12/17/2021 Aaron Seidlitz, Illinois CS 9 min read
Topics receiving recognition include Sarita Adve’s ILLIXR, recent alumna Güliz Seray Tuncay’s work in security, and much more.
Written by Aaron Seidlitz, Illinois CS
As Illinois CS research continues to sustain high levels of success, one indicator is that the papers produced here earn best paper awards at some of the most prominent computing conferences.
From the faculty guiding these projects to the students learning what it takes to develop well-regarded research, these honors are representative of the effort and dedication given to research at Illinois CS.
Congratulations to each of the following projects and the authors involved.
Best Paper Award, IEEE International Symposium on Workload Characterizations (IISWC 2021)
“A labor of love” over the last several years has led Illinois CS professor Sarita Adve, and a student research team devoted to the Illinois Extended Reality (ILLIXR) testbed and consortium, to a Best Paper Award at the IEEE International Symposium on Workload Characterization (IISWC 2021).
This paper, entitled “ILLIXR: Enabling End-to-End Extended Reality Research,” provided the initial description of the entity’s purpose. It does so, as Adve explained, by providing “the first published power/performance/quality characterization of an end-to-end extended reality system.” Additionally, it “provides a blueprint for the research my group will undertake over the next several years.”
As this team built the end-to-end system and now works to create the testbed and consortium that will become the bedrock of open-source XR research, this best paper award acknowledges the sound basis of their concept.
It also allows the group to extend the reach of their interactions.
“This best paper award is a tremendous acknowledgement, and it definitely helps as we advertise it to others who we could collaborate with through ILLIXR,” Adve said. “It’s also a direct result of the dedicated work our 13 student authors have exhibited throughout this process. We have also consulted with numerous colleagues in the industry and academia.
“It’s been an amazing collaboration; perhaps the hardest work I have done in my entire career and the most fun.”
Adve and her PhD student Huzaifa Muhammed began this process about five years ago. As they came to the realization that this was an exciting area of development across an interdisciplinary scope, eager undergraduate and graduate students filled out the team.
“These were incredibly talented students who worked well with each other and formed a great team,” Adve said. “This paper represents a lot of hard work by a lot of people for over three years.”
Best Paper Award, ACM International Conference on Systems for Energy-Efficient Buildings, Cities, and Transportation (BuildSys 2021)
Illinois CS PhD student Ashish Kashinath and his advisor, Illinois CS adjunct professor Sibin Mohan, paired together on a research paper that has roots from Kashinath’s internship with Microsoft Research, India.
It is also derived from the student’s research focus in performance, predictability and resilience challenges when commodity components are used in Cyber-Physical Systems and Networks – for instance in power grid, avionics and automotive systems.
Their paper, entitled “PIRMedic: A Physics-based Fault Diagnosis for Passive Infra-Red (PIR) Sensors,” recently earned the Best Paper Award at ACM’s BuildSys 2021.
It is built off a focus in detecting failures in PIR sensors, which can be used in modern buildings for such things as detecting occupancy in rooms (turning lights or the HVAC on or off) or in washrooms (faucets or sanitizer and towel dispensers).
“Failures of the sensors is common due to excessive use and environmental factors. Such failures can affect daily function or even result in serious problems,” Mohan said. “ The impact in this work comes from the simple, yet elegant design of our system — which uses something that is fundamental to the operation of the sensors to detect failures — and the physics of it – namely the behavior of voltages and currents at the sensor hardware during the sensing process.
“Coupled with the analysis framework developed by Ashish, this can potentially help detect failures in millions of such installations — leading to significant cost and labor savings.”
For nearly two years, this project went from formulation through experimental evaluation.
Mohan said that the project succeeded through Kashinath’s perseverance, combined with assistance from leadership at the Coordinated Science Laboratory. Additionally, the Starbucks location on campus at The Union played a big role, as the group deployed hardware and collected its data there. Finally, Illinois Facilities & Services helped provide valuable information and support.
“I think the best paper award recognizes both the novelty of this work and Ashish’s efforts in making sure the project was successful,” Mohan said. “I’m hoping that this work gets wide recognition as a result, as it rightly deserves. It has potential for impact beyond the academic circles and in real world scenarios — possibly another reason why it won the award.”
Best Student Paper Runner-Up, IEEE International Conference on Data Mining (ICDM 2021)
As a visiting scholar with Illinois CS professor Jiawei Han, professor Di Jin of Tianjin University began research on text-rich information networks through collaboration with Han’s group – thus generating a joint paper on GCN with text-rich networks presented at the Web Search and Data Mining Conference 2021.
Jin then introduced this paper’s first author, a student of his named Zhizhi Yu, to Illinois CS professor Hanghang Tong.
