Science & Mathematics

We study a linear mixing model with dependent latent components, assuming multiple data domains. Most existing models assume that the components are independent or at least uncorrelated, in line with independent component analysis (ICA). Some recent work allows for dependent components, but then makes specific assumptions such as parametric forms of dependencies, multi-view settings, or interventions, or does not recover the individual components. In contrast, we consider a multi-domain setting in which domains differ through domain-specific scalings of the components, while the distribution of the underlying latent components is the same across domains. This approach can model data collected, for example, from different sensors measuring the same process, different laboratories conducting the same experiment, different experimental conditions, or different subjects that might differ in biological or physiological factors. We show that, under sufficient domain variability, latent variables and mixing functions can be identified from second-order statistics alone. We propose the Multi-Domain Covariance Matching (MuDo-CoM) algorithm that generalizes previous methods of joint diagonalization. MuDo-CoM is validated on simulated data and a real-world fMRI dataset.

Emoji are widely treated as compact affective tokens whose meaning is summarised by a single sentiment score. We argue that this picture is incomplete. From a corpus of 16{,}575 English tweets and a per-emoji metric we call \emph{stability delta} (within-class minus between-class cosine similarity of sentence embeddings), we show that emojis used in tightly clustered topical contexts also exhibit \emph{more} polarised crowd-sourced sentiment, not less. Validated against the Kralj Novak et al.\ (2015) sentiment lexicon, the central correlation is Spearman $ρ= +0.361$ (Holm $p = 0.015$, 95\% bootstrap CI $[+0.10, +0.59]$); against the independent Ferré et al.\ (2025) Spanish 13-emotion ratings, $ρ= +0.509$ ($p = 0.0003$). The result survives permutation testing, BCa bootstrap, leave-one-out jackknife, partial correlation controlling for emoji frequency, and Holm correction across a four-test family. Topical coherence and affective dispersion are distinct dimensions of emoji meaning. As a downstream check we report leakage-free five-fold cross-validation results for sarcasm classification on the Emoji Sarcasm Dataset; all three baselines (a gated multimodal architecture, fine-tuned BERTweet features, and fine-tuned RoBERTa features) score above 0.97 macro F1, indicating the dataset has saturating emoji-label cues. We frame the classifier as supporting evidence rather than the contribution.

Predicting spatial gene expression from H&E images offers a scalable complement to experimental spatial transcriptomics. However, despite progress in spatial context modeling, how to construct subsequent context from spot states updated by earlier contextual interactions remains underexplored. We introduce HistoLoop, a looped Transformer that formulates prediction as iterative contextual inference. At each refinement step, HistoLoop aggregates the current spot states into context tokens at the corresponding resolution, processes them with a shared spatial Transformer, and retrieves spot-specific contextual information through cross-attention. A memory-mediated update integrates this information into persistent spot states, which are then used to construct context at the next, finer resolution. Under the HEST-Bench HVG50 evaluation protocol, HistoLoop achieves the highest macro-averaged gene-wise Pearson correlation among the compared methods across six cancer-specific tasks spanning five organs in HEST-1K. Ablation studies support the importance of recurrent state refinement and reconstructing context from updated spot states at progressively finer resolutions. Expression-derived clustering further suggests that HistoLoop better captures spatially organized multigene patterns.

Reliable retrieval-augmented generation (RAG) requires recognizing when neither retrieved documents nor parametric knowledge supports an answer. We introduce DeclineCraft, a dynamic abstention benchmark whose unanswerable questions target Minecraft feature additions after a corpus cut-off, with audits of possible parametric exposure. Later cut-offs support evaluation of future models. Across 27 LLMs, we measure Correct Abstention Rate (CAR; how often models acknowledge insufficient knowledge on unanswerable questions), Answerable Attempt Rate and Answer Accuracy. We find that (1) single-shot retrieval significantly raises CAR in 38 of 40 configurations (model and thinking setting) versus no retrieval. (2) Iterative RAG, in which the model self-assesses its context and can request up to two further retrieval rounds, significantly lowers CAR versus single-shot RAG in 30 of 40 configurations (mean 46.4%), with little change in Answer Accuracy (68.5% to 69.3%). Adding query rewriting and reranking to hybrid (dense & sparse) retrieval lowers CAR for all 12 models tested. (3) Larger models answer more accurately (Spearman $ρ= +0.60$ with total parameters), but neither model size nor release date shows a significant association with CAR. (4) Without retrieval, enabling thinking on the same weights lowers CAR significantly for 8 of 13 models (by 14% points on average), and the responses shift from acknowledged uncertainty to premise denial.

