Recently, Large Language Models (LLMs) have been demonstrated to possess impressive capabilities in a variety of domains and tasks. We investigate the issue of prompt design in the multi-turn text-to-SQL task and attempt to enhance the LLMs' reasoning capacity when generating SQL queries. In the conversational context, the current SQL query can be modified from the preceding SQL query with only a few operations due to the context dependency. We introduce our method called CoE-SQL which can prompt LLMs to generate the SQL query based on the previously generated SQL query with an edition chain. We also conduct extensive ablation studies to determine the optimal configuration of our approach. Our approach outperforms different in-context learning baselines stably and achieves state-of-the-art performances on two benchmarks SParC and CoSQL using LLMs, which is also competitive to the SOTA fine-tuned models.

Recently Large Language Models (LLMs) have been proven to have strong abilities in various domains and tasks. We study the problem of prompt designing in the text-to-SQL task and attempt to improve the LLMs' reasoning ability when generating SQL queries. Besides the trivial few-shot in-context learning setting, we design our chain-of-thought (CoT) prompt with a similar method to schema linking. We provide a method named ACT-SQL to automatically generate auto-CoT exemplars and thus the whole process doesn't need manual labeling. Our approach is cost-saving since we only use the LLMs' API call once when generating one SQL query. Furthermore, we extend our in-context learning method to the multi-turn text-to-SQL task. The experiment results show that the LLMs' performance can benefit from our ACT-SQL approach. Our approach achieves SOTA performance on the Spider dev set among existing in-context learning approaches.

Large language models (LLMs) have achieved impressive performance on code generation. However, for complex programming tasks, generating the correct solution in one go becomes challenging, thus some prior works have designed program repair approaches to improve code generation performance. In this work, we propose Self-Debugging, which teaches a large language model to debug its predicted program via few-shot demonstrations. In particular, we demonstrate that Self-Debugging can teach the large language model to perform rubber duck debugging; i.e., without any human feedback on the code correctness or error messages, the model is able to identify its mistakes by investigating the execution results and explaining the generated code in natural language. Self-Debugging achieves the state-of-the-art performance on several code generation benchmarks, including the Spider dataset for text-to-SQL generation, TransCoder for C++-to-Python translation, and MBPP for text-to-Python generation. On the Spider benchmark where there are no unit tests to verify the correctness of predictions, Self-Debugging with code explanation consistently improves the baseline by 2-3%, and improves the prediction accuracy on problems of the hardest level by 9%. On TransCoder and MBPP where unit tests are available, Self-Debugging improves the baseline accuracy by up to 12%. Meanwhile, by leveraging feedback messages and reusing failed predictions, Self-Debugging notably improves sample efficiency, and can match or outperform baseline models that generate more than 10x candidate programs.

For programmers, algorithms and data structures are their most essential subjects--a programmer's bread and butter if you will. If you want to enter the field of programming and hit the ground running, you'll need to master the most common data structures and boost your resume with in-demand skills. Here, we'll explore the eight most important data structures every programmer should know, including what they do and where to use them. To start, let's gain a fundamental understanding of what a data structure is. Data structures are methods of storing and organizing data in a computer system so that operations can be performed upon them more efficiently.

This paper discusses techniques for merging data files where no key field exists between the files. The paper will illustrate an approach to resolve two issues that are common to most fuzzy matching techniques: 1) how to weight proxy identifier fields, and 2) how to measure the Type One and Type Two errors of the merge estimation algorithm. A common requirement in analytics is to merge records in two or more large sets of information (i.e., thousands if not millions of records) where no exact key exists to match records between the information sets. When no exact key between the two data sets exists, a common merging solution is to use "fuzzy" matching. "Fuzzy" matching uses proxy keys as substitute keys to match records between the two data files.

Since Saturday, a massive trove of Facebook data has circulated publicly, splashing information from roughly 533 million Facebook users across the internet. The data includes things like profile names, Facebook ID numbers, email addresses, and phone numbers. It's all the kind of information that may already have been leaked or scraped from some other source, but it's yet another resource that links all that data together--and ties it to each victim--presenting tidy profiles to scammers, phishers, and spammers on a silver platter. Facebook's initial response was simply that the data was previously reported on in 2019 and that the company patched the underlying vulnerability in August of that year. But a closer look at where, exactly, this data comes from produces a much murkier picture.

Drones could someday have a sort of invisible license plate that allows local authorities to determine who the unmanned aerial system (UAS) belongs too. Pitched by Chinese drone manufacturer DJI, the concept for an electronic identification system for small drones is just one of many ideas as the Federal Aviation Administration looks into potential ways of identifying drone users. DJI suggests drones should use the radio equipment already on board most systems to transmit a unique registration number. That number would identify the drone owner to law enforcement in the event of a complaint or flight through a restricted area. Areas with restricted drone flight, such as airports, could use radio equipment to read that number and report the ID number to the authorities.

Determining the position of a robot in an environment, termed localization, is one of the challenges facing roboticist. Localization is essential to solving more complex problems such as locomotion, path planning and environmental learning. Our lab is developing a multi-agent system to use multiple small robots to accomplish tasks normally completed by larger robots. However, because of the reduced size of these robots, methods previously used to determine the position of the robot, such as GPS, cannot be employed. The problem we are facing is that we need to be able to determine the position of each of the robots in this multi-agent system simultaneously. We have developed a system to help track and identify robots using an overhead camera and LEDs, mounted on the robots, to efficiently solve the localization problem.