BEHAVE: amBiguous and Equivocal beHAviors Verification Engine
This program is tentative and subject to change.
Software transparency requires understanding not only which components a software system includes, but also which behaviors it may exhibit during execution. Component-level information can reveal what a software product is made of, but it does not explain how it acts once executed, nor whether it performs behaviors that may affect privacy, security, or user trust. Recent work introduced Equivocal Software Behaviors (ESBs), i.e., behaviors that are not necessarily malicious but may raise uncertainty about software goals, privacy, or security when left unexplained. In this short paper, we present BEHAVE, the amBiguous and Equivocal beHAviors Verification Engine, a tool designed to make ESB analysis actionable for software assessment. BEHAVE automates the collection, integration, and summarization of equivocal behavioral evidence from heterogeneous information sources, producing concise ESB profiles for individual software samples and aggregate views for software collections. By making source-specific and shared evidence explicit, BEHAVE supports the comparison of equivocal behavioral observations across information sources, operating environments, and topic-specific custom requirements extending the standard ESB taxonomy. We exercised BEHAVE in a pilot analysis of 400 selected software samples from public datasets. The results show that BEHAVE can effectively support software inspection by exposing differences among information sources and by highlighting environment-specific ESB profiles.
This program is tentative and subject to change.
Fri 18 SepDisplayed time zone: Amsterdam, Berlin, Bern, Rome, Stockholm, Vienna change
11:00 - 12:30 | Session 20 - Humans in the Loop: Teams, Tools & TradeoffsVisions and Emerging Results Track / Tool Demonstration and Data Showcase Track / Journal First Track / Research Papers Track at A53S Chair(s): Gemma Catolino University of Salerno, Fiorella Zampetti University of Sannio, Italy Theme: Human Factors & Software Teams | ||
11:00 20mPaper | "It Comes in Notebooks": Changes and Challenges When Operationalizing ML Prototypes Research Papers Track Arumoy Shome Delft University of Technology, Luís Cruz TU Delft, Diomidis Spinellis AUEB & TU Delft, Arie van Deursen TU Delft | ||
11:20 20mPaper | Smaller Models, Unexpected Costs: Trade-offs of LLM Quantization for Automated Program Repair Research Papers Track Fernando Vallecillos Ruiz Simula Research Laboratory, Giordano d'Aloisio University of L'Aquila, Max Hort Simula Research Laboratory, Luca Traini University of L'Aquila, Antinisca Di Marco University of L'Aquila, Leon Moonen Simula Research Laboratory Pre-print | ||
11:40 20mPaper | An Empirical Study of the Evolution of GitHub Actions Workflows Journal First Track Pooya Rostami Mazrae University of Mons, Alexandre Decan University of Mons; F.R.S.-FNRS, Tom Mens University of Mons, Mairieli Wessel Radboud University Link to publication DOI Pre-print | ||
12:00 10mShort-paper | BEHAVE: amBiguous and Equivocal beHAviors Verification Engine Tool Demonstration and Data Showcase Track Gregorio Dalia University of Sannio, Andrea Di Sorbo University of Sannio, Gerardo Canfora University of Sannio, Corrado Visaggio University of Foggia, Italy | ||
12:10 10mShort-paper | CMM-AISE: A Cognitive Maturity Model for AI-Assisted Software Engineering Visions and Emerging Results Track Daniel González Universidad de los Andes, Mario Linares-Vásquez Universidad de los Andes, Daniel Soto Tres Astronautas, Andrés Hurtado Blend | ||