# Research Papers — 2026-04 · bundle page 191 of 233 > Full-text markdown of papers #9501–9550 of 11612 published in 2026-04, newest first. AI-readable mirrors of the full 11612-paper 2026-04 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-04/page/191 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/190.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/192.md) ### Sharpness-Aware Surrogate Training for On-Sensor Spiking Neural Networks Authors: Maximilian Nicholson Published: 2026-04-06 Categories: cs.NE, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.09696v1 Spiking neural networks (SNNs) are a natural computational model for on-sensor and near-sensor vision, where event driven processors must operate under strict power budgets with hard binary spikes. However, models trained with surrogate gradients often degrade sharply when the smooth surrogate nonlinearity is replaced by a hard threshold at deployment; a surrogate-to-hard transfer gap that directly limits on-sensor accuracy. We study Sharpness-Aware Surrogate Training (SAST), which applies Sharpness-Aware Minimization (SAM) to a surrogate-forward SNN so that the training objective is smooth and the gradient is exact, and position it as one gap-reduction strategy under the tested settings rather than the only viable mechanism. Under explicit contraction assumptions we provide state-stability, input-Lipschitz, and smoothness bounds, together with a corresponding nonconvex convergence result. On two event-camera benchmarks, swap-only hard-spike accuracy improves from 65.7\% to 94.7\% on N-MNIST and from 31.8\% to 63.3\% on DVS Gesture. Under a hardware-aware inference simulation (INT8/INT4 weight quantization, fixed-point membrane potentials, discrete leak factors), SAST remains strong: on N-MNIST, hard-spike accuracy improves from 47.6\% to 96.9\% (INT8) and from 43.2\% to 81.0\% (INT4), while on DVS Gesture it improves from 25.3\% to 47.6\% (INT8) and from 26.0\% to 43.8\% (INT4). SynOps also decrease under the same hardware-aware setting, including 1734k$\rightarrow$1315k (N-MNIST, INT8) and 86221k$\rightarrow$4323k (DVS Gesture, INT8). These results suggest that SAST is a promising component in a broader toolbox for on-sensor spiking inference under the tested settings. --- ### Assessing Privacy Preservation and Utility in Online Vision-Language Models Authors: Karmesh Siddharam Chaudhari, Youxiang Zhu, Amy Feng, Xiaohui Liang, Honggang Zhang Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.09695v1 The increasing use of Online Vision Language Models (OVLMs) for processing images has introduced significant privacy risks, as individuals frequently upload images for various utilities, unaware of the potential for privacy violations. Images contain relationships that relate to Personally Identifiable Information (PII), where even seemingly harmless details can indirectly reveal sensitive information through surrounding clues. This paper explores the critical issue of PII disclosure in images uploaded to OVLMs and its implications for user privacy. We investigate how the extraction of contextual relationships from images can lead to direct (explicit) or indirect (implicit) exposure of PII, significantly compromising personal privacy. Furthermore, we propose methods to protect privacy while preserving the intended utility of the images in Vision Language Model (VLM)-based applications. Our evaluation demonstrates the efficacy of these techniques, highlighting the delicate balance between maintaining utility and protecting privacy in online image processing environments. Index Terms-Personally Identifiable Information (PII), Privacy, Utility, privacy concerns, sensitive information --- ### EDFNet: Early Fusion of Edge and Depth for Thin-Obstacle Segmentation in UAV Navigation Authors: Negar Fathi Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.09694v1 Autonomous Unmanned Aerial Vehicles (UAVs) must reliably detect thin obstacles such as wires, poles, and branches to navigate safely in real-world environments. These structures remain difficult to perceive because they occupy few pixels, often exhibit weak visual contrast, and are strongly affected by class imbalance. Existing segmentation methods primarily target coarser obstacles and do not fully exploit the complementary multimodal cues needed for thin-structure perception. We present EDFNet, a modular early-fusion segmentation framework that integrates RGB, depth, and edge information for thin-obstacle perception in cluttered aerial scenes. We evaluate EDFNet on the Drone Depth and Obstacle Segmentation (DDOS) dataset across sixteen modality-backbone configurations using U-Net and DeepLabV3 in pretrained and non-pretrained settings. The results show that early RGB-Depth-Edge fusion provides a competitive and well-balanced baseline, with the most consistent gains appearing in boundary-sensitive and recall-oriented metrics. The pretrained RGBDE U-Net achieves the best overall performance, with the highest Thin-Structure Evaluation Score (0.244), mean IoU (0.219), and boundary IoU (0.234), while maintaining competitive runtime performance (19.62 FPS) on our evaluation hardware. However, performance on the rarest ultra-thin categories remains low across all models, indicating that reliable ultra-thin segmentation is still an open challenge. Overall, these findings position early RGB-Depth-Edge fusion as a practical and modular baseline for thin-obstacle segmentation in UAV navigation. --- ### TaFall: Balance-Informed Fall Detection via Passive Thermal Sensing Authors: Chengxiao Li, Xie Zhang, Wei Zhu, Yan Jiang, Chenshu Wu Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.09693v1 Falls are a major cause of injury and mortality among older adults, yet most incidents occur in private indoor environments where monitoring must balance effectiveness with privacy. Existing privacy-preserving fall detection approaches, particularly those based on radio frequency sensing, often rely on coarse motion cues, which limits reliability in real-world deployments. We introduce TaFall, a balance-informed fall detection system based on low-cost, privacy-preserving thermal array sensing. The key insight is that TaFall models a fall as a process of balance degradation and detects falls by estimating pose-driven biomechanical balance dynamics. To enable this capability from low-resolution thermal array maps, we propose (i) an appearance-motion fusion model for robust pose reconstruction, (ii) physically grounded balance-aware learning, and (iii) pose-bridged pretraining to improve robustness. TaFall achieves a detection rate of 98.26% with a false alarm rate of 0.65% on our dataset with over 3,000 fall instances from 35 participants across diverse indoor environments. In 27 day deployments across four homes, TaFall attains an ultra-low false alarm rate of 0.00126% and a pilot bathroom study confirms robustness under moisture and thermal interference. Together, these results establish TaFall as a reliable and privacy-preserving approach to fall detection in everyday living environments. --- ### Tipiano: Cascaded Piano Hand Motion Synthesis via Fingertip Priors Authors: Joonhyung Bae, Kirak Kim, Hyeyoon Cho, Sein Lee, Yoon-Seok Choi, Hyeon Hur, Gyubin Lee, Akira Maezawa, Satoshi Obata, Jonghwa Park, Jaebum Park, Juhan Nam Published: 2026-04-06 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.09692v1 Synthesizing realistic piano hand motions requires both precision and naturalness. Physics-based methods achieve precision but produce stiff motions; data-driven models learn natural dynamics but struggle with positional accuracy. Piano motion exhibits a natural hierarchy: fingertip positions are nearly deterministic given piano geometry and fingering, while wrist and intermediate joints offer stylistic freedom. We present (OURS), a four-stage framework exploiting this hierarchy: (1) statistics-based fingertip positioning, (2) FiLM-conditioned trajectory refinement, (3) wrist estimation, and (4) STGCN-based pose synthesis. We contribute expert-annotated fingerings for the FürElise dataset (153 pieces, ~10 hours). Experiments demonstrate F1 = 0.910, substantially outperforming diffusion baselines (F1 = 0.121), with user study (N=41) confirming