The Capability Hardware Enhanced RISC Instructions (CHERI) architecture provides fine-grained memory protection for systems, but introduces additional hardware overheads that may negatively impact performance. Evaluating and optimizing such secure architectures require benchmarks that explicitly exercise their security mechanisms. Despite the abundance of benchmarks for unhardened systems, security-aware benchmarks for CHERI-based architectures remain scarce. We address this gap by proposing a framework for generating security-aware benchmarks for CHERI-based RISC-V systems, leveraging the TestRIG tool and applying CHERI-specific post-processing to ensure valid capability usage. As a demonstration use case, we apply the generated benchmarks to evaluate an In-Memory Computing (IMC)-based acceleration of the CHERI tagged memory. Our results show best-case speedups between 6% and 11%, while also identifying scenarios in which the acceleration proves no performance benefit.

Pretraining data selection has the potential to improve language model pretraining efficiency by utilizing higher-quality data from massive web data corpora. Current data selection methods, which rely on either hand-crafted rules or larger reference models, are conducted statically and do not capture the evolving data preferences during pretraining. In this paper, we introduce *model-aware data selection with data influence models (MATES)*, where a data influence model continuously adapts to the evolving data preferences of the pretraining model and then selects the data most effective for the current pretraining progress. Specifically, we collect oracle data influence by locally probing the pretraining model and fine-tune a small data influence model to approximate it accurately. The data influence model then predicts data influence over the whole pretraining corpus and selects the most influential data for the next pretraining stage. Experiments of pretraining 410M and 1B models on the C4 dataset demonstrate that MATES significantly outperforms random data selection on extensive downstream tasks. It doubles the gains achieved by the state-of-the-art data selection approach that leverages larger reference models and reduces the total FLOPs required to reach certain performances by half. Further analyses validate the effectiveness of the locally probed oracle data influence and the approximation with data influence models. Our code is open-sourced at https://github.com/cxcscmu/MATES.

In this paper, we study the probabilistic stability analysis of a subclass of stochastic hybrid systems, called the Planar Probabilistic Piecewise Constant Derivative Systems (Planar PPCD), where the continuous dynamics is deterministic, constant rate and planar, the discrete switching between the modes is probabilistic and happens at boundary of the invariant regions, and the continuous states are not reset during switching. These aptly model piecewise linear behaviors of planar robots. Our main result is an exact algorithm for deciding absolute and almost sure stability of Planar PPCD under some mild assumptions on mutual reachability between the states and the presence of non-zero probability self-loops. Our main idea is to reduce the stability problems on planar PPCD into corresponding problems on Discrete-time Markov Chains with edge weights. Our experimental results on PPCD benchmarks with various dynamics and number of modes, demonstrate the practical feasibility of this approach.