Verifix: Verified Repair of Programming Assignments

Automated feedback generation for introductory programming assignments is useful education. Most works try to generate correct a student program by comparing its behavior with an instructor’s reference on selected tests. In this work, our aim verifiably repairs as feedback. A student-submitted aligned and composed solution in terms of control flow, the variables two programs are automatically via predicates describing relationship between variables. When verification attempt obtained fails, we turn problem into MaxSMT whose leads minimal repair. We have conducted experiments curated from widely deployed intelligent tutoring system. Our results show that generating verified repair without sacrificing overall rate possible. fact, implementation, Verifix, shown outperform Clara, state-of-the-art tool, rate. This shows promise using high confidence pedagogy settings.