International Business Machines Corporation

Si Tong Zhao

Zhong Fang Yuan

Tong Liu

Xiao Lin Sun

Yuan Yuan Ding

Yi Chen Zhong

Xiang Yu Yang

Tian Ji Yang

Hai B Zou

Teng Sun

Ya Dong Li

Ya Juan Dang

Xueliang Zhao

Ziqiumin Wang

Jing Wen Xu

Wen Jie Hao

Yu Zui Wy You

Frank Feng

Teng Jiao Li

Xuan Yin Xia

Provided are techniques for identifying and correcting model focus drift during model training. A model is trained using an original dataset with data, and the is classified into clusters. A representation is assigned to each of the clusters. A first visualization dashboard with visualizations is generated, where each visualization represents a first data distribution of an associated cluster using the representation assigned to that cluster. The model is fine-tuned using a fine-tune dataset. A second visualization dashboard is generated by updating each visualization, where each visualization represents a second data distribution of the associated cluster. It is determined that a cluster of the clusters has focus drift based on changes between the first data distribution and the second data distribution. The focus drift is corrected by: adding the data of the cluster to the fine-tune dataset to form a combined dataset and fine-tuning the model using the combined dataset.

Identifying and correcting model focus drift during model training

In an approach for road section recognition using multi-modal cognitive mechanism, a processor receives an audio signal from a road test. A processor processes the audio signal to generate an acoustic spectrum density distribution map to identify a respective at least one road section switching point in a first mode. A processor processes a spectrogram of the audio signal to identify the respective at least one road section switching point in a second mode. A processor uses a machine learning model to predict an expected sound at each frame of the audio signal, to calculate a similarity between the expected sound and an actual sound, and to identify the respective at least one road switching point when the similarity is lower than a pre-set similarity threshold in a third mode. A processor combines results of the three modes to obtain a final set of road section switching points.

Multi-modal cognitive mechanism for road section recognition

According to one embodiment, a method, computer system, and computer program product for building an acoustic model is provided. The present invention may include performing contrastive pre-training of the acoustic model; building a dataset classifier using prompt engineering; performing a prediction process; and performing zero-shot audio prediction using the pre-trained acoustic model.

Adaptable acoustic model built with limited labeling data

Information extraction and image restructuring includes generating semantic vectors to encode portions of text extracted from a document. For each semantic vector a semantic similarity between a schema key and other text encoded therein is determined based on their respective positions within the document. An enhanced NLP model is created using the semantic vectors, each labeled according to the semantic similarity. The text, including schema key, are re-encoded as a key and candidate vectors. Key-value pairs are generated by matching the key vector with a predetermined number of candidate vectors. The enhanced NLP model, using prompt learning, is repurposed to perform a next-sentence prediction that predicts which of the candidate vectors is logically related to the schema key. Based on the next-sentence prediction, the discrete portion of text identified as the schema key and portion of text determined to be logically related thereto are output.

Information extraction and image re-ordering using prompt learning and machine-reading comprehension

In order to perform a semantic search based on a graph database, sets of nodes are selected from a plurality of nodes in a graph database. A set of nodes semantically matches a keyword in a natural language query. At least one target node is identified in the sets of nodes. A path is selected from candidate paths based on similarities between the candidate paths and a plurality of paths in the graph database. A graph query for retrieving information from the graph database is generated based on the selected path and the query target.

Semantic search based on a graph database

Growing community of inventors

Xi'an, China

Si Tong Zhao