We used DeepSeek OCR to extract every dataset from tables/charts across 500k+ AI arXiv papers for $1000 🚀
See which benchmarks are trending and discover datasets you didn't know existed
Doing the same task with Mistral OCR would've cost $7500 👀
Introducing NotebookLM for arXiv papers 🚀
Transform dense AI research into an engaging conversation
With context across thousands of related papers, it captures motivations, draws connections to SOTA, and explains key insights like a professor who's read the entire field
I would like to thank @aaipunairport for their excellent service. I forgot my laptop at security belt on Sep 26. I don’t know how They located me among hundreds or thousands of people and returned my laptop, even before I knew that I had lost it. Thank you!!!
A chatbot’s odd behavior exposed deeper risks: a legacy app, weak credentials, insecure APIs, & a compromised admin device, which led to a major security breach. See how @Qualys TotalAppSec shields your app stack to reduce exposure & strengthen resilience: https://t.co/QU8K5MBFbQ
Trending on alphaXiv today:
A 1B model just beat a 405B model at math reasoning by sampling 512 solutions and using learned verifiers to pick the best one 🚀
Not magic -- performance is highly task-dependent and requires careful pairing of models with verification strategies. But a cool result nonetheless demonstrating just how far optimized inference can take us.
Introducing TaskGen, an open-sourced task-based, memory-infused LLM agentic framework that uses an Agent to break down a task into subtasks and assigns each subtask to either an equipped function or another Agent.
Talk directly with @johntanchongmin and the TaskGen team here!
Site Updates!
- faster, more consistent load times
- support for older paper IDs: /math, /hep-th, /cs, /gr-qc
- improvements to claiming your paper with ORCID
Sharing an idea that offers an exciting use case of the AI Scientist from @SakanaAILabs! Rather than generating new ideas, one can adapt the code to pick a paper and test its results.
"Reproducibility as a platform."
Thanks lakshaytalkstocomputer and @cong_ml for the insight!
Excited to announce that alphaXiv just won the Magic Grant from the @BrownInstitute to continue making research more interactive! Beyond grateful for this opportunity and thrilled to continue connecting students and authors around the world!
@StanfordAILab@askalphaxiv Excited to be an advisor to arXiv. It's an awesome project and a great service to the community. Finally papers aren't one way any longer.
arXiv -> alphaXiv
Students at Stanford have built alphaXiv, an open discussion forum for arXiv papers. @askalphaxiv
You can post questions and comments directly on top of any arXiv paper by changing arXiv to alphaXiv in any URL!
In response to popular demand, you can now comment directly on any arXiv PDF through https://t.co/dEDfqpcYQg.
Fewer rendering issues, while still enabling interactive discussion directly on the paper!
Check out this cool addition to ArXiv, the open source scientific paper repository, which adds comments. Called alphaxiv, you can either change the URL of any https://t.co/RI2fxnMGmQ paper to https://t.co/h66lKWj6bU and see the comments or load a Chrome plug in that automatically adds them.
I hope this is incorporated in regular ArXiv soon