NEWS IN BRIEF: AI/ML FRESH UPDATES

Get your daily dose of global tech news and stay ahead in the industry! Read more about AI trends and breakthroughs from around the world

AI learns to decipher charts

Researchers from MIT and the MIT-IBM Computing Research Lab developed ChartNet, a dataset and series of open-source models that outperform commercial AI models in tasks like chart interpretation. This breakthrough could empower small firms with limited budgets to leverage AI for business trend analysis and scientific figure interpretation.

Mastering Questioning with Battleship

In 2026, AI agents excel at tasks like customer service, but struggle with complex inquiries. MIT and Harvard researchers improved AI's ability to ask questions through a "Battleship" game, leading to significant gains in performance and efficiency.

Mastering Hyperparameter Optimization on Amazon Nova Forge

Amazon Nova Forge allows users to build customized language models that blend proprietary data with curated datasets, preventing catastrophic forgetting and improving domain performance without degrading general capabilities. The tool helps navigate the challenges of hyperparameter tuning for domain-specific tasks, avoiding expensive failures and ensuring the right balance between stability and...

Enhancing Attention with Parallax Correction

A new paper introduces 'Parallax,' a parameterized Local Linear Attention mechanism that enhances efficiency without cutting compute. Parallax simplifies and improves the LLA framework, making it more efficient and easier to implement, with the potential to scale to LLM pretraining and codesign with Muon.

Secure Agentic Payments with Amazon Bedrock AgentCore

Amazon Bedrock AgentCore payments, in partnership with Coinbase and Stripe (Privy), allows agents to access paid resources on behalf of end users. AgentCore addresses risks like runaway spending and lack of end user consent in autonomous payment systems.

Boost LLM Model Loading with GPUDirect on Amazon FSx

Large language models (LLMs) on AWS GPU instances face lengthy model load times. Amazon FSx for Lustre and NVIDIA GPUDirect Storage (GDS) drastically reduce load times, improving total time to first token (TTFT) from minutes to seconds for models like Llama 3.1 with 405B parameters on AWS P6e UltraServers.

Efficient Approximation of SVR with Trimmed Kernel Ridge Regression

Kernel ridge regression (KRR) and support vector regression (SVR) are machine learning techniques that can be combined to create a sparse KRR model approximating an SVR model. This hybrid approach offers the benefits of KRR's large dataset handling and SVR's efficiency in model storage, demonstrating high predictive accuracy in a demo using the scikit KernelRidge module.