Deployable structures, essential across various engineering applications ranging from umbrellas to satellites, are evolving to include soft, morphable designs where geometry drives transformation. However, a major challenge for soft materials lies in achieving reliable actuation and stable shape retention in their deployed state. Drawing inspiration from biological growth processes, we demonstrate that irreversible plastic deformations can be leveraged to create cellular metamaterials with permanent morphing capabilities. By employing a simple actuation method, stretching and releasing the structure's ends, our approach facilitates the design of structures capable of sequential, multi-target configurations and mechanical multistability. Our methodology augments additive manufacturing to transform flat, printable designs into intricate 3D forms, with broad applications in consumer goods, healthcare, and architecture.

The problem of imaging materials with circular polarization properties is discussed within the framework of vectorial ptychography. We demonstrate, both theoretically and numerically, that using linear polarizations to investigate such materials compromises the unicity of the solution provided by this computational method. To overcome this limitation, an improved measurement approach is proposed, which involves specific combinations of elliptical polarizations. The effectiveness of this strategy is demonstrated by numerical simulations and experimental measurements on cholesteric liquid crystals films, which possess unique polarization properties. With the help of Pauli matrices algebra, our results highlight the technique's ability to discern between different types of circular polarizers, uniform vs. non-uniform, and determine their handedness.

The stability of an exponential current in water to infinitesimal perturbations in the presence of gravity and capillarity is investigated. Some new results on the generation of gravity-capillary waves are presented which supplement the previous works of Morland, Saffman \& Yuen (1991) and Young \& Wolfe (2014), namely in finite depth. To consider perturbations of much larger scales, a specific attention is paid to the stability of the exponential current only in the presence of gravity.

Chain-of-thought (CoT) reasoning not only enhances large language model performance but also provides critical insights into decision-making processes, marking it as a useful tool for monitoring model intent and planning. By proactively preventing models from acting on CoT indicating misaligned or harmful intent, CoT monitoring can be used to reduce risks associated with deploying models. However, developers may be incentivized to train away the appearance of harmful intent from CoT traces, by either customer preferences or regulatory requirements. Recent works have shown that banning mention of a specific example of reward hacking, which may be done either to make CoT presentable to users or as a naive attempt to prevent the behavior, causes obfuscation of the undesired reasoning traces but the persistence of the undesired behavior. Such obfuscation threatens the reliability of CoT monitoring. However, obfuscation of reasoning can be due to its internalization to latent space computation, or its encoding within the CoT. Here, we provide an extension to these results. First, we show that penalizing the use of specific strings within load-bearing reasoning traces causes models to substitute alternative strings. Crucially, this does not alter the underlying method by which the model performs the task, demonstrating that the model can learn to steganographically encode its reasoning. We further demonstrate that models can generalize an encoding scheme. When the penalized strings belong to an overarching class, the model learns not only to substitute strings seen in training, but also develops a general encoding scheme for all members of the class which it can apply to held-out testing strings.

We show that the distance between two qubits coupled to a structured reservoir acts as a single geometric control that can store, revive, or suppress Bell nonlocality. In a mirror-terminated guide, quantum correlations lost to the bath return at discrete recurrence times, turning a product state into a Bell-violating one without any entangling drive (only local basis rotations/readout). In the continuum limit, we derive closed-form criteria for the emergence of nonlocality from backflow, and introduce a Bell-based analogue of the BLP measure to quantify this effect. We also show how subwavelength displacements away from a decoherence-free node quadratically reduce the lifetime of a dark state or bright state, enabling highly sensitive interferometric detection. All results rely on analytically solvable models and are compatible with current superconducting and nanophotonic platforms, offering a practical route to passive, geometry-controlled non-Markovian devices.

The aggregation process of fibers suspended in a fluid remains an open question, despite its critical role in numerous natural and industrial processes. This paper presents an experimental setup designed to create fiber aggregates in a turbulent flow. The phase diagram for the aggregation states is established as a function of the experimental parameters, and the statistical properties of the resulting aggregates are characterized. Additionally, the fragmentation process is investigated, and a model is proposed to describe the fragmentation time of the aggregates, which is subsequently validated through comparison with experimental results.

Jet fragmentation is investigated through a Direct Numerical Simulation campaign using Basilisk (Popinet & collaborators 2013). The simulations span over one order of magnitude of gaseous Weber numbers (13 to 165), i.e. over the second wind-induced and atomization regimes, and the jets develop over distances up to 28 nozzle diameters. The study focuses on the size and velocity distributions of droplets, as well as their joint distribution. Two models derived from different theoretical backgrounds, the statistical description of the turbulence intermittency (Novikov & Dommermuth 1997) and the empirical description of the ligament-mediated fragmentation (Villermaux et al. 2004), are compared for describing the droplet size distribution close to the nozzle. The characteristics of the size-velocity joint distribution are explained using the vortex ring theory (Saffman 1992) which highlights two sources of fragmentation. Finally, the joint histogram of the particulate Reynolds and Ohnesorge numbers is analysed and a normalisation is suggested. It reveals that the delimitations of the droplet phase space, once properly normalised, are self-similar and independent of the gaseous Weber number, both numerically and experimentally.