Papers

Sandboxed computer environments support multi-step reasoning with tools, executable programs, and persistent files, yet their extension from language models to vision-language models (VLMs) introduces a distinct state-management problem. Visual reasoning produces intermediate image-valued evidence—crops, masks, overlays, zoomed regions, and analytic renderings—that must remain addressable without accumulating unboundedly in multimodal context. We introduce VLM-in-Sandbox, a training-free framework for agentic multimodal reasoning in controlled computer environments. Its Visual Workspace registers generated artifacts in an image ledger, maintains a bounded active visual context, and lets the model explicitly promote selected evidence for subsequent inspection. This separates visual evidence generation, performed by sandbox tools, from visual evidence management. Across seven benchmarks and four base VLMs, VLM-in-Sandbox achieves the highest sample-weighted average accuracy among Vanilla VLM, Append-only Sandbox, and the proposed method. A compiler-matched $2\times2$ study on 1,260 examples further separates model-directed visibility from bounded retention: VLM-in-Sandbox reaches 66.27% accuracy with 18.6% fewer total tokens than the automatic, retain-all control. Over all 6,350 submitted GPT-4.1-mini examples, it produces 302 rescues and 142 regressions relative to Original Append-only. A local vLLM study with prefix caching confirms that the smaller request workload also reduces uncached tokens, time to first token, and end-to-end latency. These results identify explicit visual evidence state as a central abstraction for sandboxed VLM agents.

We develop a new parametrization for generic departures from the $\Lambda$CDM model based on the Markov-Mukhanov action with non-minimal matter-gravity coupling. In this approach the cosmological constant $\Lambda$ and the gravitational coupling $G$ become dynamical with their values determined by the explicit form of the matter-gravity coupling. We use the framework to build a simple phenomenological model where dynamical dark energy is naturally obtained as a consequence of the departure of the theory from classical General Relativity at low densities. We show how the model's free parameters can be constrained from observational data and interpret the tentative observations of `phantom crossing' by the DESI collaboration in view of the proposed formalism.

In this paper we propose a universal way to define multivariate orthogonal polynomials for the positive branch of WLZZ matrix models. This allows us to generalize from the Gaussian case, previously considered in the literature, the strict superintegrability property and the Miwa deformation formulas.

Image forensics is increasingly an open-world problem: manipulations range from fully synthetic images to localized edits, splicing and swapping, while most forensic detectors remain specialized to a single manipulation family. Agentic AI has recently emerged as a promising solution. In principle, such systems can assess the reliability of individual detectors, identify out-of-scope evidence, and arbitrate conflicting reports. However, it remains unclear which components actually drive performance and whether their benefits persist under distribution shift. To answer these questions, we study a training-free agentic framework built around specialist detectors, per-detector triage, and conflict-aware evidence arbitration. Using six configurations and three multimodal large language model backbones, we dissect the role of triage, prompting, and reasoning quality on both in-distribution and out-of-distribution data. Our results show that naive detector fusion suffers from severe false-positive rates on authentic images. Triage and prompting consistently improve performance by filtering unreliable evidence and exposing detector limitations. However, the dominant factor is represented by reasoning itself: A stronger judge substantially outperforms a weaker one, particularly under distribution shift. Most notably, manipulation recall is nearly saturated across all configurations, indicating that the main challenge of open-world image forensics is not detecting manipulations, but calibrating trust in specialized forensic tools and arbitrating conflicting evidence.

Predicting when maritime systems require maintenance can be critical, avoiding hazards and costly consequences. To address this problem, this paper proposes an explainable decision-making framework that integrates a neuro-fuzzy prediction model with a two-stage explainable component. The first stage of this component produces feature-attribution explanations, using gradient-based saliency maps, and the second stage extracts local rules using a fuzzy decision tree. The proposed framework is generic and can be integrated into any deep learning-based approach, rendering it explainable. To the best of our knowledge, this is the first fuzzy logic-based framework enabling both feature-level and local rule-based explanations of black box models. This approach aims to foster trustworthiness in decision making through user-understandable machine inferences. The performance of the proposed framework using a deep residual-based neural backbone is evaluated on various general-purpose public benchmark datasets, and its utility in maritime is demonstrated in the context of early fault detection in a naval propulsion system dataset. The results indicate that it can provide predictions outperforming relevant state-of-the-art approaches, with an average AUC-ROC (Area Under the Receiver Operating Characteristic Curve) value, reaching up to 99%, while offering the advantage of explainability.

