Want to build an accurate method-level bug prediction model? You will need a noise-free dataset. In our just-accepted #MSR2026 paper, we showed the effectiveness of LLMs in providing such a dataset. Preprint: https://t.co/VWWhRsnUQm
How exactly do the Large Language Model-based systems attempt to improve software energy consumption? How do they do it for other performance indicators? My students wrote 2 short papers.
The energy paper:
https://t.co/RnX0gKMV6g
The performance paper: https://t.co/HI0lvOxycG
What does the SATD in the test code look like? Can we detect them with the existing tools? Can we detect them with LLMs? To answer these, we have manually annotated 50,000 test code comments from 1,000 open-source Java projects. Read our preprint here: https://t.co/MnQnOObVyL
We compared all method history tracking tools against a systematically developed oracle. We also introduce a new tool, and present a UI-based application that allows comparing all tools for any method.
Very excited that my MSc. student Shahidul's paper got accepted by @ICSEconf 2026.
HistoryFinder: Advancing Method-Level Source Code History Generation with Accurate Oracles and Enhanced Algorithm.
Preprint: https://t.co/PCPY8M0Wva
"Understanding the Issue Types in Open Source Blockchain-based Software Projects with the Transformer-based BERTopic", got accepted at CASCON 2025.
Preprint: https://t.co/GaISmzzbyA
Starting in September 2026, I will be recruiting two fully funded PhD students to conduct research in software quality and maintenance. For more information, please check here: https://t.co/kKKtKJ9u4B
If there is no evidence to support that Self-admitted Tech debt harms software quality and maintainability, should we stop caring about SATD? I will be presenting our #MSR2025 paper on this topic today at 2:30, room# 215.
A Bangladeshi Doctor has been serving in Gaza for the last one month, without caring about his life. May Allah protect this beautiful soul..
#GazaGenocide#Gaza
Want to know how StackOverflow has been used in SE research? Read our just accepted @JSSoftware paper "A Systematic Mapping Study of Crowd Knowledge Enhanced Software Engineering Research Using Stack Overflow". Phenomenal effort from @TanzilMH . Hats off to such dedication!
@guy_2593606@Qariablatias@lilyjayofficial It's not about what God can or can't do. It's about what God does and doesn't. For example, God can be unjust, but he is not. It doesn't fit him.
The negative maintenance impact of SATD has not been accurately established, with previous studies yielding contradictory findings. Our just accepted paper at @msrconf delves deeply into this issue to provide clarity and new insights.
preprint: https://t.co/ztYrd6okqv