We used DeepSeek-V3 to classify every AI paper on arXiv by topic (agents, VLMs, etc) 🚀

Now you can instantly filter to see what's trending in each area 🚨

The peer review process should be opened in economics. One way to do this is to embrace @alphaxiv.org www.alphaxiv.org/explore/papers

🚀The return of BERT, LLMs that beat physicians?!, and the dawn of agents to simulate users. Presenting Santa's nice list for AI! 🎅

- ModernBERT
- Superhuman performance of LLMs against Physicians (discussion with author @adamrodmanmd.bsky.social )
- LMagent: A Large-scale Multimodal Agents Society

Thank you so much @jieyusz.bsky.social and @mameister4.bsky.social! We're really glad that alphaXiv can help authors through community feedback and open discussion :)

🚀 Trending AI papers on alphaXiv this week

- Star Attention (discussion with author @shantanuacharya.bsky.social)

- One Diffusion to Generate Them All (discussion with author Duong Le from The Allen Institute for AI)

LLaVA-o1 is the first visual language model capable of systematic reasoning similar to GPT-o1 🚀

But how does it perform on multimodal math reasoning questions? 🔎

New numbers from LlaVA-o1 on the MathVision Dataset from author Guowei Xu

LLaVA-o1 (11B): 23.7%
Qwen2-VL-72B: 25.9%
Qwen2-VL-7B: 16.3%

Finding good data mixtures for LLM training can be tricky - Aioli provides a unified framework to construct pre-training data mixtures. Talk to the authors @MayeeChen @michahu.bsky.social @nicholaslourie.bsky.social @kchonyc Christopher Ré @HazyResearch directly here!