quality approaching motion capture. Expert evaluation by professional pianists (N=5) identified anticipatory motion as the key remaining gap, providing concrete directions for future improvement. --- ### CAGE: Bridging the Accuracy-Aesthetics Gap in Educational Diagrams via Code-Anchored Generative Enhancement Authors: Dikshant Kukreja, Kshitij Sah, Karan Goyal, Mukesh Mohania, Vikram Goyal Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.09691v1 Educational diagrams -- labeled illustrations of biological processes, chemical structures, physical systems, and mathematical concepts -- are essential cognitive tools in K-12 instruction. Yet no existing method can generate them both accurately and engagingly. Open-source diffusion models produce visually rich images but catastrophically garble text labels. Code-based generation via LLMs guarantees label correctness but yields visually flat outputs. Closed-source APIs partially bridge this gap but remain unreliable and prohibitively expensive at educational scale. We quantify this accuracy-aesthetics dilemma across all three paradigms on 400 K-12 diagram prompts, measuring both label fidelity and visual quality through complementary automated and human evaluation protocols. To resolve it, we propose CAGE (Code-Anchored Generative Enhancement): an LLM synthesizes executable code producing a structurally correct diagram, then a diffusion model conditioned on the programmatic output via ControlNet refines it into a visually polished graphic while preserving label fidelity. We also introduce EduDiagram-2K, a collection of 2,000 paired programmatic-stylized diagrams enabling this pipeline, and present proof-of-concept results and a research agenda for the multimedia community. --- ### Are We Recognizing the Jaguar or Its Background? A Diagnostic Framework for Jaguar Re-Identification Authors: Antonio Rueda-Toicen, Abigail Allen Martin, Daniil Morozov, Matin Mahmood, Alexandra Schild, Shahabeddin Dayani, Davide Panza, Gerard de Melo Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.09690v1 Jaguar re-identification (re-ID) from citizen-science imagery can look strong on standard retrieval metrics while still relying on the wrong evidence, such as background context or silhouette shape, instead of the coat pattern that defines identity. We introduce a diagnostic framework for wildlife re-ID with two axes: a leakage-controlled context ratio, background/foreground, computed from inpainted background-only versus foreground-only images, and a laterality diagnostic based on cross-flank retrieval and mirror self-similarity. To make these diagnostics measurable, we curate a Pantanal jaguar benchmark with per-pixel segmentation masks and an identity-balanced evaluation protocol. We then use representative mitigation families, ArcFace fine-tuning, anti-symmetry regularization, and Lorentz hyperbolic embeddings, as case studies under the same evaluation lens. The goal is not only to ask which model ranks best, but also what visual evidence it uses to do so. --- ### Face Density as a Proxy for Data Complexity: Quantifying the Hardness of Instance Count Authors: Abolfazl Mohammadi-Seif, Ricardo Baeza-Yates Published: 2026-04-06 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.09689v2 Machine learning progress has historically prioritized model-centric innovations, yet achievable performance is frequently capped by the intrinsic complexity of the data itself. In this work, we isolate and quantify the impact of instance density (measured by face count) as a primary driver of data complexity. Rather than simply observing that ``crowded scenes are harder,'' we rigorously control for class imbalance to measure the precise degradation caused by density alone. Controlled experiments on the WIDER FACE and Open Images datasets, restricted to exactly 1 to 18 faces per image with perfectly balanced sampling, reveal that model performance degrades monotonically with increasing face count. This trend holds across classification, regression, and detection paradigms, even when models are fully exposed to the entire density range. Furthermore, we demonstrate that models trained on low-density regimes fail to generalize to higher densities, exhibiting a systematic under-counting bias, with error rates increasing by up to 4.6x, which suggests density acts as a domain shift. These findings establish instance density as an intrinsic, quantifiable dimension of data hardness and motivate specific interventions in curriculum learning and density-stratified evaluation. --- ### Immunizing 3D Gaussian Generative Models Against Unauthorized Fine-Tuning via Attribute-Space Traps Authors: Jianwei Zhang, Sihan Cao, Chaoning Zhang, Ziming Hong, Jiaxin Huang, Pengcheng Zheng, Caiyan Qin, Wei Dong, Yang Yang, Tongliang Liu Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.09688v1 Recent large-scale generative models enable high-quality 3D synthesis. However, the public accessibility of pre-trained weights introduces a critical vulnerability. Adversaries can fine-tune these models to steal specialized knowledge acquired during pre-training, leading to intellectual property infringement. Unlike defenses for 2D images and language models, 3D generators require specialized protection due to their explicit Gaussian representations, which expose fundamental structural parameters directly to gradient-based optimization. We propose GaussLock, the first approach designed to defend 3D generative models against fine-tuning attacks. GaussLock is a lightweight parameter-space immunization framework that integrates authorized distillation with attribute-aware trap losses targeting position, scale, rotation, opacity, and color. Specifically, these traps systematically collapse spatial distributions, distort geometric shapes, align rotational axes, and suppress primitive visibility to fundamentally destroy structural integrity. By jointly optimizing these dual objectives, the distillation process preserves fidelity on authorized tasks while the embedded traps actively disrupt unauthorized reconstructions. Experiments on large-scale Gaussian models demonstrate that GaussLock effectively neutralizes unauthorized fine-tuning attacks. It substantially degrades the quality of unauthorized reconstructions, evidenced by significantly higher LPIPS and lower PSNR, while effectively maintaining performance on authorized fine-tuning. --- ### Grid2Matrix: Revealing Digital Agnosia in Vision-Language Models Authors: Yunkai Zhang, Linda Li, Yingxin Cui, Xiyuan Ruan, Zeyu Zheng, Kezhen Chen, Yi Zhang, Diji Yang Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.09687v2 Vision-Language Models (VLMs) excel on many multimodal reasoning benchmarks, but these evaluations often do not require an exhaustive readout of the image and can therefore obscure failures in faithfully capturing all visual details. We introduce Grid2Matrix (G2M), a controlled benchmark in which a model is shown a color grid and a color-to-number mapping, and must output the corresponding matrix. By varying grid size and the number of colors, G2M provides a simple way to increase visual complexity while minimizing semantic confounds. We find that VLMs exhibit a sharp early collapse in zero-shot end-to-end evaluation, failing on surprisingly small grids rather than degrading gradually as the task becomes denser. We probe the visual encoders of VLMs from two representative families and find that they preserve substantially more of the grid information than the corresponding end-to-end outputs. This suggests that the failure is not explained by visual encoding alone, but also reflects a gap between what remains recoverable from visual features and what is ultimately expressed in language. We term this gap \textit{Digital Agnosia}. Further analyses show that these errors are highly structured and depend strongly on how grid cells overlap with visual patch boundaries. We also find that common strategies such as model scaling and multimodal alignment do not fully eliminate this failure mode. We expect G2M to serve as a useful testbed for understanding where and how VLMs lose fine visual details, and for evaluating tasks where missing even small visual details can matter, such as tables, charts, forms, and GUIs. --- ### Spectral