We study discrete-time consensus on the Euclidean unit sphere, where weighted combinations of neighboring states are normalized to unit length. For symmetric weights on connected undirected graphs and ambient dimension $d\ge3$, earlier results establish instability of non-consensus fixed points under suitable weight conditions. For directed graphs on the unit circle, earlier results show that fixed points other than consensus and antipodal configurations occur only for a measure-zero set of weights. We show that these conclusions do not extend directly to general strongly connected directed graphs in higher dimensions. First, we construct full-rank non-consensus fixed points whose tangent eigenvalues all equal one, showing that strict linear instability need not hold. In a separate construction, for every $d\ge3$, we present a sparse strongly connected graph admitting full-rank fixed points on an open set of its permitted weights. Our main result establishes robust counterexamples to almost-global consensus: attracting rotating motions exist for $d=2$ with every $n\ge3$ and for $d=3$ with every $n\ge6$. These motions are locally asymptotically stable modulo common rotations and have asymptotic phase. They persist on open sets of strictly positive weight matrices. For each such matrix, an open set of initial configurations approaches neither consensus nor any fixed configuration. Thus the absence of full-rank non-consensus fixed points does not ensure almost-global convergence to consensus.

Cloud-accessible quantum computing has made heterogeneous Quantum Processing Units (QPUs) available through provider platforms that differ in qubit technology, calibration state, queueing behavior, and pricing model. This heterogeneity makes execution management a quantum-engineering concern: a quantum circuit may be executable on several backends, but the resulting cost, latency, fidelity, and reproducibility depend on backend selection and on how measurement shots are allocated. This paper presents the Quantum Broker, an open-source and provider-agnostic framework for Quality-of-Service-aware quantum execution management. The framework operationalizes shot-wise distribution by exploiting the statistical independence of quantum measurement shots to dispatch a single circuit across multiple heterogeneous QPUs according to user-defined constraints and optimization objectives. It combines a modular execution architecture, interchangeable backend-profiling strategies, and the Quantum QoS Specification Language (QQSpec), a declarative language for expressing QoS requirements. Dispatch planning is supported by two solver backends: a mixed-integer linear-programming (MILP) solver, and a nonlinear dual-annealing solver. Experiments on benchmark quantum circuits and noisy simulator backends show that the Quantum Broker can produce feasible QoS-aware dispatch plans under heterogeneous backend conditions and can improve QoS satisfaction with respect to single-backend and naive multi-backend baselines. The results also expose a practical trade-off between planning time and dispatch quality, with the nonlinear solver producing higher-quality allocations at increased computational cost. Overall, the Quantum Broker provides an extensible execution-management layer for engineering portable, reproducible, and QoS-aware quantum software over fragmented quantum cloud infrastructures.

Generative artificial intelligence (GenAI) is rapidly transforming knowledge work, creating new collaborative dynamics that challenge tradi-tional human-human interaction patterns. This study investigates how Transactive Memory Systems (TMS) manifest in human-GenAI collabora-tion. Through qualitative interpretive interviews with 14 knowledge work-ers and thematic analysis, we identified distinct TMS manifestations across its three main dimensions: (1) regarding specialization, expertise recogni-tion exhibited asymmetry. Humans developed awareness of GenAI capabili-ties while GenAI demonstrated only temporary recognition of human exper-tise; (2) in terms of credibility, humans formed ambivalent trust relation-ships with GenAI, simultaneously attributing expertise to the latter while implementing additional verification processes; and (3) concerning coor-dination, where humans primarily directed more passive GenAI systems lacking initiative, resulting in hierarchical interaction patterns and unidi-rectional information flow. Our findings reveal that TMS manifests differ-ently in human-GenAI compared to human-human collaboration, necessitat-ing theoretical adaptations including trust markers, acknowledging GenAI's absent metamemory, and identifying its temporary memory.

Autonomous video editing from open-world resources remains a challenging problem in multimedia content creation. Retrieval-based methods are constrained by fixed source material and lack creative flexibility, while generative models struggle with long-horizon coherence and often diverge from user intent. Both paradigms fail to replicate the structured, multi-stage reasoning that underlies professional post-production. To address these limitations, we present Auto-Director, a tool-augmented multi-agent framework that mirrors professional editorial workflows: a Planning Agent decomposes abstract topics into fact-grounded scripts via retrieval-augmented planning; a Material Agent indexes open-world footage through shot segmentation and visual tagging; an Editor Agent assembles semantically aligned clips with MLLM-driven spatial adaptation; and a Critic Agent governs an iterative Reflect-and-Revise loop that checks scene-level alignment and pacing. To rigorously evaluate this new paradigm, we introduce VEdit-Bench, a benchmark of 100 expert-curated topics with human-annotated Golden Scripts and a five-dimensional metric suite covering Narrative Logic, Factual Accuracy, Semantic Alignment, Compositional Quality, and Stylistic Cohesion. Extensive experiments demonstrate that Auto-Director significantly outperforms baseline methods in both automated metrics and human evaluation, effectively bridging the gap between amateurs and professional-grade video production.

