Abstract: Neural code models (NCMs) have demonstrated extraordinary capabilities in code intelligence tasks. Meanwhile, the security of NCMs and NCMs-based systems has garnered increasing attention.

A team of researchers has found a way to steer the output of large language models by manipulating specific concepts inside these models. The new method could lead to more reliable, more efficient, ...

Researchers went under the hood of large language models to identify and influence specific concepts They were able to identify hallucinations and improve LLM outputs They uncovered vulnerabilities, ...

Now a team from MIT and the University of California San Diego has developed a way to test whether a large language model (LLM) contains hidden biases, personalities, moods, or other abstract concepts ...

Abstract: Capacitive power transfer (CPT) has attracted attention for its merits of light coupler structure, low cost, and minimal eddy current loss. However, constrained by high operating frequencies ...

According to @bcherny, developers can run /config in Claude Code to set output styles that change the assistant’s tone and format for coding tasks, including an explanatory style for understanding ...

Crystal plasticity finite element (CPFE) code for AM SX, to simulate the deformation at dendrite scale, under the open-source MOOSE software, using C++ language. The CPFE simulation is conducted under ...