Cross-domain Named Entity Recognition (CD-NER) aims to transfer the rich knowledge in the source domain to the target domain. Recent studies adopting decomposition or generation paradigms have achieved significant performance improvements, demonstrating high accuracy in entity span detection. However, during entity type classification, models severely suffer from entity type confusion, the erroneous tendency that models classify entities of one type in the text as another similar but incorrect type. To address this issue, we first propose a Multidimensional Confusion Quantification Model (MCQM) that quantifies a model's confusion extent between entity types from three dimensions: source-target hierarchy analysis, semantic similarity analysis, and explicit data evaluation. Moreover, we propose the Progressive Bidirectional Reasoning Chain (PBRC). PBRC leverages the source-target hierarchy and confusion analysis from the MCQM to prompt the LLM to generate two-stage reasoning information. The two-stage reasoning information is utilized to augment the knowledge of the model, significantly mitigating entity type confusion and improving the model's generalization performance. Experimental results demonstrate that our method achieves new state-of-the-art results on all domains of the CrossNER dataset.

Capacity Expansion - Energy System Models (CE-ESMs) have been widely used to optimize the decarbonization of future energy systems on national and continental scales. To capture the limitations in electricity trade between different parts of the system investigated, CE-ESMs often include a representation of the transmission grid. However, CE-ESMs usually only include linearized representations of the grid and omit or strongly simplify grid stability requirements. In this work, we evaluate the N-1 voltage security of the electricity system results obtained from a CE-ESM of a decarbonized Nordic energy system in Year 2050, and analyze the effects of different grid code requirements and shunt capacitor and reactor automation on voltage security. For comparison, we evaluate a Year 2022 system based on ENTSO-E data.We show that the CE-ESM results for the Nordic countries are not N-1 voltage-secure and that, depending on the grid code and shunt automation requirements, the contingencies with voltage violations range from 1.1% to 17.4%. Furthermore, the results indicate that both Power Park Module voltage control and shunt extreme voltage automation have strong positive effects on voltage security in the modeled future system. Grid regions with low levels generation, and thus little dynamic reactive power from generators, are found to be the most at risk of being voltage-insecure.

We study rolls and homoclinic snaking in a variation of the one-dimensional Swift-Hohenberg equation, whose standard forms are prototypical order-parameter models for pattern formation in the sciences. Motivated by classes of differential equation models that involve continuous non-smooth low order nonlinear terms, we replace the standard quadratic-cubic nonlinearity by $\nu |u|^\alpha-u^3$, $\alpha \in [1,2]$ with $\nu > 0$. In the vicinity of zero, for $\alpha<2$ this nonlinearity falls outside the scope of classical Taylor expansion and bifurcation analysis. Our partially analytical and partially numerical results highlight that the non-smooth term modifies the criticality of pattern-forming bifurcations and alters the associated branches of solutions. In particular, $\alpha\in(1,2)$ implies subcriticality of roll bifurcations for any $\nu>0$. At $\alpha=1$ differentiability is lost, which has a strong impact on the bifurcations of sign-changing rolls including the disappearance of homoclinic snaking. Homoclinic snaking thus emerges non-smoothly as $\alpha$ increases from $\alpha=1$, and persists when retaining an additional, e.g., quadratic term.

In this paper, we further develop the real link Floer homology defined by the first author. We introduce a new base-pointing convention that leads to a different version of real link Floer homology and show that this new theory is related to the old one by an exact triangle. We also define a real link Floer theory for multi-based strongly invertible links, which is a strong real Heegaard invariant, and take a first step toward a real link Floer TQFT. A computer implementation for the new theory via grid diagrams was written by Zhenkun Li. We also include an appendix containing real grid homology of more than 50 small knots.