It was already recognised in the 2004 and 2014 SKA Science Case Chapters: "The Exploration of the Unknown" and "The SKA and the Unknown Unknowns", respectively, that the exploration of the unknown is an extremely important concept in realising the potential of any new instrument. This is especially true in an era of limited budgets but expanding analysis algorithms and capabilities. While this Chapter considers transients and environments, the ethos of these previous Chapters should be considered and embraced broadly and can be applied more broadly across SKAO science. We do not repeat in detail the important background these Chapters provide but encourage the reader to review them before reading this Chapter. In the last decade of radio transient astronomy, we have seen the value of these approaches with the expansion of the FRB population, the discovery of long period pulsars and transients and radio emission from white dwarfs to name just a few. With the SKAO, we should push the explored parameter space further and in new directions in all of the sensitivity, field-of-view, time, spectral, spectro-dynamic, polarisation and imaging domains. We should utilise approaches such as machine learning in allowing us to address the complexity that arises from the expanded parameter space and increased data rates and candidate sources. We will briefly summarise some of the impressive progress that the SKA pathfinders and precursors have made in these areas and what these mean for the SKA telescopes. We will also discuss ways in which it might be possible to push the current capabilities.

Machine learning-based fault detectors are often evaluated at the relay locations used for training, although local voltage and current measurements vary across power networks. This work investigates cross-relay generalization for binary detection of faults anywhere in the network using local measurements, rather than line-selective tripping. Using 50 ms windows from a simulated 110 kV transmission network, we evaluate a multilayer perceptron across all 64 source target combinations of eight relays and five random seeds. A shared simulation-level split and source-only threshold selection provide a controlled held-out evaluation without target-specific calibration. Despite nearly unchanged mean F1-scores (0.730 within-relay versus 0.734 cross-relay), the mean false-positive rate (FPR) increases from 0.088 to 0.341, while the mean false-negative rate decreases from 0.399 to 0.318. The change is strongly directional: models trained at Bus 2 relays yield a mean FPR of 0.670 at Bus 1 and Bus 3 relays, compared with 0.096 for cross-relay transfers within the same bus group. Linear probing indicates that learned representations retain coarse bus-group information, while exploratory class-conditional latent distribution analysis shows no consistent monotonic association with transfer errors. These findings demonstrate that aggregate metrics can conceal location-structured changes in false-alarm behavior, highlighting the importance of evaluating error types separately when assessing cross-relay generalization.

Realistic few-shot adaptation of medical vision-language models (VLMs) can leave low-prevalence classes without any labeled support. The training-free SS- Text+ solver remains operational in this setting, but its zero-support prototypes receive no visual adaptation and fall back to a purely textual class prototype. We in- troduce Gated Knowledge-Guided Prototype Transfer (GProtoT), a pre-repulsion module that transfers supported-class visual residuals to missing classes using VLM similarity and an ontology-based clinical distance. A validation-free evidence gate scales transfer based on semantic relevance and residual agreement, thereby limiting unsupported extrapolation. In 50 paired realistic one-shot trials on MESSIDOR with FLAIR features, severe and proliferative diabetic retinopathy were absent from every support set. Relative to SS-Text+, gated ProtoT improved missing-class recall by 5.23 points, average class accuracy by 1.37 points, and the missing/supported recall harmonic mean by 1.83 points. It reduced negative- transfer frequency from 46-48% for ungated transfer to 2%, while at the cost of reductions of 1.20 percentage points in supported-class recall and 0.45 percentage points in top-1 accuracy. These results indicate that evidence-based gating provides a safety mechanism for the knowledge source. GProtoT is based on SS- Text+ and requires no gradients, validation data, or additional model inference, of- fering a low-compute path toward accountable adaptation under incomplete clini- cal supervision. Code is available: https://github.com/sumaiyaafridi/GProtoT

Reconstructing dynamic 3D scenes from monocular video requires inferring motion from noisy and intermittently visible observations. Long-range tracks provide valuable geometric constraints, but integrating their fragmented temporal coverage into shared, continuous motion remains challenging. We introduce RayCo4D, a ray-conditioned continuous motion model for monocular 4D reconstruction that explicitly separates visual measurements from latent 3D motion states. Each track is associated with a reference-time 3D query constrained to its camera ray, and its image and depth observations constrain shared latent motion nodes through local many-to-many associations. Each node maintains an $\mathrm{SE}(3)$ pose and body-frame velocity at sparse support times. A white-noise-on-acceleration process prior couples neighboring states and defines continuous-time interpolation, allowing observations across visibility intervals to jointly constrain the same latent state sequence. The inferred shared motion is transferred to Gaussian primitives through dual-quaternion blending, while reference-anchored local $\mathrm{SE}(3)$ residuals capture fine-scale non-rigid deformation beyond the shared scaffold. Experiments on diverse monocular dynamic scenes show that RayCo4D achieves competitive novel-view synthesis quality, improves structural and temporal consistency under rapid motion and incomplete tracks, and supports continuous-time rendering.

