Seoul National University College of Engineering announced that a research team led by Professor Hyun Oh Song from the Department of Computer Science and Engineering has developed a new AI technology called KVzip that intelligently compresses the conversation memory of large language model (LLM)-based chatbots used in long-context tasks such as extended dialog and document summarization. The study is published on the arXiv preprint server.

The term conversation memory refers to the temporary storage of sentences, questions, and responses that a chatbot maintains during interaction, which it uses to generate contextually coherent replies. Using KVzip, a chatbot can compress this memory by eliminating redundant or unnecessary information that is not essential for reconstructing context. The technique allows the chatbot to retain accuracy while reducing memory size and speeding up response generation—a major step forward in efficient, scalable AI dialog systems.

Modern LLM chatbots perform tasks such as dialog, coding, and question answering using enormous contexts that can span hundreds or even thousands of pages. As conversations grow longer, however, the accumulated conversation memory increases computational cost and slows down response time.

Researchers in Shanghai have reported in a study, recently published in Science Advances, that they’ve successfully decoded Mandarin Chinese language in real time with the help of a brain-computer interface (BCI) framework, a first for BCIs working with tonal languages. The participant involved in the study was also capable of controlling a robotic arm and digital avatar and interacting with a large language model using this new system.

While most people may not want a computer reading their mind, those who are unable to speak due to neurological conditions, like strokes or amyotrophic lateral sclerosis (ALS), need to find alternative ways to communicate. Speech decoding BCIs, capable of decoding neural signals, offer a promising way to restore communication in such individuals. In addition to communication, BCIs also offer ways to control devices directly through thought. This is particularly helpful for neurological conditions in which disabilities extend beyond speech loss.

These types of devices are not exactly a novel technology, but most BCI speech decoding research has focused on English, a non-tonal language.