Edge Dynamics Reveal Functional Modes of Learning Authors: Yongzhong Xu Published: 2026-04-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.06256v1 Training dynamics during grokking concentrate along a small number of dominant update directions -- the spectral edge -- which reliably distinguishes grokking from non-grokking regimes. We show that standard mechanistic interpretability tools (head attribution, activation probing, sparse autoencoders) fail to capture these directions: their structure is not localized in parameter or feature space. Instead, each direction induces a structured function over the input domain, revealing low-dimensional functional modes invisible to representation-level analysis. For modular addition, all leading directions collapse to a single Fourier mode. For multiplication, the same collapse appears only in the discrete-log basis, yielding a 5.9x improvement in concentration. For subtraction, the edge spans a small multi-mode family. For $x^2+y^2$, no single harmonic basis suffices, but cross-terms of additive and multiplicative features provide a 4x variance boost, consistent with the decomposition (a+b)^2 - 2ab. Multitask training amplifies this compositional structure, with the $x^2+y^2$ spectral edge inheriting the addition circuit's characteristic frequency (2.3x concentration increase). These results suggest that training discovers low-dimensional functional modes over the input domain, whose structure depends on the algebraic symmetry of the task. These results suggest that spectral edge dynamics identify low-dimensional functional subspaces governing learning, whose representation depends on the algebraic structure of the task. Simple harmonic structure emerges only when the task admits a symmetry-adapted basis; more complex tasks require richer functional descriptions. --- ### Learning the Stellar Structure Equations via Self-supervised Physics-Informed Neural Networks Authors: Manuel Ballester, Santiago Lopez-Tapia, Seth Gossage, Patrick Koller, Philipp M. Srivastava, Ugur Demir, Yongseok Jo, Almudena P. Marquez, Christoph Wuersch, Souvik Chakraborty, Vicky Kalogera, Aggelos Katsaggelos Published: 2026-04-06 Categories: astro-ph.SR, astro-ph.GA, astro-ph.IM, cs.AI arXiv: https://arxiv.org/abs/2604.06255v1 Stellar astrophysics relies critically on accurate descriptions of the physical conditions inside stars. Traditional solvers such as \texttt{MESA} (Modules for Experiments in Stellar Astrophysics), which employ adaptive finite-difference methods, can become computationally expensive and challenging to scale for large stellar population synthesis ($>10^9$ stars). In this work, we present an self-supervised physics-informed neural network (PINN) framework that provides a mesh-free and fully differentiable approach to solving the stellar structure equations under hydrostatic and thermal equilibrium. The model takes as input the stellar boundary conditions (at the center and surface) together with the chemical composition, and learns continuous radial profiles for mass $M_r(r)$, pressure $P(r)$, density $ρ(r)$, temperature $T(r)$, and luminosity $L_r(r)$ by enforcing the governing structure equations through physics-based loss terms. To incorporate realistic microphysics, we introduce auxiliary neural networks that approximate the equation of state and opacity tables as smooth, differentiable functions of the local thermodynamic state. These surrogates replace traditional tabulated inputs and enable end-to-end training. Once trained for a given star, the model produces continuous solutions across the entire radial domain without requiring discretization or interpolation. Validation against benchmark \texttt{MESA} models across a range of stellar masses yields a Mean Relative Absolute Error of $3.06\%$ and an average $R^2$ score of $99.98\%$. To our knowledge, this is the first demonstration that the stellar structure equations can be solved in a fully self-supervised and data-free fashion employing PINNs. This work establishes a foundation for scalable, physics-informed emulation of stellar interiors and opens the door to future extensions toward time-dependent stellar evolution. --- ### SE-Enhanced ViT and BiLSTM-Based Intrusion Detection for Secure IIoT and IoMT Environments Authors: Afrah Gueriani, Hamza Kheddar, Ahmed Cherif Mazari, Seref Sagiroglu, Onur Ceran Published: 2026-04-06 Categories: cs.CR, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.06254v1 With the rapid growth of interconnected devices in Industrial and Medical Internet of Things (IIoT and MIoT) ecosystems, ensuring timely and accurate detection of cyber threats has become a critical challenge. This study presents an advanced intrusion detection framework based on a hybrid Squeeze-and-Excitation Attention Vision Transformer-Bidirectional Long Short-Term Memory (SE ViT-BiLSTM) architecture. In this design, the traditional multi-head attention mechanism of the Vision Transformer is replaced with Squeeze-and-Excitation attention, and integrated with BiLSTM layers to enhance detection accuracy and computational efficiency. The proposed model was trained and evaluated on two real-world benchmark datasets; EdgeIIoT and CICIoMT2024; both before and after data balancing using the Synthetic Minority Over-sampling Technique (SMOTE) and RandomOverSampler. Experimental results demonstrate that the SE ViT-BiLSTM model outperforms existing approaches across multiple metrics. Before balancing, the model achieved accuracies of 99.11% (FPR: 0.0013%, latency: 0.00032 sec/inst) on EdgeIIoT and 96.10% (FPR: 0.0036%, latency: 0.00053 sec/inst) on CICIoMT2024. After balancing, performance further improved, reaching 99.33% accuracy with 0.00035 sec/inst latency on EdgeIIoT and 98.16% accuracy with 0.00014 sec/inst latency on CICIoMT2024. --- ### FLeX: Fourier-based Low-rank EXpansion for multilingual transfer Authors: Gaurav Narasimhan Published: 2026-04-06 Categories: cs.LG, cs.AI, cs.PL arXiv: https://arxiv.org/abs/2604.06253v1 Cross-lingual code generation is critical in enterprise environments where multiple programming languages coexist. However, fine-tuning large language models (LLMs) individually for each language is computationally prohibitive. This paper investigates whether parameter-efficient fine-tuning methods and optimizer enhancements can improve cross-lingual transfer from Python to languages like Java. We fine-tune the Code Llama 7B model using low-rank adaptation (LoRA) to optimize a small subset of parameters and compare Adam and Sophia optimizers, while exploring a novel Fourier-based regularization technique. Our contributions include: (1)demonstrating that LoRA fine-tuning on a small, high-quality dataset (MBPP) can exceed the pass@1 performance of the more broadly fine-tuned Code Llama-Python-7B model (40.1% vs. 38.4%); (2) showing that while Sophia achieves faster convergence than Adam, final pass@1 scores show marginal differences; and (3) presenting evidence that Fourier-based regularization during fine-tuning significantly improves cross-lingual transfer, achieving 42.1% pass@1 on Java tasks compared to the 34.2% baseline. These findings suggest that combining LoRA, optimized training methods, and frequency-domain regularization can efficiently adapt single-language LLMs to perform well across multiple programming languages. --- ### Toward Reducing Unproductive Container Moves: Predicting Service Requirements and Dwell Times Authors: Elena Villalobos, Adolfo De Unánue T., Fernanda Sobrino, David Aké, Stephany Cisneros, Jorge Lecona, Alejandra Matadamaz Published: 2026-04-06 Categories: cs.AI, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2604.06251v1 This article presents the results of a data science study conducted at a container terminal, aimed at reducing unproductive container moves through the prediction of service requirements and container dwell times. We develop and evaluate machine learning models that leverage historical operational data to anticipate which containers will require pre-clearance handling services prior to cargo release and to estimate how long they are expected to remain in the terminal. As part of the data preparation process, we implement a classification system for cargo descriptions and