexus Search contain 130M of records, calculated PageRank for all papers and their references. Everything in a much compact format that CrossRef one's.
You may learn on how to deploy and use this database on Replicate page of standard-template-construct.x site
⚡️Paywalls Are Breaking Down
Yesterday, White House Office of Science and Technology Policy announced that "starting in 2026, any scientific publication that receives federal funding will need to be openly accessible on the day it's published."
IPFS directory with ALL LibGen books. Directory can be listed in the following way:
ipfs ls bafyb4iaibfst5f4icsns6i5gly6wlqnxggt3qm5ju3j6mghlrqrwd5bwja --resolve-type=false --size=false -s
P.S. Join , we are pinning all scholarly publications
We are building a programmatic API to retrieve any scholarly work on the planet (and beyond) and it works already. Have you ever dreamed to feed scholarly knowledge to GPT?
Preserving scientific knowledge is vital for humanity and essential for AGI building by democratizing learning datasets. And contemporary technologies make everyone capable to participate in knowledge replication:
We have significantly improved performance of STC Library and make it updatable in real time. Also, EPUB reader have been embedded right into interface, so no need in downloading dedicated apps for reading.
USA is commanding to "any authorized officer" of the independent country to "arrest and bring" a person of the third country before judge. Argentina, may I command you too?
#zlibrary
#ppi
#Argentina
We are integrating scholarly papers and AI. Welcome Cybrex - out first attempt to build langchain based model that bridges LLM and the library of Standard Template Construct.
Posted a quite detailed introduction and guidance about Standard Template Construct, IPFS and company on Reddit which is better suited for such reads. The post covers the most basic questions, hope it will be helpful.
We've launched a collaborative IPFS Cluster recently. Now, list of ~430 000 pairs DOI-CIDs of most popular, cited and fresh scholar papers (2021-2023 years included) is published.
STC contains these papers already, available for searching and reading.
Wanna neural networks trained on scientific corpus and capable to answer any scientific question, to point out inconsistencies in research, to suggest the direction of experiments? This is impossible not due to technical issues but due to copyright chains
We are building a collaborative IPFS Cluster (based on ) for scholar papers to ensure their availability forever, for everyone.
Contact us in Nexus Telegram chat if you are familiar with Linux, Docker, have >1TB disks and wish to devote space to cluster
@Adriano08570501
check it out, I believe in you. IPFS is similar to torrents, but more comfortable to users.
Earlier or later you will have to learn how to use it due to the censoring barriers wiping out the good old Internet
Guys have published guide for developers on how to create decentralized search.
Join the battle if you are developer and have any ideas about what should be stored and searched for eternity.
Just completed a mission statement of Nexus and and a list of our current projects. The world is going mad in wars, hunger and climate disaster, but we should try to put the base for what will come after.
We have released all collected papers in a single IPFS directory. Join our replication efforts:
ipfs pin add /ipns/hub.standard-template-construct.org --progress
More detail at (especially if you are stuck with pinning)
🌍 University of Chicago Press for Oriental Researchers
18,000 articles from 5 journals published by the University of Chicago Press are now available, thanks to at least four of our unsung but heroic librarians.
We've also created an unstoppable domain for STC to keep users updated. Install an extension to your browser:
and then visit
http://standard-template-construct.x
This address will be an instant redirector to the most actual version of STC
📻Part of PubMed papers searchable! 3M metadata added, search with standard queries
💭10k of Classical Philology pubs added.
🤖New AI dataset dropping soon with 300k+ books & millions of papers. Gear up your GPUs!
#AI
#PubMed
#ResearchUpdate
Indian Court Orders Telegram to Disclose Personal Details of Pirating Users
Indian friends should be careful. Your government is violating your privacy and now it is related to everyone here.
🚧 Current copyright laws hinder AI researchers by limiting their access to these crucial resources. This stalls humanity's progress and prevents us from realizing AI's full potential. We must break down these barriers! ⚙️🔗
⚡️ One Nation, One Subscription
Indian government finally decided to adopt "One Nation, One Subscription" program allowing national-wide access to local and international academic and scientific content from April 1, 2023
Memes are more studied than cannibalism!
Now you can plot # of publications for query against time in Nexus. Just as in Google Trends. Type /trends and follow guide.
Textual data are provided by CrossRef, it is up to date dump.
Even now watchdogs are failing to prevent leaking academic papers from inside universities. Prevention of leaking papers from entire country would be definetly impossible.
🔬 The success of AI research hinges on access to large amounts of high-quality data. Scholarly publications are a treasure trove of knowledge that could empower AI advancements in healthcare, education, and climate change. 🌍🩺🎓
@AnjaneyaSitaRam
We have decided not to raise funds via fiat currencies; it is our principal position. But thank you for your intentions. You can help by talking about us in the media, participating in development, or by seeding.
@Nexus_search
Dude, I wish you too pay bills for Nexus.
Though... thanks, hope you squatted the name with good intentions and won't promote cryptocurrencies or anti-science here.