WinBuzzer

AI Coding: Microsoft’s 7B X-Coder Outperforms 14B Rivals on Synthetic Data

Microsoft and Tsinghua University have developed a 7B-parameter AI coding model that outperforms 14B rivals using only ...
The Robot Report

1X launches world model enabling NEO robot to learn tasks by watching videos

With this update, 1X Technologies' NEO leverages internet-scale video data fine-tuned on robot data to perform AI tasks.

The longest-lasting tools I own are older than me

Tough and charming, some old tools will outlive their owners but not their function -- hammering, cutting, and chopping for ...

What’s the most reliable weather forecasting model? It’s complicated

Patricia Sanchez, a National Weather Service meteorologist at the Fort Worth office, said the most reliable model is ...

A smarter way to approach AI prompting

Prompts describe tasks. Rubrics define rules. Here’s how rubric-based prompting reduces hallucinations in search and content workflows.

MemRL outperforms RAG on complex agent benchmarks without fine-tuning

MemRL separates stable reasoning from dynamic memory, giving AI agents continual learning abilities without model fine-tuning ...

MIT’s new ‘recursive’ framework lets LLMs process 10 million tokens without context rot

While standard models suffer from context rot as data grows, MIT’s new Recursive Language Model (RLM) framework treats ...

Physical AI And World Models Raise The Bar On What We Call ‘Smart’ ... And LLMs Are Not Enough

Instead, physical AI needs to orchestrate a blend of on-device processing for speed and cloud computation for long-term ...

DeepSeek’s Engram Conditional Memory Shows How to Reduce AI Compute Waste

DeepSeek's new Engram AI model separates recall from reasoning with hash-based memory in RAM, easing GPU pressure so teams ...
eLife

Predicting human decision-making across task conditions via individuality transfer

Encoding individual behavioral traits into a low-dimensional latent representation enables the accurate prediction of decision-making patterns across distinct task conditions.
IEEE

Instructive Code Retriever: Learn from Large Language Model’s Feedback for Code Intelligence Tasks

Abstract: Recent studies proposed to leverage large language models (LLMs) with In-Context Learning (ICL) to handle code intelligence tasks without fine-tuning. ICL employs task instructions and a set ...