.
@shoyer We admire neuralGCM and all the contributions you are making for AI+climate modeling. Social media doesn't allow for too much nuance – what I meant to say was FourCastNet 3 is unprecedented in offering competitive skill at 6-hour resolution, with probabilistic estimates
RESEARCH
-
FourCastNet 3 Achieves Competitive Skill at 6-Hour Resolution
By
–
-
Continuous Learning and Personalization in AI Systems
By
–
In principle I agree, but in practice we have a small niche use case and this (taste) data won't make it out into the outside world. There'll be millions of people who have different taste requirements and continuous learning for them would be a massive unlock. Yes we can fine
-
RAG Applications Across Industries Explained
By
–
Want to learn more about where RAG is being applied across industries? https://
nvda.ws/3TMPiye -
Gemini Receives IMO Official Confirmation While OpenAI Awaits
By
–
It looks to me like Gemini may have got official confirmation from IMO while OpenAI did not
-
LLMs Lack Continuous Learning for Creative Quality Assessment
By
–
But don't you think that at least some of taste would come from continuous learning? We are working on a creative project now and it is a pain the LLMs are incapable of learning what is good and what is bad, while humans are learning pretty quickly, at least to some base level
-
In-context learning and competitive performance in systems
By
–
+1 Two points
1) this is really just good old in context learning
2) as already mentioned, another system without this also scored gold. -
Frontier Models Overview: Current AI Landscape Assessment
By
–
Alright, this is just frontier models. Not ranking the companies or products.
-

Qwen3-235B-A22B-2507 released with performance boost
By
–

Qwen released Qwen3-235B-A22B-2507, a minor upgrade to Qwen3-235B-A22B, which provides a significant performance boost. Already available on Qwen chat
-
AI Struggles with 2D Grid-Based Square Problems
By
–
I hear AI *hates* problems based on 2D grids of squares
-

Qwen3 Model Excels at Reasoning and Creative Writing Tasks
By
–
Very nice post-training lesson from the new Qwen3 model: – A pure reasoning model is expected to perform better on pure reasoning benchmarks. Nothing surprising. – We previously saw that pure reasoning models also perform very well on non-reasoning tasks, like creative writing.