Three new species of small (~25–200 g) tribosphenic marsupials, including a new genus, are described from the Early Miocene deposits of the Riversleigh World Heritage Area. Phantasmodon travouilloni, Phantasmodon minuferox, and Keeunidae gen. and sp. indet. share dental apomorphies with the late Oligocene marsupials Keeuna woodburnei and Ankotarinja tirarensis (currently Australidelphia incertae sedis), and with Australia’s oldest and most plesiomorphic marsupial, the early Eocene Djarthia murgonensis. Parsimony analyses recover these taxa as a distinct clade, warranting recognition of a new marsupial order, Keeunamorphia. In contrast, Bayesian analyses do not support the inclusion of D. murgonensis in this group, instead placing it as a stem australidelphian or stem marsupial. When early Eocene South American and European metatherians are excluded, Keeunamorphia new order appears to be the most basal (parsimony analysis) or one of the most basal (Bayesian analysis) lineages of Australidelphia. However, when these taxa are included in parsimony analysis, Keeunamorphia weakly links to South American sternbergiids and European herpetotheriids, whereas Bayesian analysis suggests keeunamorphians (excluding D. murgonensis) are more closely related to dasyuromorphians than to peramelemorphians, microbiotherians, D. murgonensis, and non-australidelphian metatherians. Keeunamorphia n. ord. may represent some of Australia’s earliest marsupials, with possible origins in Gondwana and a record spanning more than 35 million years before its apparent extinction in the Middle Miocene.

UUID: http://zoobank.org/a2699457-5b2c-4f0c-b8b6-dd41401c3a17

This article explores patterns of agreement and word order in the Central Algonquian language Border Lakes Ojibwe. This variety of Ojibwe shows alternations between VOS and VSO word orders and complex interactions between probes on Voice, Infl, and C. I show that the behaviour of lower probes feeds and bleeds the possible agreement and movement relations on each subsequent probe. These complex interactions culminate with the peripheral agreement marker on C, which shows a curious pattern that I dub reverse omnivority, where the probe agrees with lower ranked arguments over higher ranked arguments regardless of whether it is a logical subject or object. The core theoretical innovation is an extension of the syntactic operation agree to encode a relativized EPP, which captures variation and restrictions on movement and the possibility of movement and feature copying being independent. The account provides a strong case for Ojibwe as a configurational language, and is shown to capture variation in agreement and word order in various corners of Ojibwe and beyond.

Marc Bloch devient historien à l’époque où se constituent les sciences sociales. On s’intéresse ici à ses années de formation et aux influences intellectuelles – Karl Marx, Numa Denis Fustel de Coulanges, Émile Durkheim – qui ont façonné son imaginaire sociologique (marqué par les débats des années 1860-1910) et une pratique de l’histoire orientée par la quête de la « structure sociale ». Le médiéviste paraît l’inventeur de cette notion qui prend une place croissante dans sa recherche et lui permet d’explorer à la fois les relations entre les phénomènes sociaux et l’articulation entre réalités matérielles et système de représentations. Marc Bloch tente ainsi d’élaborer une approche « structurale » ou « structurelle » de la société féodale, qui semble dialoguer davantage avec les réflexions de Marcel Mauss et de Maurice Halbwachs qu’avec celles de Lucien Febvre. Pourtant, ce moment structural de l’historiographie française fut largement relégué dans l’ombre après la mort tragique de l’historien.

In a non-generative approach to artificial intelligence in an artistic practice, this work looks at mapping processing data from artificial neural networks (NNs) onto sound and visuals. One aim of this practice-based piece of research is to paradoxically offer insights into how these ubiquitous, yet notoriously opaque algorithms operate, by exposing the audience to the intrinsic unintelligibility of their processes. The other is to use these vast amounts of abstract data as a creative starting point for audiovisual artworks, referring to aesthetic traditions that have emerged from the need to make use and potentially make sense of such extensive masses of information, and from ones that have developed sounds that have gradually become associated with digital and post-digital worlds and other exterior and abstract concepts. At the heart of the whole work is a link and cross-fertilisation between the use of sounds and visuals aesthetically associated with errors and digital malfunction and the use of actual ‘waste’ data (from NN training), which acts as a trace of their operation.