Satellite-based continuous-variable quantum key distribution (CV-QKD) is fundamentally constrained by finite communication windows, channel attenuation, and statistical fluctuations arising from a limited number of exchanged quantum states. This work develops a finite-size CV-QKD framework for satellite quantum communication under a hybrid noise model that jointly incorporates satellite geometry, atmospheric and propagation losses, quantum Poissonian noise coexisting with classical additive white Gaussian noise, detector imperfections, and composable finite-key security under collective attacks. A single canonical finite-size secret key rate (SKR) expression is derived in which the finite-size correction is stated in a strictly decreasing form in the block size, removing the ambiguity between the segmented and the block-level formulations used in earlier treatments. The finite-size accounting is further extended to the quantum bit error ratio (QBER) through an explicit parameter estimation penalty, so that the feasibility constraint inherits the same block-size dependence as the key rate. The framework quantifies the influence of reconciliation efficiency, transmission efficiency, detection efficiency, receiver aperture, beam waist, zenith angle, wavelength, and effective communication time. Numerical results reveal a significant performance gap between finite-size and asymptotic operation, indicating that asymptotic analysis can overestimate the feasible satellite altitude by a factor of several, and that the achievable range is governed by the number of signals actually exchanged rather than by any single physical channel parameter. These results establish block size as a first-class design variable for satellite CV-QKD mission dimensioning.

Spatial reasoning asks a model to work out where objects are in 3D space, how they relate to one another, and how they move. Agentic VLMs have recently brought large improvements on such tasks by letting the model actively gather evidence with 3D perception tools instead of answering in a single pass. Even so, on many questions the tools return the right evidence and the model still does not arrive at the correct answer: the gap between model and tools, where raw readings become usable conclusions, is left unfilled. We present S-Agent 2.0, a spatial reasoning agent that equips an off-the-shelf general coding agent with a command-line perception toolkit, and we design the interface around one principle: help the VLM close the evidence loop: acquire scene evidence, remember it in a persistent scene memory, and interpret each reading in plain spatial language. This design keeps the full flexibility of a coding agent, and every reading it acts on stays grounded in the tools' geometry. On MMSI-Bench and ReVSI, the toolkit improves a strong open-source model, Qwen3.8-27B, by +13.9 and +20.5 points over the same agent without tools, and a strong closed-source model, Gemini-3.8-Flash, by +11.2 and +9.6 points. With the open backbone S-Agent 2.0 reaches 66.5 on MMSI-Bench and 82.3 on ReVSI, ahead of the specialized system SpatialClaw on the same backbone (62.6 and 74.5), and with Gemini-3.8-Flash it reaches 74.1 and 90.5 against its 67.0 and 79.7. Both tool gains shrink as the backbone grows more capable, but SpatialClaw's shrinks far more on ReVSI, from +13.2 to +2.7, while ours stays at +9.6. S-Agent 2.0 also works across different general coding agent harnesses, such as Codex and Claude Code.

In speculative decoding, a lightweight draft model proposes tokens that the target model verifies in parallel, speeding up generation without changing the target distribution. D-PACE (Wu et al., 2026) trains such drafters with a loss that weights each position by its contribution to the expected accepted length. Both its weights and its cross-entropy (CE) loss use only the draft's probability of the recorded target token, treating the target as a point mass. Under rejection sampling, however, a sampled draft token is accepted with probability equal to the overlap of the draft and target distributions, one minus their total variation (TV). We derive D-PAL (Dynamic Position-wise Acceptance Loss) from the exact expected accepted length, using this overlap as the acceptance signal. This yields overlap-based position weights with log acceptance (LA) as the per-position loss, and the objective reduces to D-PACE for a point-mass target. Since the weights are detached, any per-position loss can replace LA in practice. We show that forward KL divergence and Rényi divergence of order 1/2 match or exceed LA, and that Rényi-1/2 provably bounds the true acceptance rate from above and below. Replacing CE with Rényi-1/2 improves acceptance length under both D-PACE's weights and ours. On six benchmarks with a Qwen3-4B target, D-PAL raises mean acceptance length over D-PACE from 4.50 to 4.57 at T=0 and from 4.20 to 4.28 at T=1, and consistently achieves higher throughput at both temperatures.