Yuming Li — Page 3 | Searcharxiv

arXiv2022

A Novel DeBERTa-based Model for Financial Question Answering Task

As a rising star in the field of natural language processing, question answering systems (Q&A Systems) are widely used in all walks of life. Compared with other scenarios, the applicationin financial scenario has strong requirements in the traceability and interpretability of the Q&A systems. In addition, since the demand for artificial intelligence technology has gradually shifted from the initial computational intelligence to cognitive intelligence, this research mainly focuses on the financial numerical reasoning dataset - FinQA. In the shared task, the objective is to generate the reasoning program and the final answer according to the given financial report containing text and tables. We use the method based on DeBERTa pre-trained language model, with additional optimization methods including multi-model fusion, training set combination on this basis. We finally obtain an execution accuracy of 68.99 and a program accuracy of 64.53, ranking No. 4 in the 2022 FinQA Challenge.

cs.CL↗

arXiv2019

Lauca: Generating Application-Oriented Synthetic Workloads

The synthetic workload is essential and critical to the performance evaluation of database systems. When evaluating the database performance for a specific application, the similarity between synthetic workload and real application workload determines the credibility of evaluation results. However, the workload currently used for performance evaluation is difficult to have the same workload characteristics as the target application, which leads to inaccurate evaluation results. To address this problem, we propose a workload duplicator (Lauca) that can generate synthetic workloads with highly similar performance metrics for specific applications. To the best of our knowledge, Lauca is the first application-oriented transactional workload generator. By carefully studying the application-oriented synthetic workload generation problem, we present the key workload characteristics (transaction logic and data access distribution) of online transaction processing (OLTP) applications, and propose novel workload characterization and generation algorithms, which guarantee the high fidelity of synthetic workloads. We conduct extensive experiments using workloads from TPC-C, SmallBank and micro benchmarks on both MySQL and PostgreSQL databases, and experimental results show that Lauca consistently generates high-quality synthetic workloads.

cs.DB↗