For a family $\mathcal{F}$ of $k$-graphs, $ex_k(n,\mathcal{F})$ denotes the maximum number of edges in an $n$-vertex $\mathcal{F}$-free $k$-graph. Let $M_{s+1}^k$ denote a matching of size $s+1$ in $k$-uniform hypergraphs. Recently, Alon and Frankl (JCTB, 2024) determined $ex_2(n,\{M_{s+1}^2,K_{\ell+1}\})$ for all $n\geq 2s+1$ and $\ell\geq 2$. For every non-bipartite graph $F$, Gerbner (JGT, 2024) determined $ex_2(n,\{M_{s+1}^2,F\})$ for sufficiently large $n$. In this paper, we investigate this problem for different ranges of the matching parameter. First we prove that for every graph $F$ with $\chi(F)>3$, there exist constants $\beta>0$ and $s_0$ such that $ex_2(n,\{M^2_{s+1}, F\})=ex_2(2s+1,F)$ for $\max\{s_0,n/2-\beta n\}<s<n/2$. For integers $\ell\ge k\ge3$, let $\mathcal{K}_{\ell+1}^k$ be the family of all $k$-graphs $F$ with at most $\binom{\ell+1}{2}$ edges for which there is an $(\ell+1)$-set $L$ such that every pair of vertices of $L$ is covered by an edge of $F$, and let $H_{\ell+1}^k$ be the $k$-uniform hypergraph obtained from the complete graph $K_{\ell+1}$ by enlarging each edge with a set of $k-2$ new vertices, which is a member of $\mathcal{K}_{\ell+1}^k$. We determine $ex_k\bigl(n,\mathcal{K}_{\ell+1}^k\cup\{M_{s+1}^k\}\bigr)$ for $s< \frac{n}{8(k-1)^{k-2}(\ell-1)^3}$. For sufficiently large $s$, we also determine $ex_k\bigl(n,\{M_{s+1}^k,H_{\ell+1}^k\}\bigr)$ for $\frac{n}{k}-\beta n<s\leq \frac{n-k}{k}$ and $s< \frac{n}{8(k-1)^{k-2}(\ell-1)^3}$.

Large language models (LLMs), when acting as agents, are expected to take observed data in context, infer the latent state space underlying the world, and leverage it for downstream prediction. However, prior work demonstrated that LLMs struggle to use representations learned in context on a graph tracking task, where the model needs to construct a representation of the graph governing data generation process and use it for subsequent predictions. In this paper, we show that extending this to few-shot settings, where each demonstration is generated from a different world with either the same or different graph topologies, enhances its prediction on 6 models from 4 model families. To understand this improvement, we linearly probe a low-dimensional world representation that encodes graph information in the hidden states. Notably, we find that few-shot demonstrations relocate the world representation and increase its predictive use. Specifically, for each model, these world representations shift in directions nearly orthogonal to their original subspace, and interventions on these representations selectively impair performance more than interventions on other subspaces. Consistent with this insight, we show that few-shot demonstrations with observations from different worlds improve performance on ARC-AGI-1&2, web agent tasks, and Othello. Our findings elucidate the role and internal mechanisms of few-shot demonstrations in in-context world modeling. More broadly, our work advances our understanding of how LLM agents learn from in-context observations and provides implications for their further improvement.

Strongly lensed gravitational waves provide exceptionally precise measurements of relative macroimage arrival times, making their time delays a promising time-domain probe of gravity on galaxy-cluster scales. This precision places stringent demands on the control of lensing systematic effects. We present a proof-of-concept pipeline for constraining a phenomenological modification, $D$, of the galaxy cluster Weyl potential using the macroimage time-delay span of a simulated sample of binary neutron star (BNS) sources and cluster lenses described by a Navarro–Frenk–White cluster halo and a singular isothermal sphere for the brightest cluster galaxy (BCG). Closure tests recover the injected general-relativistic (GR) limit. Adopting a fiducial 5 per cent precision on the reconstructed time-delay span, we assess controlled mis-specifications of the halo mass and concentration, BCG contribution, source and lens positions, redshifts, line-of-sight environment, projected ellipticity and secondary cluster-scale structure. Within the adopted gravity model, neglecting halo ellipticity, secondary cluster-scale structure, cluster mis-centring and line-of-sight perturbations can induce significant, strongly configuration-dependent biases in the inferred parameter $D$. Biased estimates of the cluster halo mass and concentration, as well as systematic BNS source-position errors at the level considered here, produce significant coherent shifts. By contrast, uncertainties in the BCG contribution and in the lens and source redshifts are comparatively subdominant. These results highlight the need for explicit treatment of the above uncertainties before cluster-lensed gravitational wave time delays can be used as precision tests of gravity.

