Recently, Mitsui Chemicals announced that it has started a collaborative project with Blueqat Co., Ltd. to combine advanced technologies such as natural language processing and quantum computing to accelerate new applications of Mitsui Chemicals products.
Mitsui Chemicals said the project used a natural language processing model called sentence embedding for simple contrastive learning. SimCSE is a high-performance model that creates vector expressions to mine the meaning of documents and other texts. Mitsui Chemicals added that the model is capable of calculating high-precision semantic similarities between different texts and is expected to be used in areas such as information retrieval and question-answering systems.
Mitsui Chemicals said: "The partner company has successfully compressed the data of the SimCSE model effectively by using tensor network technology. Tensor networks are used for various tasks in quantum computing, such as compressing quantum circuit data and performing quantum simulations." The model is expected to find practical uses in areas such as patent searches and searches for new applications for Mitsui Chemicals products. Mitsui Chemicals said it will promote digital transformation by combining machine learning technology with quantum technology.
