“ELF: Embedded Language Flows” Most diffusion LMs still treat language as discrete tokens, or keep forcing continuous states back into token space during generation. This paper runs diffusion almost entirely in continuous embedding space, then convert to tokens only at the
RESEARCH
-
Comparative analysis of GQA, MLA, and Gated DeltaNet model architectures
By
–
Yes! It would be interesting whether this happens in GQA:SWA models only though or also in MLA or Gated DeltaNet models, for example.
-
TACO: A New Training-Free Framework for Terminal AI Agents
By
–
What if your terminal agent could learn to ignore the noise and keep only what matters? Researchers from University of Manchester, HKUST, and Beihang University present TACO — a plug-and-play, training-free framework that automatically discovers and refines compression rules
-
Team of agents solving a physics problem
By
–
watching a team of agents tackling a hard theoretical physics problem is quite mesmerizing – self-correcting, deriving hard equations, computing intermediate results, re-estimating the best approach https://t.co/EySjaCtUFZ pic.twitter.com/RhUmNXkGLB
— Thomas Wolf (@Thom_Wolf) 13 mai 2026Watching a team of agents tackle a challenging theoretical physics problem is quite mesmerizing—they self-correct, derive complex equations, compute intermediate results, and re-evaluate the best approach.
-
Machine Learning Applied to Platinum Fuel Cell Catalyst Research
By
–
Platinum Fuel Cell Catalyst Research Using Machine Learning! #BigData #Analytics #AI #MachineLearning #DataScience #IoT #IIoT #Python #RStats #TensorFlow #JavaScript #ReactJS #CloudComputing #Serverless #DataScientist #Linux #Programming #Coding #100DaysofCode
-
Using AI assistants for LaTeX formula extraction in research papers
By
–
Need a quick way to extract a formula?
— alphaXiv (@askalphaxiv) 13 mai 2026
alphaXiv’s assistant actually supports LaTeX generation!
Just simply highlight the equations in the paper, and ask the assistant for the LaTeX code, then copy it to the workspace of your choice. pic.twitter.com/mQd0a0w6VzNeed a quick way to extract a formula? alphaXiv’s assistant actually supports LaTeX generation! Just simply highlight the equations in the paper, and ask the assistant for the LaTeX code, then copy it to the workspace of your choice.
-
Learning LLM Architectures Through Scratch Implementation and Model Evaluation
By
–
A little talk on what we can learn from implementing LLM architectures from scratch in Python and PyTorch. And how I approach new open-weight models, compare them against reference implementations etc:
-
δ-mem: An efficient online memory mechanism for LLMs
By
–
// δ-mem: Efficient Online Memory for LLMs // One of the more elegant memory mechanisms I've seen this month. Most long-term memory work either inflates context or retrains the model. This paper shows a tiny external state, coupled directly into the attention computation, can