Enhancing Web Accessibility: Automated Detection of Issues with Generative AI
Websites are integral to people’s daily lives, with billions in use today. However, due to limited awareness of accessibility and its guidelines, developers often release web apps that are inaccessible to people with disabilities, who make up around 16% of the global population. To ensure a baseline of accessibility, software engineers rely on automated checkers that assess a webpage’s compliance based on predefined rules. Unfortunately, these tools typically cover only a small subset of accessibility guidelines and often overlook violations that require a semantic understanding of the webpage. The advent of generative AI, known for its ability to comprehend textual and visual content, has created new possibilities for detecting accessibility violations. We began by studying the most widely used guideline, WCAG, to determine the testable success criteria that generative AI could address. This led to the development of an automated tool called \name, which extracts elements from a page related to each success criterion and inputs them into an LLM prompted to detect accessibility issues on the web. Evaluations of \name showed its effectiveness, with a precision of 95.2% and a recall of 87.69%. Additionally, when tested on real websites, \name identified an average of 8 more types of accessibility violations than the combination of existing tools.
Mon 23 JunDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
10:30 - 12:20 | Bug DetectionResearch Papers / Industry Papers / Demonstrations / Journal First at Aurora B Chair(s): Lingming Zhang University of Illinois at Urbana-Champaign | ||
10:30 20mTalk | Yuga: Automatically Detecting Lifetime Annotation Bugs in the Rust Language Journal First Vikram Nitin Columbia University, Anne Mulhern Red Hat Inc, Sanjay Arora Red Hat Inc, Baishakhi Ray Columbia University | ||
10:50 10mTalk | SpecChecker-Int: An Extensible Concurrency Bugs Detection Tool for Interrupt-driven Embedded Software Demonstrations Boxiang Wang Beijing Sunwise Information Technology Ltd, Chao Li Beijing Institute of Control Engineering; Beijing Sunwise Information Technology, Rui Chen Beijing Institute of Control Engineering; Beijing Sunwise Information Technology, Sheng Wang Beijing Sunwise Information Technology Ltd, Chunpeng Jia Beijing Sunwise Information Technology Ltd, Mengfei Yang China Academy of Space Technology | ||
11:00 20mTalk | dl²: Detecting Communication Deadlocks in Deep Learning Jobs Industry Papers Yanjie Gao Microsoft Research, Jiyu Luo University of Science and Technology of China, Haoxiang Lin Microsoft Research, Hongyu Zhang Chongqing University, Ming Wu Zero Gravity Labs, Mao Yang Microsoft Research DOI Pre-print | ||
11:20 20mTalk | Detecting Metadata-Related Bugs in Enterprise Applications Research Papers Md Mahir Asef Kabir Virginia Tech, Xiaoyin Wang University of Texas at San Antonio, Na Meng Virginia Tech DOI | ||
11:40 20mTalk | ROSCallBaX: Statically Detecting Inconsistencies In Callback Function Setup of Robotic Systems Research Papers Sayali Kate Purdue University, Yifei Gao Purdue University, Shiwei Feng Purdue University, Xiangyu Zhang Purdue University DOI | ||
12:00 20mTalk | Enhancing Web Accessibility: Automated Detection of Issues with Generative AI Research Papers Ziyao He University of California, Irvine, Syed Fatiul Huq University of California, Irvine, Sam Malek University of California at Irvine DOI |
Aurora B is the second room in the Aurora wing.
When facing the main Cosmos Hall, access to the Aurora wing is on the right, close to the side entrance of the hotel.