IndexCache, a new sparse attention optimizer, delivers 1.82x faster inference on long-context AI models

Researchers at Tsinghua University and Z.ai built IndexCache to eliminate redundant computation in sparse attention models ...

Google reveals algorithms to address AI memory challenges; memory and storage stocks drop

Google unveils TurboQuant, PolarQuant and more to cut LLM/vector search memory use, pressuring MU, WDC, STX & SNDK.
Science Daily

Study shows addressing working memory can help students with math difficulty improve word problem-solving skills

Working memory is like a mental chalkboard we use to store temporary information while executing other tasks. Scientists worked with more than 200 elementary students to test their working memory, ...
Decrypt

Google Shrinks AI Memory With No Accuracy Loss—But There's a Catch

The technique reduces the memory required to run large language models as context windows grow, a key constraint on AI ...

Nvidia says it can shrink LLM memory 20x without changing model weights

Nvidia's KV Cache Transform Coding (KVTC) compresses LLM key-value cache by 20x without model changes, cutting GPU memory ...
News Medical

MERLIN algorithm unlocks immune cell location memory in organs

A new AI-based method reconstructs spatial information about where immune cells were originally located in an organ, even after these cells have been removed from the tissue and analyzed individually.
Scientific Research Publishing

SymPcNSGA-Testing: A Hybrid Approach to Mitigate Path Explosion in Software Programs ()

To address these shortcomings, we introduce SymPcNSGA-Testing (Symbolic execution, Path clustering and NSGA-II Testing), a ...
Hosted on MSN

Planning post-stroke care: Calculator predicts six-month memory and thinking problems

Researchers at the University of Oxford have developed a "Stroke Cognition Calculator," a new tool designed to estimate a person's chance of having thinking and memory problems six months after a ...