Robotic foundation models achieve impressive performance on standard manipulation benchmarks, yet these evaluations typically assume clean, timely, and consistent visual observations throughout execution. We introduce LIBERO-VPro, a benchmark for systematically evaluating the closed-loop visual robustness of robotic foundation models by perturbing the visual evidence available during execution. LIBERO-VPro covers four complementary dimensions, including Visual Evidence Degradation, Camera Staleness, Visual Source Consistency, and Task-Relevant Scene Variation, spanning 12 challenge categories, 96 experimental settings, and 3,296 task-condition cases. We evaluate three vision-language-action models and three world-action models over approximately 196,000 simulated episodes, complemented by 200 real-world rollouts on a Franka Research 3. Our results reveal that strong nominal performance can mask substantial weaknesses in visual grounding and adaptation. Models often remain successful despite severe object-level occlusion, yet degrade sharply when local interaction cues are disrupted or familiar spatial priors are violated. They are also highly sensitive to stale or missing observations and struggle when changed task preconditions require behavioral adaptation. Finally, VLAs and WAMs exhibit distinct robustness profiles, showing that visual robustness is multi-dimensional and architecture-dependent. LIBERO-VPro provides a systematic diagnostic framework for developing robotic foundation models that can more reliably ground and adapt their actions under challenging visual conditions.

We reduce Lalonde's pseudo-telepathy inner product game from six to five dimensions, keeping its four Alice questions and three Bob questions and reducing each answer alphabet to five. We prove that the reduced game self-tests its nonmaximally entangled state and all 35 measurement effects. The self-test is also robust: for a strategy winning with probability at least $1-\varepsilon$ we obtain $d_{\mathrm{ext}}\le\min\{1,2^{21}\sqrt{\varepsilon}\}$, where $d_{\mathrm{ext}}=\sqrt{1-\Xi}$ and $\Xi$ is the optimal squared fidelity with the reference state under local extraction channels. A compactness argument then gives simultaneous qualitative robustness for all measurement actions on the state under common local isometries. As a consequence of the self-test, no maximally entangled state of any finite dimension can win the game perfectly.

AI coding agents increasingly support software development beyond code completion, including planning, implementation, testing, and repository-level task execution. Their practical use, however, often remains only weakly connected to established software engineering practices. The aim of this work is to develop and evaluate a lightweight, spec-driven lifecycle for governed agentic software engineering. The lifecycle combines established software engineering practices with repository-local guidance through specifications, AGENTS.md, and phase-specific agent skill files. The approach was developed in the context of the FHNW course AI-assisted Software Development and applied by students to business-oriented software use cases. Its educational and practical applicability is explored through a student survey combining closed rating items with open-ended questions. The contribution of this work is a process-oriented framework that enables AI coding agents to operate with bounded autonomy within an explicit, reviewable, and test-oriented software development lifecycle.

In this paper, we consider the asymptotic density of $T_c(x):=\#\{p\leq x:P^+(p-1)\geq p^c\}$, where $P^+(n)$ denote the largest prime factor of $n$. We show that for $x\rightarrow\infty$, for any $0<c\leq 1/2$, one has \begin{align*} \mathop{\lim\inf}_{x\rightarrow\infty}\frac{T_c(x)}{\pi(x)}&\geq\max\left(1-\frac{16}{5}\rho\left(\frac{1}{c}\right),1-c+g(c)\right), \end{align*} where $g(c)>0$ for $11/32<c\leq1/2$, and $g(c)=0$ for $0<c\leq11/32$. In particular, we have $g(1/2)>0.0436$. This improves a previous result by Liu-Wu-Xi (2020), who showed \begin{align*} \mathop{\lim\inf}_{x\rightarrow\infty}\frac{T_c(x)}{\pi(x)}&\geq\max\left(1-4\rho\left(\frac{1}{c}\right),1-c\right), \end{align*} for $0<c\leq 1/2$.

Graph Retrieval-Augmented Generation (GraphRAG) has remarkably enhanced large language models on complex reasoning by leveraging structured entity topologies. However, existing frameworks heavily rely on standard autoregressive language models where the nature of inherent sequential generation severely hinders overall inference efficiency. Inspired by Diffusion Language Models (DLMs) that offer massive parallelism via continuous refine-in-parallel decoding, we aim to accelerate GraphRAG in the discrete space. However, it remains non-trivial for two challenges. First, partially denoised drafts are highly dynamic and uncertain, making dynamic graph grounding non-trivial. Second, raw denoising states are inherently noisy and unstable, making synchronous graph retrieval and multi-hop aggregation computationally prohibitive. To this end, we present LADDER, a novel framework that bridges diffusion language modeling with GraphRAG through graph-guided parallel decoding. Specifically, (i) we propose an event-driven self-clocking retrieval, inspired by our key insight that 88% of target entities emerge early in the partially denoised state, leading final commitment by an average of 5.7-9.6 steps. This mechanism dynamically triggers graph retrieval only when the set of graph-linkable entities expands, yielding an asynchronous self-clocking policy that bypasses learned gates or heuristic thresholds. (ii) An incomplete-query graph propagation module is designed to process the newly emerging entity queries using a specialized graph foundation model, continuously aggregating multi-hop evidence to sharpen parallel predictions and accelerate overall decoding convergence. Extensive experiments on three challenging multi-hop QA benchmarks show that LADDER raises average exact match from 39.6% to 45.2% while achieving a 4.1x latency reduction.

