Compiler fuzzing tools such as Csmith have uncovered many bugs in compilers by randomly sampling programs from a generative model. The success of these tools is often attributed to their ability to generate unexpected corner case inputs that developers tend to overlook during manual testing. At the same time, their chaotic nature makes fuzzer-generated test cases notoriously hard to interpret, which has lead to the creation of input simplification tools such as C-Reduce (for C compiler bugs). In until now unrelated work, researchers have also shown that human-written software tends to be rather repetitive and predictable to language models. Studies show that developers deliberately write more predictable code, whereas code with bugs is relatively unpredictable. In this study, we ask the natural questions of whether this high predictability property of code also, and perhaps counter-intuitively, applies to fuzzer-generated code. That is, we investigate whether fuzzer-generated compiler inputs are deemed unpredictable by a language model built on human-written code and surprisingly conclude that it is not. To the contrary, Csmith fuzzer-generated programs are more predictable on a per-token basis than human-written C programs. Furthermore, bug-triggering tended to be more predictable still than random inputs, and the C-Reduce minimization tool did not substantially increase this predictability. Rather, we find that bug-triggering inputs are unpredictable relative to Csmith's own generative model. This is encouraging; our results suggest promising research directions on incorporating predictability metrics in the fuzzing and reduction tools themselves.

The feature interaction problem arises when two or more independent features interact with each other in an undesirable manner. Feature interactions remain a challenging and important problem in emerging domains of cyber-physical systems (CPS), such as intelligent vehicles, unmanned aerial vehicles (UAVs) and the Internet of Things (IoT), where the outcome of an unexpected interaction may result in a safety failure. Existing approaches to resolving feature interactions rely on priority lists or fixed strategies, but may not be effective in scenarios where none of the competing feature actions are satisfactory with respect to system requirements. This paper proposes a novel synthesis-based approach to resolution, where a conflict among features is resolved by synthesizing an action that best satisfies the specification of desirable system behaviors in the given environmental context. Unlike existing resolution methods, our approach is capable of producing a desirable system outcome even when none of the conflicting actions are satisfactory. The effectiveness of the proposed approach is demonstrated using a case study involving interactions among safety-critical features in an autonomous drone.

We introduce deep convolutional and recurrent neural networks for end-to-end, open-vocabulary text reading on historical maps. A text detection network predicts word bounding boxes at arbitrary orientations and scales. The detected word images are then normalized for a robust recognition network. Because accurate recognition requires large volumes of training data but manually labeled data is relatively scarce, we introduce a dynamic map text synthesizer providing a practically infinite stream of training data. Results are evaluated on a labeled data set of 30 maps featuring over 30,000 text labels.