Large language model (LLM) activations are notoriously difficult to understand, with most existing techniques using complex, specialized methods for interpreting them. Recent work has proposed a simpler approach known as LatentQA: training LLMs to directly accept LLM activations as inputs and answer arbitrary questions about them in natural language. However, prior work has focused on narrow task settings for both training and evaluation. In this paper, we instead take a generalist perspective. We evaluate LatentQA-trained models, which we call Activation Oracles (AOs), in far out-of-distribution settings and examine how performance scales with training data diversity. We find that AOs can recover information fine-tuned into a model (e.g., biographical knowledge or malign propensities) that does not appear in the input text, despite never being trained with activations from a fine-tuned model. Our main evaluations are four downstream tasks where we can compare to prior white- and black-box techniques. We find that even narrowly-trained LatentQA models can generalize well, and that adding additional training datasets (such as classification tasks and a self-supervised context prediction task) yields consistent further improvements. Our best AOs match or exceed white-box baselines on all four tasks and the best overall baseline on 3 of 4. These results suggest that diversified training to answer natural-language queries imparts a general capability to verbalize information about LLM activations.
We investigate a framework for train-free MRI segmentation based on Topological Data Analysis. The pipeline proceeds in three steps, first identifying the whole object to segment via automatic thresholding, then detecting a distinctive subset whose topology is known in advance, and finally deducing the various components of the segmentation. A key ingredient is the extraction of approximate representative cycles from persistence diagrams, which provides an interpretable link between persistent features and anatomical components. To clarify the method’s scope, we make the underlying topological and intensity assumptions explicit, quantify when they hold on real data, and analyze typical failure modes. We evaluate the approach on glioblastoma and on fetal cortical plate segmentation, with comparisons to unsupervised and deep-learning references. By operating without large annotated datasets, the method is well suited to scarce-data settings and provides an interpretable baseline and practical initialization for expert refinement or learning-based pipelines.
We present the first edge Er measurements in negative triangularity (NT) plasmas in the Tokamak & agrave; Configuration Variable (TCV). The Doppler backscattering measurements of v perpendicular to approximate to Er/B reveal a significant impact of triangularity on the Er well: in Ohmic, neutral beam injection, and electron cyclotron resonance heated discharges, the Er well and associated ErxB shear are stronger in NT-shaped plasmas compared to their positive triangularity (PT) counterpart. This suggests a connection to the concomitant NT performance gain relative to PT L-mode.
Polarity is a fundamental reciprocal duality of n-dimensional projective geometry which associates to points polar hyperplanes, and more generally k-dimensional convex bodies to polar (n-1-k)-dimensional convex bodies. It is well-known that the Legendre-Fenchel transformation of functions can be interpreted from the polarity viewpoint of their graphs using an extra dimension. In this paper, we first show that generic polarities induced by quadratic polarity functionals can be expressed either as deformed Legendre polarity or as the Legendre polarity of deformed convex bodies, and be efficiently manipulated using linear algebra on (n+2)× (n+2) matrices operating on homogeneous coordinates. Second, we define polar divergences using the Legendre polarity and show that they generalize the Fenchel-Young divergence or equivalent Bregman divergence. This polarity study brings new understanding of the core reference duality in information geometry. Last, we show that the total Bregman divergences can be considered as a total polar Fenchel-Young divergence from which we newly exhibit the reference duality using dual polar conformal factors.
We provide a first-principle construction of baryons as quantum Hall droplets in single-flavor holographic QCD. The baryons are described as charged D6-branes with a circular boundary on a flavor D8-brane in the Type IIA backgrounds dual to the confining and non-confining phases. The holographic description allows us to calculate precisely their properties, such as mass and size. We also consider other objects with baryonic charge, such as vortons, domain walls with holes, and "sandwich vortons", and discuss the relative (meta)stability of all these configurations.