perform deduplication of consignee records to improve data consistency and feature quality. These predictive capabilities provide valuable inputs for strategic planning and resource allocation in yard operations. Across multiple temporal validation periods, the proposed models consistently outperform existing rule-based heuristics and random baselines in precision and recall. These results demonstrate the practical value of predictive analytics for improving operational efficiency and supporting data-driven decision-making in container terminal logistics. --- ### DISSECT: Diagnosing Where Vision Ends and Language Priors Begin in Scientific VLMs Authors: Dikshant Kukreja, Kshitij Sah, Karan Goyal, Mukesh Mohania, Vikram Goyal Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.06250v1 When asked to describe a molecular diagram, a Vision-Language Model correctly identifies ``a benzene ring with an -OH group.'' When asked to reason about the same image, it answers incorrectly. The model can see but it cannot think about what it sees. We term this the perception-integration gap: a failure where visual information is successfully extracted but lost during downstream reasoning, invisible to single-configuration benchmarks that conflate perception with integration under one accuracy number. To systematically expose such failures, we introduce DISSECT, a 12,000-question diagnostic benchmark spanning Chemistry (7,000) and Biology (5,000). Every question is evaluated under five input modes -- Vision+Text, Text-Only, Vision-Only, Human Oracle, and a novel Model Oracle in which the VLM first verbalizes the image and then reasons from its own description -- yielding diagnostic gaps that decompose performance into language-prior exploitation, visual extraction, perception fidelity, and integration effectiveness. Evaluating 18~VLMs, we find that: (1) Chemistry exhibits substantially lower language-prior exploitability than Biology, confirming molecular visual content as a harder test of genuine visual reasoning; (2) Open-source models consistently score higher when reasoning from their own verbalized descriptions than from raw images, exposing a systematic integration bottleneck; and (3) Closed-source models show no such gap, indicating that bridging perception and integration is the frontier separating open-source from closed-source multimodal capability. The Model Oracle protocol is both model and benchmark agnostic, applicable post-hoc to any VLM evaluation to diagnose integration failures. --- ### SALLIE: Generation-Free Hidden-State Detection of Jailbreaks and Prompt Injections Across Text and Vision Authors: Guy Azov, Ofer Rivlin, Guy Shtar Published: 2026-04-06 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2604.06247v2 Large Language Models (LLMs) and Vision-Language Models (VLMs) are vulnerable to jailbreaks and prompt injections delivered through text or images. Existing defenses often narrow threat coverage or add inference cost through input transformations, repeated model queries, generated responses, or threat-specific detectors. We present SALLIE (Single-Pass Activation Lookup for Layerwise Input Evaluation), a white-box, generation-free detector with a common architecture for textual and visual jailbreaks and prompt injections. Using one backbone forward pass, SALLIE extracts last-token residual-stream representations, applies layer-wise k-nearest-neighbor probes, and averages their scores across a contiguous layer range, without modifying the backbone or generating a response. To distinguish reusable detector configurations from deployment-specific tuning, we compare shared, threshold-only, and fully calibrated regimes across backbones and modalities. We evaluate SALLIE on compact open-weight VLMs (Gemma-3-4b-it, Phi-3.5-vision-instruct, SmolVLM2-2.2B-Instruct) using heterogeneous held-out test data spanning both threat families and modalities. Fully calibrated SALLIE-Phi achieves an aggregate F1 of 0.87 and higher aggregate balanced accuracy than RCS-KCD on all three backbones under a matched-calibration protocol. On Phi-3.5 visual inputs, threshold-only calibration yields an F1 of 0.99 with zero observed false positives, versus 0.54 for zero-shot Gemini-2.5-Flash-Lite. On Gemma-4-E2B-it, a backbone held out from representation-hyperparameter selection, the same procedure raises F1 from 0.62 to 0.79 on text and 0.68 to 0.98 on visual inputs. Representation hyperparameters can thus transfer across some backbones, while operating points remain backbone-, modality-, and distribution-dependent. We further report FPR/FNR tradeoffs, latency, and representation ablations. --- ### SALLIE: Safeguarding Against Latent Language & Image Exploits Authors: Guy Azov, Ofer Rivlin, Guy Shtar Published: 2026-04-06 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2604.06247v1 Large Language Models (LLMs) and Vision-Language Models (VLMs) remain highly vulnerable to textual and visual jailbreaks, as well as prompt injections (arXiv:2307.15043, Greshake et al., 2023, arXiv:2306.13213). Existing defenses often degrade performance through complex input transformations or treat multimodal threats as isolated problems (arXiv:2309.00614, arXiv:2310.03684, Zhang et al., 2025). To address the critical gap for a unified, modal-agnostic defense that mitigates both textual and visual threats simultaneously without degrading performance or requiring architectural modifications, we introduce SALLIE (Safeguarding Against Latent Language & Image Exploits), a lightweight runtime detection framework rooted in mechanistic interpretability (Lindsey et al., 2025, Ameisen et al., 2025). By integrating seamlessly into standard token-level fusion pipelines (arXiv:2306.13549), SALLIE extracts robust signals directly from the model's internal activations. At inference, SALLIE defends via a three-stage architecture: (1) extracting internal residual stream activations, (2) calculating layer-wise maliciousness scores using a K-Nearest Neighbors (k-NN) classifier, and (3) aggregating these predictions via a layer ensemble module. We evaluate SALLIE on compact, open-source architectures - Phi-3.5-vision-instruct (arXiv:2404.14219), SmolVLM2-2.2B-Instruct (arXiv:2504.05299), and gemma-3-4b-it (arXiv:2503.19786) - prioritized for practical inference times and real-world deployment costs. Our comprehensive evaluation pipeline spans over ten datasets and more than five strong baseline methods from the literature, and SALLIE consistently outperforms these baselines across a wide range of experimental settings. --- ### No-reference based automatic parameter optimization for iterative reconstruction using a novel search space aware crow search algorithm Authors: Poorya MohammadiNasab, Ander Biguri, Philipp Steininger, Peter Keuschnigg, Lukas Lamminger, Agnieszka Lach, S M Ragib Shahriar Islam, Anna Breger, Clemens Karner, Carola-Bibiane Schönlieb, Wolfgang Birkfellner, Sepideh Hatamikia Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.06246v1 Iterative reconstruction technique's ability to reduce radiation exposure by using fewer projections has attracted significant attention. However, these methods typically require a precise tuning of several hyperparameters, which can have a major impact on reconstruction quality. Manually setting these parameters is time-consuming and increases the workload for human operators. In this paper, we introduce a novel fully automatic parameter optimization framework that can be applied to a wide range of Cone-beam computed tomography (CBCT) iterative reconstruction algorithms to determine optimal parameters without requiring a reference reconstruction. The proposed method incorporates a modified crow search algorithm (CSA) featuring a superior set-dependent local search mechanism, a search-space-aware global search strategy, and an objective-driven balance between local and global search. Additionally, to ensure an effective initial population, we propose a chaotic diagonal linear uniform initialization scheme that accelerates algorithm convergence. The performance of the proposed framework was evaluated on three imaging machines and four real datasets, as well as three different iterative reconstruction methods with the highest number of tunable parameters, representing