Baidu open-sourced 10 versions of its ERNIE 4.5 model family. Key highlights: —Models from 300M to 424B params
—424B version beats DeepSeek V3 on 22/28 benchmarks
—“Heterogeneous” training architecture for increased efficiency
—Apache 2.0 license
RESEARCH
-

Baidu Open-Sources 10 ERNIE 4.5 Model Versions, 424B Beats DeepSeek V3
By
–
-
Microsoft’s MAI Diagnostic Orchestrator Outperforms Doctors in Medical Cases
By
–
Microsoft just dropped MAI Diagnostic Orchestrator — a medical AI that outperforms doctors.
— Rowan Cheung (@rowancheung) 1 juillet 2025
It solved 85.5% of 304 cases vs. just 20% by experienced physicians (5–20 yrs)
Also delivered greater cost savings than human doctorspic.twitter.com/wCB3bpbsuiMicrosoft just dropped MAI Diagnostic Orchestrator — a medical AI that outperforms doctors. It solved 85.5% of 304 cases vs. just 20% by experienced physicians (5–20 yrs) Also delivered greater cost savings than human doctors
-

Multiple LLMs Solve Previously Unsolvable ARC-AGI-2 Examples Together
By
–
Many ARC-AGI-2 examples that were unsolvable by any single LLM were solved by combining multiple LLMs. In some cases, an initially incorrect attempt by o4-mini is used by R1-0528 and Gemini-2.5-Pro as a hint to get to the correct solution. ARC-AGI-2 code: https://
github.com/SakanaAI/ab-mc
ts-arc2
… -

Graph Data Modeling with Python: Practical Guide for Data Scientists
By
–
Graph Data Modeling with #Python — A practical guide to curating, analyzing, and modeling data with graphs: http://
amzn.to/3PYxCPF v/ @PacktPublishing ——
#DataScience #GraphDB #LinkedData #AI #MachineLearning #DataScientist
——
𝒦𝑒𝓎 𝐹𝑒𝒶𝓉𝓊𝓇𝑒𝓈: Transform relational -

Alexandr Wang Becomes Meta Chief AI Officer
By
–
I’m excited to be the Chief AI Officer of @Meta
, working alongside @natfriedman
, and thrilled to be accompanied by an incredible group of people joining on the same day. Towards superintelligence -
AB-MCTS: Inference-Time Scaling for Frontier AI Model Cooperation
By
–
Inference-Time Scaling and Collective Intelligence for Frontier AIhttps://t.co/3qSUaEixQU
— hardmaru (@hardmaru) 1 juillet 2025
We developed AB-MCTS, a new inference-time scaling algorithm that enables multiple frontier AI models to cooperate, achieving promising initial results on the ARC-AGI-2 benchmark.… pic.twitter.com/9guSv7Dtv0Inference-Time Scaling and Collective Intelligence for Frontier AI https://
sakana.ai/ab-mcts/ We developed AB-MCTS, a new inference-time scaling algorithm that enables multiple frontier AI models to cooperate, achieving promising initial results on the ARC-AGI-2 benchmark. -

Microsoft AI diagnostic study strengths limitations analysis
By
–
This thread has a very good summary of the strengths and limitations of the recent Microsoft study of AI diagnostic ability.
-
Stanford Expert on AI Use in Academic Settings
By
–
How are Stanford experts using AI? In this new @Stanford series, @StanfordHAI Senior Fellow @Rbaltman explains why he permits AI in classes, rejects it for recommendation letters, and feels both cautious and optimistic about its future:
-
AI Mars and Immortality: Are We Dreaming Big Enough?
By
–
A.I., Mars and Immortality: Are We Dreaming Big Enough? | Interesting Ti… https://
youtu.be/vV7YgnPUxcU?si
=uepcf84-pWT5OEg1
… via @YouTube -
Meta recruits AI experts pursuing superintelligence development
By
–
Meta poaches 11 AI all-stars as Zuckerberg teases 'Superintelligence' https://
youtu.be/0vgV9jwdTcw?si
=0k8jza9rSlpY2RxX
… via @YouTube