The paper introduces SLIM-init, a monocular visual‑inertial initialization method that avoids explicit 3D reconstruction by using 2D line features and vanishing points. It provides robust rotation constraints through line‑derived vanishing points and improves translation estimation with a line epipolar residual and a line‑normal projection residual. Experiments on public benchmarks and custom degenerate‑motion sequences show that SLIM-init achieves higher accuracy and robustness than existing initialization techniques.
By Junwan Choi, Woongrae Jo, Dong-Uk Seo, Jinwoo Jeon, Hyun Myung
The paper presents two optimizations for forced alignment of audio to text using the Viterbi algorithm. The first optimization applies the Hirschberg algorithm to reduce memory usage from 140 GB to 5 MB for three‑hour inputs and speeds up alignment to one‑third the time of torchaudio on a CPU. The second optimization models alignment as a constrained random walk, enabling pruning that yields an additional 2× speedup on inputs longer than 20 minutes while maintaining over 98 % alignment accuracy.
By Lawry Sorenson, Michael Crandall, Eric K. Ringger, Stephen D. Richardson
The paper presents a bi‑modal speech‑level transformer that eliminates segment‑level labeling and introduces a hierarchical attention interpretation method. By using gradient‑weighted attention maps from all attention layers, the model provides both speech‑level and sentence‑level explanations of depression detection. Experimental results show the transformer outperforms a segment‑level model (p=0.854 vs. 0.732, r=0.947 vs. 0.808, F1=0.897 vs. 0.768).
By Qingkun Deng, Saturnino Luz, Sofia de la Fuente Garcia
The paper introduces a clinician‑grounded evaluation platform called InterviewPlayground, which uses a memory‑augmented patient simulator to assess AI‑assisted psychiatric intake systems. It supports comparison across different interviewing styles, reduces clinician workload, and measures clinically relevant performance. In a pilot study, a GPT‑based intake interviewer captured more relevant items but made more unfounded inferences and missed safety concerns compared to clinicians.
By King Shi, Amanda Li, Jonathan Ivey, Synthia Qia Wang, Guan Gui, Hyunseo Kim, Peter Zandi, Jason Straub, Jacob Taylor, Ananya Joshi
SafeStyle is a training‑free framework that injects calibrated style residuals into frozen diffusion models for reference‑guided stylization. It estimates style‑supported and content‑associated subspaces from small calibration sets, then transports purified style evidence across adaptive spatial granularity while limiting its influence with a residual‑norm budget. Experiments on texture‑ and geometry‑dominant styles show high DINO style similarity (0.432–0.474) with minimal semantic leakage (0.8%).
By Zhangping Yang, Min Li, Song Yan, Rong Gao, Xinliang Bi, Guanye Xiong, Yujie He
Samsone is a family of small audio language models (SALMs) designed for on‑device inference, with the flagship Samsone‑134M setting a new state‑of‑the‑art for its size class across multiple benchmarks. The paper also presents Samsone‑99M and Samsone‑356M to study scaling laws, showing that these compact models achieve performance competitive with much larger counterparts. The authors train the models on publicly available data and release training code, weights, mobile‑optimized checkpoints, and an open‑source Android app for real‑time inference.
By Piotr Masztalski, Micha{\l} K. Grzeszczyk, Olaf Sikorski
CIPL (Channel Inversion for Privacy Leakage) is a channel-aware framework designed to evaluate black-box privacy leakage in large language model agents. It models the leakage process through stages of sensitive source, selection, assembly, execution, observation, and extraction, assessing how selected sensitive units become attacker-recoverable outputs. Experiments across memory, retrieval, and tool-mediated targets, plus a live-agent case study, reveal that recoverability depends on factors beyond storage labels, such as observation surface, prompt alignment, retrieval depth, and provider behavior, and that a semantic audit can uncover disclosures missed by exact matching.
By Tao Huang, Guosen Wu, Guolong Zheng, Jiayang Meng, Chen Hou, Xu Yang, Xuechao Yang, Feng Xia
The paper introduces FedDCN, a federated deep clustering network that jointly optimizes reconstruction and clustering losses for high‑dimensional, heterogeneous data. It addresses challenges of non‑IID client data by generating synthetic augmentations and applying geometric regularization to align latent spaces. Experiments show the method’s effectiveness under both IID and non‑IID settings, and the authors outline future research directions.
By Morris Stallmann, Charalampos S. Kouzinopoulos, Marcin Pietrasik, Anna Wilbik
The paper investigates the limits of the maximal coding rate reduction (MCR²) framework for out‑of‑distribution (OOD) generalisation. It shows that MCR² can lead to complete prediction failure under distribution shift, even when a perfectly stable feature is available, and that adding invariance principles from IRM or REx does not resolve this issue. The authors conclude that additional assumptions or learning principles are needed to guarantee stable OOD predictions with MCR².
By Menghui Zhou, Gaoshan Bi, Vitaveska Lanfranchi, Po Yang
XCalib is an unsupervised dense registration framework that aligns thermal and visible video streams by optimizing virtual pinhole camera parameters and predicted monocular depth, thereby restricting spatial displacements to physically valid projection geometries. It introduces a novel registration paradigm using camera parameterization as an implicit regularizer, a robust Normalized Edges Correlation (NEC) metric for cross‑spectral alignment, and demonstrates superior temporal stability and alignment accuracy on public ADAS datasets compared to unconstrained dense flow baselines.
By Aurelien Godet, Gabriel Jobert, Mauro Dalla Mura
The paper investigates whether machine learning models can uncover physical patterns in atomistic data without relying on traditional physics-based inductive biases such as geometric locality or graph structures. By training a general-purpose architecture on molecular simulation data, the authors demonstrate that the model autonomously learns interatomic interaction strengths resembling classical electrostatics and identifies interaction cutoffs aligned with established physical models. The study also reports predictable neural scaling behavior and competitive accuracy on certain metrics compared to physics-informed architectures, suggesting that explicit priors may only be necessary when empirically justified.
