Machine Learning for Information Extraction
2 repos
Machine learning systems and models designed to extract structured information from unstructured text and documents. The cluster centers on neural architectures and training approaches for tasks like relation extraction, question answering over tables, table-to-text generation, and semantic role labeling. While specific languages aren't recorded, the central repositories (flare-fsrl, flare-ma, flare-cfa, flare-finred, flare-tatqa, flare-tsa) reflect work on few-shot learning, multi-task adaptation, and curriculum learning strategies for improving extraction performance across diverse information types.