Extending the recursion principles of a formal system is an enticing but dangerous endeavour with a well-documented history of leading to consistency bugs. Nakano's guarded recursion is an elegant, type-based approach to soundly extend type theory with a powerful recursion principle. This makes guarded recursion useful for many applications, from programming with infinite structures such as streams to reasoning about advanced programming language features using synthetic guarded domain theory. Sadly, guarded recursion is not directly supported by any major interactive theorem prover, which leaves users of guarded recursion with unmechanised pen-and-paper proofs or mechanisations that depend on unmaintained theorem provers. In this paper, we present an implementation of guarded recursion as an embedded language in Lean consisting of a simply-typed lambda calculus for definitions and a higher-order logic for reasoning. Using Lean's excellent support for metaprogramming, our language allows users to write guarded recursive definitions in an intuitive syntax and to prove properties about them using a dedicated proof mode. We give our language a presheaf model, which we use to prove the soundness of our language and to allow users to export guarded recursive definitions and their theorems into standard Lean developments. To demonstrate the usefulness of our language, we present several case studies for programming and reasoning with guarded recursion.

We establish a saddle-order proposition for the ensemble-averaged generalized free energy in reduced Euclidean black hole ensembles. Under regularity and stability assumptions, the leading fluctuation contribution is determined by the order of the first stabilizing term around the dominant saddle. An ordinary critical point is governed by a quartic soft mode, giving a quarter-temperature contribution instead of the familiar half-temperature Gaussian result. We verify this analytically for four-dimensional RN-AdS and Kerr-AdS black holes and map their thermodynamic controls to a common quartic normal form, whose scaling function connects the critical and Gaussian regimes along the zero-field one-well direction. The proposition also extends to coupled collective modes with weighted-homogeneous leading potentials. These results provide a unified description of reduced-ensemble fluctuations through black hole criticality.

The scaling of GPUs for AI and HPC workloads is increasingly constrained by the capacity, bandwidth, and thermal limits of both 2.5D HBM-GPU and direct-stacked 3D HBM-on-GPU integration. This work establishes the thermal design envelope for 3D volumetric DRAM-on-GPU integration, in which vertically oriented DRAM dies and interleaved cooling cavities reshape heat flow and memory interfacing above the GPU. Using a package-level thermal model anchored to a consistent HBM-on-GPU baseline and driven by a realistic reticle-scale non-uniform GPU power map, we quantify the key parameters governing thermal feasibility. Stack height is the dominant limiter of peak temperature, while cooling-cavity conductivity shifts the feasible region, and mold insertion and stack orientation further modulate thermal behavior. A distributed memory-controller and network-on-chip tier introduces only a moderate thermal penalty. Although die-level parallelism increases bandwidth, the reduction in simulated training time saturates once execution becomes compute-bound. These results define a bounded co-design space across bandwidth, capacity, and thermal constraints for 3D volumetric DRAM-on-GPU integration.

We determine the maximum stable volume of bubbles attached to flat surfaces as a function of contact radius, contact angle, and capillary length. By solving the Young-Laplace equation using a shooting method, we calculate equilibrium bubble shapes and determine when bubbles detach through three distinct modes: necking at the cavity, sideways instability, and spreading followed by necking on the surrounding surface. Our predicted detachment volumes agree well with experimental measurements of electrolytic, boiling, and carbonated bubbles, as well as pendant drops. These results bring together and extend previous analytical estimates and, by symmetry, also apply directly to pendant drops.

