Yes, linked to in my post above: https://
research.google/blog/google-re
search-2025-bolder-breakthroughs-bigger-impact/
… That blog itself has links to individual blog post entries or papers about different specific pieces of work.
RESEARCH
-
Google Research 2025: Linked Blog Post on Breakthroughs and Impact
By
–
-
Unsupervised Learning Confusion: Training on Test Data Explained
By
–
Man discovers unsupervised learning and confuses it with what "training on test" actually means.
-
How AI Models Learn: Debunking Data Reproduction Myths
By
–
I’m tired of explaining how AI models learn. They don’t copy or reproduce data in the way you think they do.
-
Firefly and Licensed Data Training in Generative AI
By
–
What about models like firefly that are trained on licensed data?
-

Stanford Hiring Data Science Postdocs in Causal Inference
By
–
We're hiring data science postdocs! Work on interdisciplinary projects in causal inference, ML, econometrics, and more. Recent PhDs with coding expertise and cross-disciplinary experience are encouraged to apply: https://
datascience.stanford.edu/programs/data-
science-fellows
… -
Computer Vision: Much More Than Just Artificial Intelligence
By
–
Computer Vision isn’t “just AI.”
— Satya Mallick (@LearnOpenCV) 19 décembre 2025
It’s how machines interpret visual data—from cameras to X‑rays to telescopes—solving tasks like detection, recognition, OCR, and 3D reconstruction. #ComputerVision #AI #MachineLearning #DeepLearning #ImageProcessing pic.twitter.com/KhVODmTBBUComputer Vision isn't "just AI."
It's how machines interpret visual data—from cameras to X‑rays to telescopes—solving tasks like detection, recognition, OCR, and 3D reconstruction. #ComputerVision #AI #MachineLearning #DeepLearning #ImageProcessing -
Increasing Context Length and Model Capacity: Expensive Trade-offs
By
–
Either
a) you increase the context length and/or let the model crawl through stored data
b) you start of with a model that has large enough capacity and periodically update the weights
Both options are expensive, so I am not sure they actually want this right now. -
Mixture of Experts Models: Current Perspectives and Efficiency
By
–
Poll: How Are You Thinking About Mixture of Experts (MoE) Models Today? MoE architectures scale model intelligence by routing each request to a subset of specialized “experts,” improving efficiency. Which best matches your current perspective?
-
Who Is Shaping AI’s Future? Key Leaders Discuss
By
–
Who is really shaping the future of AI? with @AGraylin
, @sama
, @jruwitch and @TEDchris -

ODB-dLLM: Faster Parallelizable Diffusion-Based Language Models
By
–
Diffusion-based LLMs are fast and parallelizable, but bidirectional attention makes inference expensive due to repeated prefill and decoding. Enter ODB-dLLM, a dual-boundary framework with adaptive prefill length prediction and dLLM-specific jump-share speculative decoding.