R-4B Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning
RESEARCH
-

EmbodiedOneVision: Vision-Text-Action Pretraining for Robot Control
By
–
EmbodiedOneVision Interleaved Vision-Text-Action Pretraining for General Robot Control
-
Physics counting argument for high performance robotics
By
–
Thx for your comments. Re: “Conceptual robotics can largely be solved using data, but high performance robotics cannot….There are reasons and theory behind this insight, that I will return to in some other text.” I’m curious: is there a counting argument from physics?
-

Jet-Nemotron and PostNAS Optimize LLM Efficiency-Accuracy Trade-offs
By
–
Jet-Nemotron & PostNAS: Redefining the Efficiency–Accuracy Frontier in LLMs @NVIDIA
’s latest research introduces Jet-Nemotron, a family of hybrid-architecture LMs powered by Post Neural Architecture Search (PostNAS). Instead of costly pretraining from scratch, PostNAS starts -

High School Mathematics Foundations for Data Science Machine Learning
By
–
High school #Mathematics — foundation for #DataScience and #MachineLearning →
http://
bit.ly/31x9U14
————
#AI #ML #DataScientist #Calculus #LinearAlgebra #Probability #Statistics #Python
————
Learn Beginning, Intermediate, and Advanced Calculus here: http://
amzn.to/3KFVOjG -
Image Generation Flaws Need Comprehensive Benchmark Testing
By
–
Yes, that makes sense. Image generation is deeply flawed today. Would love to see a range of these questions used across a variety of imagegen and done over time. Would be an interesting benchmark.
-
New In-House AI Models: Token Architecture and Efficiency
By
–
Dove into our two new in-house models, token sculpture, scale vs efficiency, and more with @ReedAlbergotti and @semafor this week
-

Measuring AI Progress: Benchmarks and Advancement Over Time
By
–
It is worth measuring a wide range of benchmarks to see strengths and weaknesses. But benchmarks need to be repeated over time to measure progress (I have never claimed progress on all of them, btw). This is your prompt in Midjourney v1. Clearly large advances since then.
-
AI Critics Dismiss Technology as Bubble Destined to Pop
By
–
I dunno, I interact with the criticisms & critics of AI all the time (and have written some influential criticisms myself, especially on education), but I find a large number of critics really do believe that AI is useless and will go away once some notional bubble pops.