Context: Large language models (LLMs) can generate unit tests quickly, but high structural coverage does not establish that those tests execute reliably or detect faults. Existing evidence often treats coverage as the principal outcome and rarely compares prompt strategies and models through mutation testing at class level. Objective: This study examines how prompt strategy and model choice shape the executability, structural coverage, fault-detection effectiveness, and structural quality of LLM-generated Python test suites relative to human-written suites. Method: We evaluate multiple prompt strategies across a diverse set of current LLM configurations on the ClassEval benchmark. The evaluation combines execution outcomes, line and branch coverage, Cosmic Ray mutation scores, and structural quality indicators. Primary analyses treat successful execution as a prerequisite; paired comparisons use only classes shared by the relevant executable subsets. Results: Structural coverage is consistently near its ceiling and offers little discrimination among configurations. Executability varies substantially. The proposed prompt performs strongly for mutation score, but no prompt dominates across models. Model choice explains more variation than prompt choice, and their interaction shows that prompt effectiveness depends on the selected model. Human and LLM suites are evaluated on unequal executable subsets, so their relative mutation scores do not establish superiority. Conclusion: Reliable assessment of LLM-generated tests should treat executability as a gate and combine coverage with mutation testing and structural quality indicators. In practice, model selection should precede prompt tuning.

We establish a discrete $H^1$-Korn inequality for finite element spaces satisfying a second order weak continuity condition. This implies that classical $H^2$-nonconforming elements satisfy the discrete Korn inequality. Based on this result, we develop a stabilized Morley finite element method for linear strain gradient elasticity using only piecewise quadratic polynomials. Thanks to the discrete $H^1$-Korn inequality, this method does not require any penalty parameter. We prove that the method is robust with respect to both $\lambda$ and $\iota$. Numerical experiments confirm the theoretical results.

We present a new systematic method to design an algebraic swapping lemma observer for nonlinear systems which are affine in the state. Four are the main features of the new observer: (i) Far superior transient performance compared to standard asymptotically convergent observers; (ii) structural simplicity - its construction relies solely on basic linear filtering and a single matrix inversion; (iii) applicability to non-uniformly-completely-observable systems, for which classical methods are not applicable; (iv) straightforward extension to the cases of unknown time-varying parameters and measurement delay. The core idea exploited in the paper is to combine the construction of a generalized parameter estimation-based observer with a simple filtering technique to derive the algebraic observer. The resulting design is shown to apply to a class of state-affine systems satisfying a weak generic condition. Moreover, it is established that for time-invariant systems, this condition is always satisfied globally.

An Intraoperative Hypotension (IOH) event is a frequent complication during administration of general anaesthesia with serious downstream consequences, yet clinical management remains reactive and not predictive. Existing predictive models, however, ignore drug infusion history as a valuable signal for prediction despite its direct pharmacological relevance. Our model achieves an Area Under the Receiver Operating Characteristic curve (AUROC) of 0.7360 and an Area Under the Precision-Recall Curve (AUPRC) of 0.1794, representing a 2.73-fold lift over the random guessing AUPRC baseline (0.0657), with the removal of propofol and remifentanil effect-site concentrations resulting in a 13.9% AUPRC drop compared to the full model. This is consistent with the hypothesis that pharmacokinetic trajectories encode impending haemodynamic changes before they manifest in the Mean Arterial Pressure (MAP). Additionally, this paper shows that training without lead-gap filtering degraded AUROC by 16.7%, empirically confirming that unfiltered models learn to detect ongoing hypotension rather than predict future events. Finally, a Mamba-based architecture achieves the aforementioned high prediction performance while maintaining a constant memory footprint across a range of sequence lengths, unlike the quadratic VRAM overhead typical of vanilla Transformers, making it the more practical choice for continuous intraoperative deployment.

While Mamba-based models have shown strong potential for long sequence modeling, adapting them to vision is challenging due to the requirement of local neighborhood correlations and multi-directional spatial contexts for visual understanding. In this paper, we present LiAuto-MindViT, a novel hybrid vision backbone that synergizes the strengths of CNNs, Mamba, and Transformers. The core of our design is the Adaptive Bidirectional Mamba (ABM), which eliminates the directional bias of unidirectional SSMs through bidirectional selective scanning with learnable alpha blending, enabling content-adaptive directional fusion without the overhead of exhaustive multi-path routing. To further accelerate inference, we propose a deployment-friendly Reparameterized ConvSE (RepConvSE) module that leverages structural reparameterization to reduce latency and memory access overhead. Extensive experiments demonstrate that LiAuto-MindViT achieves state-of-the-art performance on image classification, object detection, and semantic segmentation while enabling efficient inference through reparameterization.

