Abstract:This paper addresses the challenge of Database Entity Recognition (DB-ER) in Natural Language Queries (NLQ). We present several key contributions to advance this field: (1) a human-annotated benchmark for DB-ER task, derived from popular text-to-sql benchmarks, (2) a novel data augmentation procedure that leverages automatic annotation of NLQs based on the corresponding SQL queries which are available in popular text-to-SQL benchmarks, (3) a specialized language model based entity recognition model using T5 as a backbone and two down-stream DB-ER tasks: sequence tagging and token classification for fine-tuning of backend and performing DB-ER respectively. We compared our DB-ER tagger with two state-of-the-art NER taggers, and observed better performance in both precision and recall for our model. The ablation evaluation shows that data augmentation boosts precision and recall by over 10%, while fine-tuning of the T5 backbone boosts these metrics by 5-10%.
Abstract:The recent meteoric advancements in large language models have showcased a remarkable capacity for logical reasoning and comprehension. These newfound capabilities have opened the door to a new generation of software, as has been made obvious through the innumerable ways they are being applied in the industry. This research focuses on harnessing and guiding the upgraded power of LLMs to construct a framework that can serve as an intermediary between a user and their user interface. By comprehending a user's needs through a thorough analysis of natural textual inputs, an effectively crafted LLM engine can classify the most likely available application, identify the desired UI component and subsequently execute the user's expected actions. This integration can evolve static UI systems into highly dynamic and adaptable solutions, introducing a new frontier of intelligent and responsive user experiences. Such a framework can fundamentally shift how users accomplish daily tasks, skyrocket efficiency, and greatly reduce cognitive load.