By Tobias Kreiman, Yutong Bai, Fadi Atieh, Elizabeth Weaver, Eric Qu, Aditi S. Krishnapriyan
The paper proposes a Bayesian inverse problem formulation for unsupervised anomaly detection, aiming to infer the most probable corruption causing each observation. This approach yields a probabilistic anomaly score based on the energy of inferred corruption parameters and provides a unified framework that recovers several existing methods as special cases. Experiments demonstrate improved object-class AUROC on the MVTec AD dataset and strong detection performance on a brain MRI benchmark, with additional estimates of pathology intensity, bias, and geometry.
By James Myles, Matthew Baugh, Johanna P. M\"uller, Bernhard Kainz, Yingzhen Li
Fact Grounded Attention (FGA) is a new architectural change that injects verifiable knowledge directly into the transformer’s pre‑softmax attention scores, turning large language models into deterministic truth‑tellers. By modifying the attention mechanism itself, FGA prevents hallucinations for facts that exist in its knowledge base, unlike prior methods that patch output or prepend retrieved text. Experiments on 1,107 technical queries show accuracy rising from 6.3% with vanilla Llama 3.2 to 99.7% with FGA, and knowledge updates can be applied in under one second without retraining.
By Aayush Gupta, Manish Choudhary
The paper introduces GDMRG, a Graph-Augmented Dual-Stream Medical Report Generation framework that incorporates a Topological Knowledge Internalization module using a Graph Convolutional Network to encode disease co-occurrence priors. It employs a dual-stream classifier—one branch generating diagnostic prompts under topological constraints and an auxiliary branch dynamically calibrating decision boundaries for imbalanced samples—alongside a Diagnosis-Guided Spatial Attention mechanism to align visual features with clinical semantics. Experiments on MIMIC-CXR show competitive clinical efficacy and natural language fluency, with strong zero-shot performance on IU X-Ray.
By Moyu Tang, Shangkun Sima, Chupei Tang, Junxiao Kong, Di Wang, Tianchi Lu
The study investigates how people evaluate AI-generated financial advice by conducting a randomized vignette experiment with 285 U.S. adults. Participants were presented with consistent financial recommendations delivered in three styles—AI, expert, and online community—alongside source labels. The results show that advice style most strongly influenced message and safety appraisals, expert labels increased perceived source knowledge, and decision context shaped risk and safety judgments, with these appraisals explaining a large portion of overall quality, trust, and intended reliance.
By Aryan Ramchandra Kapadia, Eshwar Chandrasekharan, Koustuv Saha
The paper extends the analysis of Joint-Embedding Predictive Architectures (JEPAs) beyond Euclidean latent spaces to Riemannian manifolds. It shows that when latent variables lie on a sphere and the target distribution matches this spherical geometry, every optimal representation recovers the latent state up to an orthogonal transformation, demonstrating that Gaussian uniqueness is not universal. Experiments confirm that geometrically compatible targets improve linear recovery, especially in high-dimensional toroidal settings.
By L\'eo Nicollier (CB, ATT), Enric Meinhardt-Llopis (CB), Marc Pic (ATT), Pablo Mus\'e (CB, IFUMI), Gabriele Facciolo (CB)
The paper introduces Riemannian Neural Hamiltonian Flows, a generative model that extends Hamiltonian normalizing flows to Riemannian manifolds by combining a fixed kinetic energy, a learned scalar potential, and a geodesic leapfrog integrator. It provides an analysis showing how the learned Hamiltonian can be interpreted through an implicit profile and a matched potential, with special cases such as isotropic Gaussian and local harmonic analysis around modes. Experiments on Euclidean, hyperbolic, and spherical spaces demonstrate competitive sample quality and computational efficiency compared to a Riemannian continuous normalizing flow, while confirming the interpretability of the learned potential.
By Vincent Souveton
The paper evaluates machine learning (ML) models for predicting critical heat flux (CHF) in square rod bundles using the CTF subchannel code and the EPRI rod bundle CHF database. It compares pure and hybrid residual correction models in local and semilocal forms, finding that tube-trained ML models transfer well to rod bundle geometries and outperform traditional empirical correlations and lookup tables. The local hybrid LUT model shows the best overall performance, while the semilocal pure ML model remains highly competitive, indicating that significant improvements in rod bundle CHF prediction are achievable even with tube-only training data.
By Aidan Furlong, Vinicius de Melo Monteiro, Robert Salko, Juliana Pacheco Duarte, Xu Wu
ServeGuard is a supply‑chain primitive that allows a publisher to ship a proof‑carrying adapter for an open‑weight language model, proving in zero‑knowledge that the adapter contains no hidden backdoor channel in the monitor’s blind subspace. The proof is inexpensive because it relies on a deterministic function of the public base model, and the served residual is the model’s own public floor. The system lets consumers or regulators verify the absence of this class of hidden channels without revealing the certified read factor or trusting the publisher.
By Dominik Dahlem, Rui Vieira
The paper develops a nonparametric kernel estimator for tangent vector field regression on a Riemannian manifold without boundary, using parallel transport to align responses before averaging. It derives uniform bias, covariance, and stochastic rates, and constructs a simultaneous confidence tube by whitening the tangent norm into a unit‑variance Gaussian field, whose Gumbel limit yields an explicit intrinsic constant. The method is validated through simulations on various manifolds and applied to reconstruct global wind data, demonstrating spatially varying uncertainty.
By Xiaotian Chang, Yangdi Jiang, Qirui Hu