Research & Papers
MIT researchers teach AI models to interpret charts
MIT researchers have developed ChartNet, a new training dataset designed to improve how vision-language models...
MIT News AI
Research & Papers · Other Companies
Research & Papers
How to Speed Up Transformer Training Using NVIDIA Apex (FusedAdam, FusedLayerNorm) and Native torch.amp
This article discusses optimizing Transformer model training performance using NVIDIA Apex tools including FusedAdam...
MarkTechPost
Research & Papers
Turing Award winner Richard Sutton says pure generative AI can't do real science
Turing Award winner Richard Sutton argues that pure generative AI lacks the ability to evaluate its own results,...
The Decoder
Research & Papers
Nvidia bets big on physical AI at GTC Taipei with a new world model, driving brain, and open humanoid robot
Nvidia launched new AI models at GTC Taipei including Cosmos 3 world model, Alpamayo 2 Super driving model, and an open...
The Decoder
Research & Papers
Parallax: A Parameterized Local Linear Attention That Keeps Softmax and Adds a Learned Covariance Correction Branch
Parallax is a new parameterized local linear attention mechanism that improves upon Local Linear Attention by replacing...
MarkTechPost
Research & Papers
Trajectory Releases a Concurrent Multi-LoRA Training Stack for Continual Learning, Reporting a 2.81× Experiment-Throughput Gain
Trajectory, in collaboration with UC Berkeley Sky Lab and Anyscale, developed a concurrent multi-LoRA training stack...
MarkTechPost
Research & Papers
Making AI chatbots helpful weakens their ability to simulate human behavior, large-scale study finds
A large-scale study with 208,000 participants found that training language models to be helpful chatbots paradoxically...
The Decoder
Research & Papers
Terence Tao argues AI could bring division of labor to math for the first time in history
Mathematician Terence Tao argues that AI could fundamentally transform mathematics by enabling division of labor for...
The Decoder
Research & Papers
Genesis AI Releases Nyx, Quadrants, and Genesis World 1.0 Physics Platform for Scalable Robotics Foundation Model Evaluation
Genesis AI released Genesis World 1.0, a four-component simulation platform for robotics foundation model evaluation,...
MarkTechPost
Research & Papers
Meet mKernel: A Multi-GPU, Multi-Node Fused Kernel Library for GPU-Driven Communication
UC Berkeley's UCCL team has released mKernel, a fused kernel library that combines intra-node NVLink communication,...
MarkTechPost
Research & Papers
RSI is the new AGI — and it’s just as hard to pin down
Multiple AI labs are pursuing recursive self-improvement (RSI) as a path toward artificial general intelligence, but...
TechCrunch AI
Research & Papers
A Coding Guide to Implement a pgvector-Powered Semantic, Hybrid, Sparse, and Quantized Vector Search System
This tutorial demonstrates how to build a vector search system using pgvector in PostgreSQL, integrating semantic,...
MarkTechPost
Research & Papers
Sakana AI Proposes DiffusionBlocks: a Block-wise Training Framework That Converts Residual Networks into Independently Trainable Denoising Modules
Sakana AI proposes DiffusionBlocks, a novel framework that converts residual networks into independently trainable...
MarkTechPost
Research & Papers
MEMO: A Modular Framework for Training a Dedicated Memory Model on New Knowledge Without Modifying LLM Parameters
Researchers from NUS, MIT, and A*STAR introduce MEMO, a modular framework that trains a separate memory model to encode...
MarkTechPost
Research & Papers
Design a High-Precision Retrieve-and-Rerank Pipeline with ZeroEntropy Zerank-2 Reranker
This tutorial demonstrates how to build a high-precision retrieve-and-rerank pipeline using the ZeroEntropy Zerank-2...
MarkTechPost
Research & Papers
Stability AI Releases Stable Audio 3: A Family of Fast Latent Diffusion Models for Audio Generation and Editing
Stability AI has released Stable Audio 3, a family of latent diffusion models for audio generation and editing with...
MarkTechPost
Research & Papers
Import AI 458: Reckoning with the future; and a singularity story
An article from Import AI 458 discussing potential AI-driven breakthroughs and developments expected in the coming...
Import AI
Research & Papers
Design a Complete Multimodal RLVR Pipeline with Open-MM-RL, Vision-Language Prompting, Reward Scoring, and GRPO Export
This tutorial demonstrates how to build a complete multimodal reinforcement learning pipeline using the Open-MM-RL...
MarkTechPost
Research & Papers
Step by Step Guide to Build and Compare FedAvg and FedProx Federated Learning on Non-IID CIFAR-10 with NVIDIA FLARE
This tutorial demonstrates how to build and compare FedAvg and FedProx federated learning algorithms using NVIDIA FLARE...
MarkTechPost