the most challenging senario. The results indicate that the proposed method could outperform manual settings and CSA, with an 4.19% improvement in average fitness and 4.89% and 3.82% improvements on CHILL@UK and RPI_AXIS, respectively, which are two benchmark no-reference learning-based quality metrics. In addition, the qualitative results clearly show the superiority of the proposed method by maintaining fine details sharply. The overall performance of the proposed framework across different comparison scenarios demonstrates its effectiveness and robustness across all cases. --- ### CraterBench-R: Instance-Level Crater Retrieval for Planetary Scale Authors: Jichao Fang, Lei Zhang, Michael Phillips, Wei Luo Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.06245v1 Impact craters are a cornerstone of planetary surface analysis. However, while most deep learning pipelines treat craters solely as a detection problem, critical scientific workflows such as catalog deduplication, cross-observation matching, and morphological analog discovery are inherently retrieval tasks. To address this, we formulate crater analysis as an instance-level image retrieval problem and introduce CraterBench-R, a curated benchmark featuring about 25,000 crater identities with multi-scale gallery views and manually verified queries spanning diverse scales and contexts. Our baseline evaluations across various architectures reveal that self-supervised Vision Transformers (ViTs), particularly those with in-domain pretraining, dominate the task, outperforming generic models with significantly more parameters. Furthermore, we demonstrate that retaining multiple ViT patch tokens for late-interaction matching dramatically improves accuracy over standard single-vector pooling. However, storing all tokens per image is operationally inefficient at a planetary scale. To close this efficiency gap, we propose instance-token aggregation, a scalable, training-free method that selects K seed tokens, assigns the remaining tokens to these seeds via cosine similarity, and aggregates each cluster into a single representative token. This approach yields substantial gains: at K=16, aggregation improves mAP by 17.9 points over raw token selection, and at K=64, it matches the accuracy of using all 196 tokens with significantly less storage. Finally, we demonstrate that a practical two-stage pipeline, with single-vector shortlisting followed by instance-token reranking, recovers 89-94% of the full late-interaction accuracy while searching only a small candidate set. The benchmark is publicly available at hf.co/datasets/jfang/CraterBench-R. --- ### Coverage Optimization for Camera View Selection Authors: Timothy Chen, Adam Dai, Maximilian Adang, Grace Gao, Mac Schwager Published: 2026-04-06 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2604.05259v1 What makes a good viewpoint? The quality of the data used to learn 3D reconstructions is crucial for enabling efficient and accurate scene modeling. We study the active view selection problem and develop a principled analysis that yields a simple and interpretable criterion for selecting informative camera poses. Our key insight is that informative views can be obtained by minimizing a tractable approximation of the Fisher Information Gain, which reduces to favoring viewpoints that cover geometry that has been insufficiently observed by past cameras. This leads to a lightweight coverage-based view selection metric that avoids expensive transmittance estimation and is robust to noise and training dynamics. We call this metric COVER (Camera Optimization for View Exploration and Reconstruction). We integrate our method into the Nerfstudio framework and evaluate it on real datasets within fixed and embodied data acquisition scenarios. Across multiple datasets and radiance-field baselines, our method consistently improves reconstruction quality compared to state-of-the-art active view selection methods. Additional visualizations and our Nerfstudio package can be found at https://chengine.github.io/nbv_gym/. --- ### Extending Tabular Denoising Diffusion Probabilistic Models for Time-Series Data Generation Authors: Umang Dobhal, Christina Garcia, Sozo Inoue Published: 2026-04-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.05257v1 Diffusion models are increasingly being utilised to create synthetic tabular and time series data for privacy-preserving augmentation. Tabular Denoising Diffusion Probabilistic Models (TabDDPM) generate high-quality synthetic data from heterogeneous tabular datasets but assume independence between samples, limiting their applicability to time-series domains where temporal dependencies are critical. To address this, we propose a temporal extension of TabDDPM, introducing sequence awareness through the use of lightweight temporal adapters and context-aware embedding modules. By reformulating sensor data into windowed sequences and explicitly modeling temporal context via timestep embeddings, conditional activity labels, and observed/missing masks, our approach enables the generation of temporally coherent synthetic sequences. Compared to baseline and interpolation techniques, validation using bigram transition matrices and autocorrelation analysis shows enhanced temporal realism, diversity, and coherence. On the WISDM accelerometer dataset, the suggested system produces synthetic time-series that closely resemble real world sensor patterns and achieves comparable classification performance (macro F1-score 0.64, accuracy 0.71). This is especially advantageous for minority class representation and preserving statistical alignment with real distributions. These developments demonstrate that diffusion based models provide effective and adaptable solutions for sequential data synthesis when they are equipped for temporal reasoning. Future work will explore scaling to longer sequences and integrating stronger temporal architectures. --- ### Protecting and Preserving Protest Dynamics for Responsible Analysis Authors: Cohen Archbold, Usman Hassan, Nazmus Sakib, Sen-ching Cheung, Abdullah-Al-Zubaer Imran Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05256v1 Protest-related social media data are valuable for understanding collective action but inherently high-risk due to concerns surrounding surveillance, repression, and individual privacy. Contemporary AI systems can identify individuals, infer sensitive attributes, and cross-reference visual information across platforms, enabling surveillance that poses risks to protesters and bystanders. In such contexts, large foundation models trained on protest imagery risk memorizing and disclosing sensitive information, leading to cross-platform identity leakage and retroactive participant identification. Existing approaches to automated protest analysis do not provide a holistic pipeline that integrates privacy risk assessment, downstream analysis, and fairness considerations. To address this gap, we propose a responsible computing framework for analyzing collective protest dynamics while reducing risks to individual privacy. Our framework replaces sensitive protest imagery with well-labeled synthetic reproductions using conditional image synthesis, enabling analysis of collective patterns without direct exposure of identifiable individuals. We demonstrate that our approach produces realistic and diverse synthetic imagery while balancing downstream analytical utility with reductions in privacy risk. We further assess demographic fairness in the generated data, examining whether synthetic representations disproportionately affect specific subgroups. Rather than offering absolute privacy guarantees, our method adopts a pragmatic, harm-mitigating approach that enables socially sensitive analysis while acknowledging residual risks. --- ### EAGLE: Edge-Aware Graph Learning for Proactive Delivery Delay Prediction in Smart Logistics Networks Authors: Zhiming Xue, Menghao Huo, Yujue Wang Published: 2026-04-06 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05254v1 Modern logistics networks generate rich operational data streams at every warehouse node and transportation lane -- from order timestamps and routing records to shipping manifests -- yet predicting delivery delays remains