Together, this team then worked for about nine months on a paper entitled “AS-GCN: Adaptive Semantic Architecture of Graph Convolutional Networks for Text-Rich Networks,” which Tong believes demonstrates the power of multidisciplinary research. All seven researchers included on the project have diverse backgrounds and come from four different institutes.
The resulting honor from ICDM, a top data mining conference in IEEE, represents all areas of data mining – including algorithms, software, systems and applications.
“Our work develops an end-to-end deep learning architecture that unifies neural topic modes – a powerful technique in text mining – and graph convolutional networks – which is a popular tool in deep graph learning – to enhance text-rich network representation,” Tong said.
Outstanding Doctoral Dissertation Award Runner-Up, ACM Special Interest Group on Security, Audit and Control (SIGSAC 2021)
Recent alumna Güliz Seray Tuncay (PhD CS ’19) became thoroughly interested in computer security during the second year of her PhD studies. At that point, she began finding security issues and vulnerabilities in “pretty much anything I touched, be it a communication protocol, an Android API, or a new system introduced in a paper I just started reading.”
After having presented her work on the security issues of mobile browsers at ACM Computer and Communications Security (CCS), she also identified numerous serious security vulnerabilities in some of Android's key security mechanisms. Her work on this topic received a Distinguished Paper Award from the Network and Distributed System Security Symposium (NDSS). She presented her follow-up work at the USENIX Security Symposium. Her discoveries were recognized and rewarded by Google and her work led to changes in the official releases of Android, the most popular mobile platform as of now.
Despite this success, Tuncay emphasized that the dissertation process is still a grueling one. Her work with PhD advisor Carl Gunter helped her through the process, and she couldn’t help but feel extremely fulfilled upon receiving ACM SIGSAC 2021 runner-up award for Outstanding Doctoral Dissertation – as it represents the research community’s overwhelming belief in the value of this work.
The title of her dissertation was, “Practical Least Privilege for Cross-Origin Interactions on Mobile Operating Systems.”
“Many of the products we use today, be it a phone, or a wearable device, or any other complex system really, are unfortunately not always designed with a security-first mindset,” Tuncay said. “My work disclosed serious security vulnerabilities that I discovered in critical parts of mobile operating systems, more specifically in Android, and offered practical countermeasures that can tackle the identified issues effectively and efficiently.”
Best Paper Runner-Up, ACM Conference on Computer and Communications Security (CCS 2021)
Illinois CS PhD students Peiyao Sheng and Gerui Wang worked together for the past six months to further research a topic that became the focus of a recent paper that earned Best Paper Runner-Up at the ACM Conference on Computer and Communications Security (CCS 2021). It is titled “BFT Protocol Forensics.”
Both students based this work off other published work on the topic that they found to be narrow and worthy of further investigation.
“The most impactful development is our collaboration with the Diem team, formerly Facebook’s blockchain system,” Sheng said. “We apply our theoretical analysis to this real-world blockchain system and implement a lightweight module on it. The proposal received active feedback for deployment in Diem.”
“When we explored the relationship of our definition of forensics property and several existing BFT protocols, our conclusion was that we can equip those protocols through our property with little difficulty,” Wang said. “This means that simple and understandable changes to existing protocols can have this additional yet useful property – forensics.”
Working with three others through online meetings that often included advisors, the group’s Best Paper Runner-Up Award served as proof that the collaboration worked.
“CCS is a top conference in the area of security, and it is one of the few most prominent conferences for my research area,” Wang said. “We thought the paper fit well with this conference, and to receive this award is an encouraging result.”
“I’m glad to see other people felt interested in our work. It’s a strong encouragement for me to explore my interest in research and share it with others,” Sheng said.
Best Paper Runner-Up, ACM Conference on Computer and Communications Security (CCS 2021)
For more than a year, Illinois CS professor Ling Ren and PhD students Sourav Das and Zhoulun Xiang have worked on a topic that has led to numerous paths.
Initial work included a distributed randomness beacon, to an algorithm that efficiently disseminates data in a distributed network of nodes, and, finally, focusing on the problem of data dissemination.
This led to a recent paper entitled, “Asynchronous Data Dissemination and its Applications,” which earned Best Paper runner-up at CCS 2021.
“The biggest impact of the project is that the new algorithm solves a fundamental problem that has been overlooked. It has many applications in cryptography and distributed computing,” Ren said. “In our CCS paper, we already used our new algorithm to improve a variety of important algorithms, such as reliable broadcast and verifiable secret sharing.
“We have since identified a few other research directions that can benefit from this work, and we are currently working through those topics.”
That CCS is the flagship conference on computer security, seems a fitting result to Ren.
“I am very impressed with the perseverance of the students. We had some decent designs of the randomness beacon project but they kept pushing for a better and more elegant design, which seemed really difficult at the time. This effort eventually led us to ask the right questions and invent the new algorithm,” Ren said.