We show that the existence of non-zero tropical forms of degree at least two implies that the tropical Chow group of points of an integral affine manifold is infinite-dimensional. This can be seen as a tropical analogue of classical results of Mumford and Roitman for Chow groups of smooth (complex) projective algebraic varieties. We also show that the existence of tropical 1-forms on integral affine surfaces does not imply infinite dimensionality by considering the case of a tropical Klein bottle.

Motivation and gap: PID-family controllers remain a pragmatic choice for many robotic systems due to their simplicity and interpretability, but tuning stable, high-performing gains is time consuming and typically non-transferable across robot morphologies, payloads, and deployment conditions. Fuzzy gain scheduling can provide interpretable online adjustment, yet its per-joint scaling and consequent parameters are platform dependent and difficult to tune systematically.

Proposed approach: We propose a hierarchical framework for cross-platform tuning of a learnable fuzzy gain-scheduled PID (LF-PID). The controller uses shared fuzzy membership partitions to preserve common error semantics, while learning per-joint scaling and Takagi–Sugeno consequent parameters that schedule PID gains online. Combined with physics-constrained virtual robot synthesis, meta-learning provides cross-platform initialization from robot physical features, and a lightweight reinforcement learning (RL) stage performs deployment-specific refinement under dynamics mismatch. Starting from three base simulated platforms, we generate 232 physically valid training variants via bounded perturbations of mass ($\pm$10%), inertia ($\pm$15%), friction ($\pm$20%), and damping ($\pm$30%).

Results and insight: We evaluate cross-platform generalization on two distinct systems (a 9-DOF serial manipulator and a 12-DOF quadruped) under multiple disturbance scenarios. The RL adaptation stage improves tracking performance on top of the meta-initialized controller, with up to 80.4% error reduction in challenging high-load joints (12.36$^\circ$$\rightarrow$2.42$^\circ$) and 19.2% improvement under parameter uncertainty. We further identify an optimization ceiling effect: online refinement yields substantial gains when the meta-initialized baseline exhibits localized deficiencies, but provides limited improvement when baseline quality is already uniformly strong.

The link between climate change and gender is intricate, particularly in developing countries, where women are disproportionately affected due to lower adaptive capacity and economic opportunities. This study investigates the gendered economic impacts of climate change in South Africa using a recursive dynamic computable general equilibrium model. Results reveal that climate-induced shocks reduce productivity, labour demand and economic growth, leading to higher prices and reduced purchasing power. Female-headed households face greater poverty impacts than their male counterparts. These findings highlight the need for gender-responsive social protection policies that support women in vulnerable sectors such as agriculture and tourism. Targeted interventions such as combining compensatory transfers with investments in education, skills development and access to resilient livelihoods are essential to mitigate gender inequality and strengthen women’s economic resilience in the face of climate change.

Over the past two decades, architectural historiography in Iran has undergone a significant disciplinary transformation. This article examines the effect of the Iranian architectural studies (IAS) MA program establishment at Shahid Beheshti University (est. 2005) in this process. It reconstructs the historical context of architectural education in Iran since the mid-twentieth century, when architectural history occupied a marginal position, and then examines the establishment of the IAS program as a turning point in this trajectory through an analysis of its curriculum, pedagogical strategies, and scholarly output. The article argues that the program has institutionalized architectural historiography, cultivated new generations of scholars and educators, and encouraged greater attention within Iranian schools of architecture to the social, cultural, and contextual dimensions of architecture, bringing them closer to comprehensive schools of the built environment.

Longtermists such as William MacAskill maintain that if we are concerned about human well-being and the longterm future, we should embrace continued technological advancement and development, and that, indeed, one of the most significant threats to longterm human well-being and the well-being of other sentient creatures is technological stagnation – remaining at roughly the present technological level. This paper argues that we should be much more skeptical regarding claims that link continued technological development with positive contributions to well-being. We should take seriously that even if technological development has contributed significantly to improvements in well-being in the past, that we may reach or have already reached a point at which future technological development no longer does so, and that, in that event, thinking about longterm ethical concerns actually recommends developing the capability to slow down, stop, or even roll back (some) technological development.