predominantly reactive. Existing predictive approaches typically treat this problem either as a tabular classification task, ignoring network topology, or as a time-series anomaly detection task, overlooking the spatial dependencies of the supply chain graph. To bridge this gap, we propose a hybrid deep learning framework for proactive supply chain risk management. The proposed method jointly models temporal order-flow dynamics via a lightweight Transformer patch encoder and inter-hub relational dependencies through an Edge-Aware Graph Attention Network (E-GAT), optimized via a multi-task learning objective. Evaluated on the real-world DataCo Smart Supply Chain dataset, our framework achieves consistent improvements over baseline methods, yielding an F1-score of 0.8762 and an AUC-ROC of 0.9773. Across four independent random seeds, the framework exhibits a cross-seed F1 standard deviation of only 0.0089 -- a 3.8 times improvement over the best ablated variant -- achieving the strongest balance of predictive accuracy and training stability among all evaluated models. --- ### Spike Hijacking in Late-Interaction Retrieval Authors: Karthik Suresh, Tushar Vatsa, Tracy King, Asim Kadav, Michael Friedrich Published: 2026-04-06 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2604.05253v1 Late-interaction retrieval models rely on hard maximum similarity (MaxSim) to aggregate token-level similarities. Although effective, this winner-take-all pooling rule may structurally bias training dynamics. We provide a mechanistic study of gradient routing and robustness in MaxSim-based retrieval. In a controlled synthetic environment with in-batch contrastive training, we demonstrate that MaxSim induces significantly higher patch-level gradient concentration than smoother alternatives such as Top-k pooling and softmax aggregation. While sparse routing can improve early discrimination, it also increases sensitivity to document length: as the number of document patches grows, MaxSim degrades more sharply than mild smoothing variants. We corroborate these findings on a real-world multi-vector retrieval benchmark, where controlled document-length sweeps reveal similar brittleness under hard max pooling. Together, our results isolate pooling-induced gradient concentration as a structural property of late-interaction retrieval and highlight a sparsity-robustness tradeoff. These findings motivate principled alternatives to hard max pooling in multi-vector retrieval systems. --- ### DualDiffusion: A Speculative Decoding Strategy for Masked Diffusion Models Authors: Satyam Goyal, Kushal Patel, Tanush Mittal, Arjun Laxman Published: 2026-04-06 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2604.05250v1 Masked Diffusion Models (MDMs) offer a promising alternative to autoregressive language models by enabling parallel token generation and bidirectional context modeling. However, their inference speed is significantly limited by the inability to cache key-value pairs due to bidirectional attention, requiring $O(N^2)$ computations at each generation step. While recent methods like FastDLLM and DkvCache improve inference speed through attention approximations and caching strategies, they achieve speedups at the cost of generation quality. We propose DualDiffusion, a speculative decoding framework for MDMs that combines fast drafter models (using efficient approximations) with slower, more accurate verifier models. By running multiple steps of a lightweight drafter followed by a single verification step, DualDiffusion achieves a superior Pareto frontier between generation steps and accuracy compared to existing approaches. We evaluate our method on MMLU and GSM8K, demonstrating that DualDiffusion maintains high accuracy while reducing the number of generation steps required, effectively pushing the quality-efficiency trade-off curve for masked diffusion language models. --- ### Improving Sparse Memory Finetuning Authors: Satyam Goyal, Anirudh Kanchi, Garv Shah, Prakhar Gupta Published: 2026-04-06 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2604.05248v1 Large Language Models (LLMs) are typically static after training, yet real-world applications require continual adaptation to new knowledge without degrading existing capabilities. Standard approaches to updating models, like full finetuning or parameter-efficient methods (e.g., LoRA), face a fundamental trade-off: catastrophic forgetting. They modify shared dense representations, causing interference across tasks. Sparse Memory Finetuning (SMF) offers a promising alternative by localizing updates to a small subset of parameters in explicit memory layers. In this work, we present an open-source pipeline to retrofit existing pretrained models (Qwen-2.5-0.5B) with sparse memory modules, enabling effective continual learning on consumer hardware. We extend prior work by introducing a theoretically grounded slot-selection mechanism based on Kullback-Leibler (KL) divergence, which prioritizes memory updates for informationally "surprising" tokens relative to a background distribution. Our experiments demonstrate that our retrofitted models can acquire new factual knowledge with minimal forgetting of held-out capabilities, validating the sparse update hypothesis in a practical setting. --- ### Exemplar Retrieval Without Overhypothesis Induction: Limits of Distributional Sequence Learning in Early Word Learning Authors: Jon-Paul Cacioli Published: 2026-04-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05243v1 Background: Children do not simply learn that balls are round and blocks are square. They learn that shape is the kind of feature that tends to define object categories -- a second-order generalisation known as an overhypothesis (1, 2). What kind of learning mechanism is sufficient for this inductive leap? Methods: We trained autoregressive transformer language models (3.4M-25.6M parameters) on synthetic corpora in which shape is the stable feature dimension across categories, with eight conditions controlling for alternative explanations. Results: Across 120 pre-registered runs evaluated on a 1,040-item wug test battery, every model achieved perfect first-order exemplar retrieval (100%) while second-order generalisation to novel nouns remained at chance (50-52%), a result confirmed by equivalence testing. A feature-swap diagnostic revealed that models rely on frame-to-feature template matching rather than structured noun-to-domain-to-feature abstraction. Conclusions: These results reveal a clear limitation of autoregressive distributional sequence learning under developmental-scale training conditions. --- ### XMark: Reliable Multi-Bit Watermarking for LLM-Generated Texts Authors: Jiahao Xu, Rui Hu, Olivera Kotevska, Zikai Zhang Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2604.05242v2 Multi-bit watermarking has emerged as a promising solution for embedding imperceptible binary messages into Large Language Model (LLM)-generated text, enabling reliable attribution and tracing of malicious usage of LLMs. Despite recent progress, existing methods still face key limitations: some become computationally infeasible for large messages, while others suffer from a poor trade-off between text quality and decoding accuracy. Moreover, the decoding accuracy of existing methods drops significantly when the number of tokens in the generated text is limited, a condition that frequently arises in practical usage. To address these challenges, we propose \textsc{XMark}, a novel method for encoding and decoding binary messages in LLM-generated texts. The unique design of \textsc{XMark}'s encoder produces a less distorted logit distribution for watermarked token generation, preserving text quality, and also enables its tailored decoder to reliably recover the encoded message with limited tokens. Extensive experiments across diverse downstream tasks show that \textsc{XMark} significantly improves decoding accuracy while preserving the quality of watermarked text, outperforming prior methods. The code is at https://github.com/JiiahaoXU/XMark. --- ### Curvature-Aware Optimization for High-Accuracy Physics-Informed Neural Networks Authors: Anas Jnini, Elham Kiyani, Khemraj Shukla, Jorge F. Urban, Nazanin Ahmadi Daryakenari, Johannes