In a hexagonal crystal the $a$ and $c$ axes change their lengths at different rates with increasing temperature, and the thermal expansion is described by two coefficients, one for each axis. The equilibrium lattice parameters at each temperature are found by minimizing the Helmholtz free energy over the lattice parameters, and the coefficients are their logarithmic derivatives with respect to temperature. This report describes a procedure for this minimization as implemented in phonopy. The free energy is computed with harmonic force constants at a finite set of values of $a$ and $c$, and a free-energy surface over $a$ and $c$ is fitted to these values. In a variant, the harmonic force constants are replaced by temperature-dependent force constants from a self-consistent harmonic approximation, with the forces computed by a machine-learning potential fitted at each of these values. The $\alpha$ and $\omega$ phases of titanium are the worked example. Both calculations give negative thermal expansion of the $c$ axis of $\alpha$-Ti at low temperature and none in $\omega$-Ti, and they differ most for the $c$ axis of $\alpha$-Ti. The axial coefficients are harder to compute than the volumetric one, because the two axes are coupled in the free energy. In both phases of titanium the thermal effect that lengthens $a$ also shortens $c$. The coefficient of the $c$ axis is then the sum of two terms of opposite sign that partly cancel, and a small error in either term gives a large relative error in the coefficient. The report measures how much each setting of the procedure changes the thermal expansion coefficients, with the $c$ axis of $\alpha$-Ti as the most sensitive case.

Stellar flares are brief, intense increases in brightness resulting from magnetic reconnection events in stellar atmospheres. In this paper, we have developed a pipeline to detect and characterize flares in short-duration (approximately 6.2-hour) light curves from the Zwicky Transient Facility (ZTF) Extended Deep-Drilling survey. This method involves robust detrending with Tukey's biweight estimator, followed by sigma-based thresholding and visual scrutiny to identify flare candidates. It is applied to 28 million light curves, and we found 331 flaring events from 310 stars across evolutionary stages and spectral types. We estimated flare energies by assuming a blackbody emission model for 234 flares with reliable distance, and flare morphologies were classified based on flare modeling using the template by Mendoza et al.(2022). We obtained a power law correlation between energy and duration with an index of $0.19 \pm 0.01$. We also find similar power law correlation between the flare energy and $T_{decay}$, $T_{rise}$ and FWHM. This study demonstrates the power of high-cadence ground-based surveys for statistical and morphological flare studies across a broad stellar sample.

The enormous recorded data at high-energy physics (HEP) colliders make the accurate identification of physics objects a bottleneck in disentangling event topologies, where machine learning has become a standard tool for jet tagging. In this work, we examine whether Vision-Language Models (VLMs) can provide a common interface for structured jet analysis from a single physics-informed image. Each JetClass jet becomes a 224x224 RGB image with four panels encoding p_T flow of all constituents; charged-hadron, neutral-hadron, and electromagnetic composition; p_T-weighted impact-parameter significances with displaced-track multiplicity; and signed-track p_T densities with local p_T^k-weighted jet-charge asymmetry. Four open VLMs are adapted with low-rank adaptation (LoRA) for class classification across QCD, Higgs, W/Z, and top jets, six-field attribute prediction, and cross-panel consistency with replaced-panel localization, evaluated on 24000 balanced JetClass test jets. Ablations identify the impact-parameter lifetime signature of heavy-flavor tagging as dominant, with jet charge separating the nearly mass-degenerate hadronic W and Z. Zero-shot recall stays near chance (3.33%-10.57%), once adapted, Task 1 Macro recall grows monotonically with training size, and Gemma4-E4B reaches 75.05% (74.87% F_1), 93.31% Task 2 field-mean, and 99.88%/99.75% binary/localization recall. Reallocating budget toward W/Z jets raises Zqq recall by up to 9.83 points at near-constant Macro recall. Transfer to broader-topology JetClass-II and real-data Aspen Open Jets (probing the simulation-to-data gap) is retained: with 3000 target jets, Task 1 recall exceeds 70% on JetClass-II and every Task 2/3 metric exceeds 92%. Thus rasterizing energy flow, species, displacement, and jet charge, with parameter-efficient adaptation, supports jet analysis through one instruction-conditioned interface for future collider data.

We prove that the first-order theory of nonabelian free groups does not admit a model companion.