Lev Sorokin

Lev Sorokin, Ivan Vasilev, Samuele Pasini

This report summarizes the results of the first edition of the Large Language Model (LLM) Testing competition, held as part of the DeepTest workshop at ICSE 2026. Four tools competed in benchmarking an LLM-based car manual information retrieval application, with the objective of identifying user inputs for which the system fails to appropriately mention warnings contained in the manual. The testing solutions were evaluated based on their effectiveness in exposing failures and the diversity of the discovered failure-revealing tests. We report on the experimental methodology, the competitors, and the results.

Philipp Habicht, Lev Sorokin, Abdullah Saydemir et al.

In-Car Conversational Question Answering (ConvQA) systems significantly enhance user experience by enabling seamless voice interactions. However, assessing their accuracy and reliability remains a challenge. This paper explores the use of Large Language Models (LLMs) alongside advanced prompting techniques and agent-based methods to evaluate the extent to which ConvQA system responses adhere to user utterances. The focus lies on contextual understanding and the ability to provide accurate venue recommendations considering user constraints and situational context. To evaluate utterance-response coherence using an LLM, we synthetically generate user utterances accompanied by correct and modified failure-containing system responses. We use input-output, chain-of-thought, self-consistency prompting, and multi-agent prompting techniques with 13 reasoning and non-reasoning LLMs of varying sizes and providers, including OpenAI, DeepSeek, Mistral AI, and Meta. We evaluate our approach on a case study involving restaurant recommendations. The most substantial improvements occur for small non-reasoning models when applying advanced prompting techniques, particularly multi-agent prompting. However, reasoning models consistently outperform non-reasoning models, with the best performance achieved using single-agent prompting with self-consistency. Notably, DeepSeek-R1 reaches an F1-score of 0.99 at a cost of 0.002 USD per request. Overall, the best balance between effectiveness and cost-time efficiency is reached with the non-reasoning model DeepSeek-V3. Our findings show that LLM-based evaluation offers a scalable and accurate alternative to traditional human evaluation for benchmarking contextual understanding in ConvQA systems.

Lev Sorokin, Damir Safin, Shiva Nejati

Search-based software testing (SBST) is a widely adopted technique for testing complex systems with large input spaces, such as Deep Learning-enabled (DL-enabled) systems. Many SBST techniques focus on Pareto-based optimization, where multiple objectives are optimized in parallel to reveal failures. However, it is important to ensure that identified failures are spread throughout the entire failure-inducing area of a search domain and not clustered in a sub-region. This ensures that identified failures are semantically diverse and reveal a wide range of underlying causes. In this paper, we present a theoretical argument explaining why testing based on Pareto optimization is inadequate for covering failure-inducing areas within a search domain. We support our argument with empirical results obtained by applying two widely used types of Pareto-based optimization techniques, namely NSGA-II (an evolutionary algorithm) and OMOPSO (a swarm-based Pareto-optimization algorithm), to two DL-enabled systems: an industrial Automated Valet Parking (AVP) system and a system for classifying handwritten digits. We measure the coverage of failure-revealing test inputs in the input space using a metric that we refer to as the Coverage Inverted Distance quality indicator. Our results show that NSGA-II-based search and OMOPSO are not more effective than a naïve random search baseline in covering test inputs that reveal failures. The replication package for this study is available in a GitHub repository.

Lev Sorokin, Radouane Bouchekir, Tewodros A. Beyene et al.

An assurance case has become an integral component for the certification of safety-critical systems. While manually defining assurance case patterns can be not avoided, system-specific instantiations of assurance case patterns are both costly and time-consuming. It becomes especially complex to maintain an assurance case for a system when the requirements of the System-Under-Assurance change, or an assurance claim becomes invalid due to, e.g., degradation of a systems component, as common when deploying learning-enabled components. In this paper, we report on our preliminary experience leveraging the tool integration framework Evidential Tool Bus (ETB) for the construction and continuous maintenance of an assurance case from a predefined assurance case pattern. Specifically, we demonstrate the assurance process on an industrial Automated Valet Parking system from the automotive domain. We present the formalization of the provided assurance case pattern in the ETB processable logical specification language of workflows. Our findings show that ETB is able to create and maintain evidence required for the construction of an assurance case.

Lev Sorokin, Matteo Biagiola, Andrea Stocco

Scenario-based testing with driving simulators is extensively used to identify failing conditions of automated driving assistance systems (ADAS). However, existing studies have shown that repeated test execution in the same as well as in distinct simulators can yield different outcomes, which can be attributed to sources of flakiness or different implementations of the physics. In this paper, we present MultiSim, a novel approach to multi-simulation ADAS testing based on a search-based testing approach that leverages an ensemble of simulators to identify failure-inducing, simulator-agnostic test scenarios. During the search, each scenario is evaluated jointly on multiple simulators. Scenarios that produce consistent results across simulators are prioritized for further exploration, while those that fail on only a subset of simulators are given less priority, as they may reflect simulator-specific issues rather than generalizable failures. Our empirical study, which involves testing three lane-keeping ADAS on different pairs of three widely used simulators, demonstrates that MultiSim outperforms single-simulator testing by achieving, on average, a higher rate of simulator-agnostic failures by 66%. Compared to a state-of-the-art multi-simulator approach that combines the outcome of independent test generation campaigns obtained in different simulators, MultiSim identifies, on average, up to 3.4X more simulator-agnostic failing tests and higher failure rates. To avoid the costly execution of test inputs on which simulators disagree, we propose to predict simulator disagreements and bypass test executions. Our results show that utilizing a surrogate model during the search retains the average number of valid failures and also improves efficiency. Our findings indicate that combining an ensemble of simulators is a promising approach for the automated cross-replication in ADAS testing.

Rafael Giebisch, Ken E. Friedl, Lev Sorokin et al.

In-car conversational systems bring the promise to improve the in-vehicle user experience. Modern conversational systems are based on Large Language Models (LLMs), which makes them prone to errors such as hallucinations, i.e., inaccurate, fictitious, and therefore factually incorrect information. In this paper, we present an LLM-based methodology for the automatic factual benchmarking of in-car conversational systems. We instantiate our methodology with five LLM-based methods, leveraging ensembling techniques and diverse personae to enhance agreement and minimize hallucinations. We use our methodology to evaluate CarExpert, an in-car retrieval-augmented conversational question answering system, with respect to the factual correctness to a vehicle's manual. We produced a novel dataset specifically created for the in-car domain, and tested our methodology against an expert evaluation. Our results show that the combination of GPT-4 with the Input Output Prompting achieves over 90 per cent factual correctness agreement rate with expert evaluations, other than being the most efficient approach yielding an average response time of 4.5s. Our findings suggest that LLM-based testing constitutes a viable approach for the validation of conversational systems regarding their factual correctness.