Muller, Marius Zeinhofer, George Em Karniadakis Published: 2026-04-06 Categories: cs.LG, cs.AI, math.NA, math.OC arXiv: https://arxiv.org/abs/2604.05230v1 Efficient and robust optimization is essential for neural networks, enabling scientific machine learning models to converge rapidly to very high accuracy -- faithfully capturing complex physical behavior governed by differential equations. In this work, we present advanced optimization strategies to accelerate the convergence of physics-informed neural networks (PINNs) for challenging partial (PDEs) and ordinary differential equations (ODEs). Specifically, we provide efficient implementations of the Natural Gradient (NG) optimizer, Self-Scaling BFGS and Broyden optimizers, and demonstrate their performance on problems including the Helmholtz equation, Stokes flow, inviscid Burgers equation, Euler equations for high-speed flows, and stiff ODEs arising in pharmacokinetics and pharmacodynamics. Beyond optimizer development, we also propose new PINN-based methods for solving the inviscid Burgers and Euler equations, and compare the resulting solutions against high-order numerical methods to provide a rigorous and fair assessment. Finally, we address the challenge of scaling these quasi-Newton optimizers for batched training, enabling efficient and scalable solutions for large data-driven problems. --- ### From Governance Norms to Enforceable Controls: A Layered Translation Method for Runtime Guardrails in Agentic AI Authors: Christopher Koch Published: 2026-04-06 Categories: cs.AI, cs.HC, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2604.05229v1 Agentic AI systems plan, use tools, maintain state, and produce multi-step trajectories with external effects. Those properties create a governance problem that differs materially from single-turn generative AI: important risks emerge dur- ing execution, not only at model development or deployment time. Governance standards such as ISO/IEC 42001, ISO/IEC 23894, ISO/IEC 42005, ISO/IEC 5338, ISO/IEC 38507, and the NIST AI Risk Management Framework are therefore highly relevant to agentic AI, but they do not by themselves yield implementable runtime guardrails. This paper proposes a layered translation method that connects standards-derived governance objectives to four control layers: governance objectives, design- time constraints, runtime mediation, and assurance feedback. It distinguishes governance objectives, technical controls, runtime guardrails, and assurance evidence; introduces a control tuple and runtime-enforceability rubric for layer assignment; and demonstrates the method in a procurement-agent case study. The central claim is modest: standards should guide control placement across architecture, runtime policy, human escalation, and audit, while runtime guardrails are reserved for controls that are observable, determinate, and time-sensitive enough to justify execution-time intervention. --- ### Active Measurement of Two-Point Correlations Authors: Max Hamilton, Daniel Sheldon, Subhransu Maji Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05227v1 Two-point correlation functions (2PCF) are widely used to characterize how points cluster in space. In this work, we study the problem of measuring the 2PCF over a large set of points, restricted to a subset satisfying a property of interest. An example comes from astronomy, where scientists measure the 2PCF of star clusters, which make up only a tiny subset of possible sources within a galaxy. This task typically requires careful labeling of sources to construct catalogs, which is time-consuming. We present a human-in-the-loop framework for efficient estimation of 2PCF of target sources. By leveraging a pre-trained classifier to guide sampling, our approach adaptively selects the most informative points for human annotation. After each annotation, it produces unbiased estimates of pair counts across multiple distance bins simultaneously. Compared to simple Monte Carlo approaches, our method achieves substantially lower variance while significantly reducing annotation effort. We introduce a novel unbiased estimator, sampling strategy, and confidence interval construction that together enable scalable and statistically grounded measurement of two-point correlations in astronomy datasets. --- ### RoboPlayground: Democratizing Robotic Evaluation through Structured Physical Domains Authors: Yi Ru Wang, Carter Ung, Evan Gubarev, Christopher Tan, Siddhartha Srinivasa, Dieter Fox Published: 2026-04-06 Categories: cs.RO, cs.AI, cs.CL, cs.HC arXiv: https://arxiv.org/abs/2604.05226v1 Evaluation of robotic manipulation systems has largely relied on fixed benchmarks authored by a small number of experts, where task instances, constraints, and success criteria are predefined and difficult to extend. This paradigm limits who can shape evaluation and obscures how policies respond to user-authored variations in task intent, constraints, and notions of success. We argue that evaluating modern manipulation policies requires reframing evaluation as a language-driven process over structured physical domains. We present RoboPlayground, a framework that enables users to author executable manipulation tasks using natural language within a structured physical domain. Natural language instructions are compiled into reproducible task specifications with explicit asset definitions, initialization distributions, and success predicates. Each instruction defines a structured family of related tasks, enabling controlled semantic and behavioral variation while preserving executability and comparability. We instantiate RoboPlayground in a structured block manipulation domain and evaluate it along three axes. A user study shows that the language-driven interface is easier to use and imposes lower cognitive workload than programming-based and code-assist baselines. Evaluating learned policies on language-defined task families reveals generalization failures that are not apparent under fixed benchmark evaluations. Finally, we show that task diversity scales with contributor diversity rather than task count alone, enabling evaluation spaces to grow continuously through crowd-authored contributions. Project Page: https://roboplayground.github.io --- ### fastml: Guarded Resampling Workflows for Safer Automated Machine Learning in R Authors: Selcuk Korkmaz, Dincer Goksuluk, Eda Karaismailoglu Published: 2026-04-06 Categories: stat.CO, cs.LG, stat.AP, stat.ML arXiv: https://arxiv.org/abs/2604.05225v1 Preprocessing leakage arises when scaling, imputation, or other data-dependent transformations are estimated before resampling, inflating apparent performance while remaining hard to detect. We present fastml, an R package that provides a single-call interface for leakage-aware machine learning through guarded resampling, where preprocessing is re-estimated inside each resample and applied to the corresponding assessment data. The package supports grouped and time-ordered resampling, blocks high-risk configurations, audits recipes for external dependencies, and includes sandboxed execution and integrated model explanation. We evaluate fastml with a Monte Carlo simulation contrasting global and fold-local normalization, a usability comparison with tidymodels under matched specifications, and survival benchmarks across datasets of different sizes. The simulation demonstrates that global preprocessing substantially inflates apparent performance relative to guarded resampling. fastml matched held-out performance obtained with tidymodels while reducing workflow orchestration, and it supported consistent benchmarking of multiple survival model classes through a unified interface. --- ### Attribution Bias in Large Language Models Authors: Eliza Berman, Bella Chang, Daniel B. Neill, Emily Black Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05224v2 As Large Language Models (LLMs) are increasingly used to support search and information retrieval, it is critical that they accurately attribute content to its original authors. In this work, we introduce AttriBench, the first fame- and demographically-balanced quote attribution benchmark dataset. By explicitly balancing author fame and demographics, AttriBench enables controlled investigation of demographic