MMBench-GUI Hierarchical Multi-Platform Evaluation Framework for GUI Agents
RESEARCH
-

VLMs Outperform with Visual Data Alone Over Numerical
By
–
Interestingly, VLMs perform better with plot alone compared to plot + data! This paper shows that GPT‑4.1 and Claude 3.5 excels at scatterplot analysis, and could often perform better only by looking at the plot, without reading the actual numerical data.
-

SJTU’s Neural Architecture Search Claims Superintelligence Amid Criticism
By
–
SJTU's latest paper claims artificial superintelligence for neural architecture search. But critics on alphaXiv call it inconsistent and "buzzword maxxing"
– Seemingly AI generated sections of the paper
– Relies on objective benchmarks and metrics it critiques -
Same Model, Different Names: Testing Human Perception Bias
By
–
it would be really funny / telling if we dropped the same model twice with two differently suggestive model names and people got really excited about identifyin the difference
-
Book Clusters and Attention Mechanisms in AI
By
–
book clusters is great. roughly ~attention/meaning.
-

GLM-4.5 Models Released with MIT License by Zai
By
–
Pretty decent pelicans from the new GLM-4.5 and GLM-4.5 Air models. Both models are MIT licensed, released by @Zai_org this morning https://
simonwillison.net/2025/Jul/28/gl
m-45/
… -
Image pretraining impact on text-based AGI evaluation benchmarks
By
–
maybe our crux here is that most of the 'AGI evals' are text-based. so my point is that adding image pretraining doesn't help on any of the (text-based) evals.
-
MLPs and Transformers: Key Components for AI Intelligence
By
–
ah i guess you could argue CV gave us MLPs which are a part of transformer which is important for intelligence. that's a fair point
-
First Real Artificial Superintelligence Emergence Discussed
By
–
This Might Be the FIRST Real Artificial Superintelligence… https://
youtu.be/vgUHu-SF4Hk?si
=nNbNHy5BD4yDy9Yx
… via @YouTube -
Linguistics Approach Limitations in AI Development
By
–
that's actually a good point. linguistics turned out not to help either.
