Foundation Models
Pretraining, architectures, scaling behavior, data, and representation learning.
|
Discover, Share & Explore Amazing Blogs
Discover high-quality research blogs with broad technical insight across AI research and engineering. Follow the latest academic and industry thinking through curated, source-linked reading.
Why Blog?
Transform your research insights into accessible knowledge. Share discoveries, methodologies, and breakthroughs that advance scientific understanding and inspire future researchers.
Why BlogrXiv?
A dedicated platform for researchers to publish, discover, and collaborate on cutting-edge research. Built by researchers, for researchers, with advanced tools for knowledge sharing and discovery.
BlogrXiv For Researcher
Empowering researchers with tools for academic writing, peer collaboration, and knowledge dissemination. Join a community of scholars advancing human knowledge through open, accessible research communication.
Selected reading
Recently Published Blogs
Technically substantive writing selected for clarity, specificity, and lasting research value.
Taxonomy
Browse by category
A compact research taxonomy spanning models, systems, evaluation, and scientific practice.
LLM & MLLM
Reasoning, instruction tuning, multimodal language models, and evaluation.
Multimodal Systems
Vision-language systems, perception, grounding, video, and embodied understanding.
Visual Generation
Diffusion, image and video synthesis, controllability, and generative evaluation.
World Models
Learned simulators, dynamics, planning representations, and spatial reasoning.
AI Agents
Tool use, memory, orchestration, computer use, autonomy, and agent evaluation.
Efficient AI
Optimization, inference, training systems, compression, kernels, and hardware.
Trustworthy AI
Safety, alignment, interpretability, robustness, auditing, and governance.
Research Craft
Evaluation methodology, scientific process, research taste, and field-building.
Research voices
Bloggers worth following
Researchers and technical writers represented by multiple high-signal pieces in the index.
Selection protocol
Curation over aggregation.
Every indexed piece should help a technical reader understand a mechanism, method, system behavior, or research lesson.
- 01
Technical contribution
Teaches more than the announcement or abstract alone.
- 02
Specificity
Provides enough evidence and detail to support technical judgment.
- 03
Durable value
Remains useful after the immediate release cycle has passed.
Open index
Know a piece that belongs here?
Submit rigorous technical writing with a short note explaining its research value.