Common conception, shattered again. Normalization isn’t fundamental to Transformers, and this approach keeps getting better This paper introduces Dynamic erf (Derf), which further improves normalization-free Transformers With strong empirical results while generalizing better
@askalphaxiv
-
Correcting LLM-as-a-Judge Bias Statistical Calibration Method
By
–
How to Properly do LLM-as-a-Judge Raw LLM-as-a-Judge scores are inherently biased due to how LLMs would often make mistakes This paper proposes a simple statistical method to correct the scores and calculate valid confidence intervals via a human-verified calibration set
-
ArXiv paper abstract on machine learning or scientific research
By
–
https://
alphaxiv.org/abs/2511.21140 -
Depth Anything 3: Unified 3D Reconstruction Model ByteDance
By
–
Depth Anything 3 Bytedance Seed proposes a simple & scalable unified model for recovering 3D visual space from any number of views with or without known camera poses. Absolute top notch quality with SoTA 3D reconstruction and camera pose estimation
-
AI4Science Panel at NeurIPS: Future of AI-Enabled Discovery
By
–
AI is transforming scientific discovery at incredible speed Join us at NeurIPS for our AI4Science Panel with Rafael Gómez-Bombarelli, @jeffclune @james_y_zou as they discuss what the future of AI-enabled science might look like. Link to signup below
-
Meta’s Three-Gate Theory Explains Sparse RL Updates in LLMs
By
–
This Meta paper just gave an empirical proof of why RL updates in LLMs are so sparse! They proposed this "Three-Gate Theory” & showed that pretrained models have highly structured optimization landscapes, so geometry-aware RL is much better than heuristic based SFT methods
-
EGGROLL: Gradient-Free LLM Training Achieves 100x Throughput
By
–
Could evolution replace gradients for LLM training? Introducing EGGROLL, a new paper that makes gradient-free training actually practical 100x training throughput vs vanilla ES 100x memory reduction 91% inference speed
Pure integer training Trending #1 on alphaXiv -
Papers With Code SOTA Returns with Gemini 3 AI Research
By
–
Papers With Code SOTA is back 🚀
— alphaXiv (@askalphaxiv) 20 novembre 2025
We used Gemini 3 to process millions of charts and tables across arXiv + the web to surface state-of-the-art AI research in reasoning, computer-use, OCR, and more
Track the best-performing methods and see which benchmarks are getting adopted pic.twitter.com/9yKJXL35K7Papers With Code SOTA is back We used Gemini 3 to process millions of charts and tables across arXiv + the web to surface state-of-the-art AI research in reasoning, computer-use, OCR, and more Track the best-performing methods and see which benchmarks are getting adopted
-
AI Research Organization Seeks Talent at NeurIPS Conference
By
–
Our mission is to help you navigate the rapidly evolving frontier of AI research and seamlessly apply the latest research to your projects. If this aligns with your mission, shoot a note to hiring@alphaxiv.org or come meet us at NeurIPS!
-
alphaXiv Bridges AI Research-to-Practice Divide with Integrated Platform
By
–
alphaXiv is solving this by bridging the AI research-to-practice divide. We’re bringing together papers, benchmarks, and implementations into a single platform that makes it seamless for any AI practitioner to discover and build on top of the latest research.