A groundbreaking paper published @nature.com today—the first large health language model—predicts >1,000 diseases for a person and when.
Implications here:
erictopol.substack.com/p/dawn-of-a-...

New paper accepted at ACL Findings! TL;DR: While language models generally predict sentences describing possible events to have a higher probability than impossible (animacy-violating) ones, this is not robust for generally unlikely events and is impacted by semantic relatedness. 1/3

www.marktechpost.com/2025/02/10/s... #AI #artificialintelligence

Here is Tülu 3 405B 🐫 our open-source post-training model that surpasses the performance of DeepSeek-V3! It demonstrates that our recipe, which includes RVLR scales to 405B - with performance on par with GPT-4o, & surpassing prior open-weight post-trained models of the same size including Llama 3.1.

Model-Based Transfer Learning for Contextual Reinforcement Learning

Use a gaussian process on top of RL to choose the best training set that generalizes for CL in the same task, the GP is used to model the generalization gap and do BO

See, I told you GPs were cool!

arxiv.org/abs/2408.04498