To help evaluate and understand the latent capabilities of language models, this paper introduces an approach using optimized input embeddings, or 'soft prompts,' as a metric of conditional distance between a model and a target behavior. The technique aims to facilitate latent capability discovery as a part of automated red teaming/evaluation suites and to provide quantitative feedback about the accessibility of potentially concerning behaviors in a way that may scale to powerful future models, including those which may otherwise be capable of deceptive alignment. An evaluation framework using soft prompts is demonstrated in natural language, chess, and pathfinding, and the technique is extended with generalized conditional soft prompts to aid in constructing task evaluations.

Modeling structure and behavior of software systems plays a crucial role in the industrial practice of software engineering. As with other software engineering artifacts, software models are subject to evolution. Supporting modelers in evolving software models with recommendations for model completions is still an open problem, though. In this paper, we explore the potential of large language models for this task. In particular, we propose an approach, retrieval-augmented generation, leveraging large language models, model histories, and retrieval-augmented generation for model completion. Through experiments on three datasets, including an industrial application, one public open-source community dataset, and one controlled collection of simulated model repositories, we evaluate the potential of large language models for model completion with retrieval-augmented generation. We found that large language models are indeed a promising technology for supporting software model evolution (62.30% semantically correct completions on real-world industrial data and up to 86.19% type-correct completions). The general inference capabilities of large language models are particularly useful when dealing with concepts for which there are few, noisy, or no examples at all.

Reinforcement learning with human feedback (RLHF) has become the dominant method to align large models to user preferences. Unlike fine-tuning, for which there are many studies regarding training data memorization, it is not clear how memorization is affected by or introduced in the RLHF alignment process. Understanding this relationship is important as real user data may be collected and used to align large models; if user data is memorized during RLHF and later regurgitated, this could raise privacy concerns. In this work, we analyze how training data memorization can surface and propagate through each phase of RLHF. We focus our study on code completion models, as code completion is one of the most popular use cases for large language models. We find that RLHF significantly decreases the chance that data used for reward modeling and reinforcement learning is memorized, in comparison to aligning via directly fine-tuning on this data, but that examples already memorized during the fine-tuning stage of RLHF, will, in the majority of cases, remain memorized after RLHF.

Modeling structure and behavior of software systems plays a crucial role, in various areas of software engineering. As with other software engineering artifacts, software models are subject to evolution. Supporting modelers in evolving models by model completion facilities and providing high-level edit operations such as frequently occurring editing patterns is still an open problem. Recently, large language models (i.e., generative neural networks) have garnered significant attention in various research areas, including software engineering. In this paper, we explore the potential of large language models in supporting the evolution of software models in software engineering. We propose an approach that utilizes large language models for model completion and discovering editing patterns in model histories of software systems. Through controlled experiments using simulated model repositories, we conduct an evaluation of the potential of large language models for these two tasks. We have found that large language models are indeed a promising technology for supporting software model evolution, and that it is worth investigating further in the area of software model evolution.

In the context of verification of data-aware processes (DAPs), a formal approach based on satisfiability modulo theories (SMT) has been considered to verify parameterised safety properties of so-called artifact-centric systems. This approach requires a combination of model-theoretic notions and algorithmic techniques based on backward reachability. We introduce here a variant of one of the most investigated models in this spectrum, namely simple artifact systems (SASs), where, instead of managing a database, we operate over a description logic (DL) ontology expressed in (a slight extension of) RDFS. This DL, enjoying suitable model-theoretic properties, allows us to define DL-based SASs to which backward reachability can still be applied, leading to decidability in PSPACE of the corresponding safety problems.