Ученые KAIST разработали технологии, устраняющие галлюцинации искусственного интеллекта

However, in real-world environments, these models can misinterpret the physical characteristics of sensors, mistakenly identify objects, or claim to hear sounds that are not actually present simply because a certain object appears in a video. Researchers call these errors hallucinations. Knowing that, a new technology is in development that corrects such information confusion and physical misperceptions in AI, said researchers at Korea Advanced Institute of Science and Technology (KAIST). The first technology developed by the research team is the Diverse Negative Attributes (DNA) optimization method, which helps AI accurately understand the physical characteristics of special camera sensors such as thermal, depth, and X-ray sensors. Mistakes in Images Existing AI models often failed to understand the physical meaning of such images, for example by mistaking bright areas in thermal images for simple light reflection. Researchers team built VS-TDX, the first comprehensive benchmark for evaluating diverse vision sensors, and used the types of wrong answers that AI frequently produces as learning signals to help the model internalize the characteristics of each sensor. As a result, the AI gained a “new eye” that allows it to accurately infer the state of objects even in darkness or smoke. The second technology is Modality-Adaptive Decoding (MAD), a control method that blocks hallucinations caused by confusion between visual and auditory information at the source. This technology prevents AI from mistakenly claiming it hears a sound that does not actually exist simply because a certain object appears in a video. MAD works by having the AI self-assess whether vision or audio is more important for a given task, and then increasing the weight of the more relevant modality in real time. A key advantage of this technology is it can immediately suppress hallucination errors without costly model retraining, as it is training-free. Instead of retraining AI models at large scale with massive computing resources, the research team maximized cost efficiency by introducing the DNA method, which enables fine adjustment with only a small amount of data, and the MAD plug-in approach, which requires no additional training at all. Technology Applications These technologies could apply to autonomous vehicles operating at night or in bad weather, robots performing missions in smoke-filled environments, and unmanned aerial vehicles using thermal cameras. They also could be useful in fields that process multiple types of sensor information together, such as airport X-ray security screening and medical image analysis. “This research is significant because it reduces AI’s sensory bias and misperceptions without large-scale retraining,” Ro said. “It will serve as a foundation for building multimodal AI that can be trusted in real-life and industrial settings.” This achievement was notable for its continuity, with Sangyun Chung, a doctoral student in KAIST’s School of Electrical Engineering, participating as first author in both studies. Dr. Youngjun Yoo also participated as co-first author in the DNA study. Click here to view the paper.