bias in quote attribution. Using this dataset, we evaluate 11 widely used LLMs across different prompt settings and find that quote attribution remains a challenging task even for frontier models. We observe large and systematic disparities in attribution accuracy between race, gender, and intersectional groups. We further introduce and investigate suppression, a distinct failure mode in which models omit attribution entirely, even when the model has access to authorship information. We find that suppression is widespread and unevenly distributed across demographic groups, revealing systematic biases not captured by standard accuracy metrics. Our results position quote attribution as a benchmark for representational fairness in LLMs. --- ### Attribution Bias in Large Language Models Authors: Eliza Berman, Bella Chang, Daniel B. Neill, Emily Black Published: 2026-04-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05224v1 As Large Language Models (LLMs) are increasingly used to support search and information retrieval, it is critical that they accurately attribute content to its original authors. In this work, we introduce AttriBench, the first fame- and demographically-balanced quote attribution benchmark dataset. Through explicitly balancing author fame and demographics, AttriBench enables controlled investigation of demographic bias in quote attribution. Using this dataset, we evaluate 11 widely used LLMs across different prompt settings and find that quote attribution remains a challenging task even for frontier models. We observe large and systematic disparities in attribution accuracy between race, gender, and intersectional groups. We further introduce and investigate suppression, a distinct failure mode in which models omit attribution entirely, even when the model has access to authorship information. We find that suppression is widespread and unevenly distributed across demographic groups, revealing systematic biases not captured by standard accuracy metrics. Our results position quote attribution as a benchmark for representational fairness in LLMs. --- ### On the Geometry of Positional Encodings in Transformers Authors: Giansalvo Cirrincione Published: 2026-04-06 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2604.05217v1 Neural language models process sequences of words, but the mathematical operations inside them are insensitive to the order in which words appear. Positional encodings are the component added to remedy this. Despite their importance, positional encodings have been designed largely by trial and error, without a mathematical theory of what they ought to do. This paper develops such a theory. Four results are established. First, any Transformer without a positional signal cannot solve any task sensitive to word order (Necessity Theorem). Second, training assigns distinct vector representations to distinct sequence positions at every global minimiser, under mild and verifiable conditions (Positional Separation Theorem). Third, the best achievable approximation to an information-optimal encoding is constructed via classical multidimensional scaling (MDS) on the Hellinger distance between positional distributions; the quality of any encoding is measured by a single number, the stress (Proposition 5, Algorithm 1). Fourth, the optimal encoding has effective rank r = rank(B) <= n-1 and can be represented with r(n+d) parameters instead of nd (minimal parametrisation result). Appendix A develops a proof of the Monotonicity Conjecture within the Neural Tangent Kernel (NTK) regime for masked language modelling (MLM) losses, sequence classification losses, and general losses satisfying a positional sufficiency condition, through five lemmas. Experiments on SST-2 and IMDB with BERT-base confirm the theoretical predictions and reveal that Attention with Linear Biases (ALiBi) achieves much lower stress than the sinusoidal encoding and Rotary Position Embedding (RoPE), consistent with a rank-1 interpretation of the MDS encoding under approximate shift-equivariance. --- ### Hierarchical Mesh Transformers with Topology-Guided Pretraining for Morphometric Analysis of Brain Structures Authors: Yujian Xiong, Mohammad Farazi, Yanxi Chen, Wenhui Zhu, Xuanzhao Dong, Natasha Lepore, Yi Su, Raza Mushtaq, Stephen Foldes, Andrew Yang, Yalin Wang Published: 2026-04-06 Categories: cs.CV, q-bio.NC arXiv: https://arxiv.org/abs/2604.05215v1 Representation learning on large-scale unstructured volumetric and surface meshes poses significant challenges in neuroimaging, especially when models must incorporate diverse vertex-level morphometric descriptors, such as cortical thickness, curvature, sulcal depth, and myelin content, which carry subtle disease-related signals. Current approaches either ignore these clinically informative features or support only a single mesh topology, restricting their use across imaging pipelines. We introduce a hierarchical transformer framework designed for heterogeneous mesh analysis that operates on spatially adaptive tree partitions constructed from simplicial complexes of arbitrary order. This design accommodates both volumetric and surface discretizations within a single architecture, enabling efficient multi-scale attention without topology-specific modifications. A feature projection module maps variable-length per-vertex clinical descriptors into the spatial hierarchy, separating geometric structure from feature dimensionality and allowing seamless integration of different neuroimaging feature sets. Self-supervised pretraining via masked reconstruction of both coordinates and morphometric channels on large unlabeled cohorts yields a transferable encoder backbone applicable to diverse downstream tasks and mesh modalities. We validate our approach on Alzheimer's disease classification and amyloid burden prediction using volumetric brain meshes from ADNI, as well as focal cortical dysplasia detection on cortical surface meshes from the MELD dataset, achieving state-of-the-art results across all benchmarks. --- ### Boxer: Robust Lifting of Open-World 2D Bounding Boxes to 3D Authors: Daniel DeTone, Tianwei Shen, Fan Zhang, Lingni Ma, Julian Straub, Richard Newcombe, Jakob Engel Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05212v1 Detecting and localizing objects in space is a fundamental computer vision problem. While much progress has been made to solve 2D object detection, 3D object localization is much less explored and far from solved, especially for open-world categories. To address this research challenge, we propose Boxer, an algorithm to estimate static 3D bounding boxes (3DBBs) from 2D open-vocabulary object detections, posed images and optional depth either represented as a sparse point cloud or dense depth. At its core is BoxerNet, a transformer-based network which lifts 2D bounding box (2DBB) proposals into 3D, followed by multi-view fusion and geometric filtering to produce globally consistent de-duplicated 3DBBs in metric world space. Boxer leverages the power of existing 2DBB detection algorithms (e.g. DETIC, OWLv2, SAM3) to localize objects in 2D. This allows the main BoxerNet model to focus on lifting to 3D rather than detecting, ultimately reducing the demand for costly annotated 3DBB training data. Extending the CuTR formulation, we incorporate an aleatoric uncertainty for robust regression, a median depth patch encoding to support sparse depth inputs, and large-scale training with over 1.2 million unique 3DBBs. BoxerNet outperforms state-of-the-art baselines in open-world 3DBB lifting, including CuTR in egocentric settings without dense depth (0.532 vs. 0.010 mAP) and on CA-1M with dense depth available (0.412 vs. 0.250 mAP). --- ### Integration of Object Detection and Small VLMs for Construction Safety Hazard Identification Authors: Muhammad Adil, Mehmood Ahmed, Muhammad Aqib, Vicente A. Gonzalez, Gaang Lee, Qipei Mei Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05210v1 Accurate and timely identification of construction hazards around workers is essential for preventing workplace accidents. While large vision-language models (VLMs) demonstrate strong contextual reasoning capabilities, their high computational requirements limit their applicability in near real-time construction hazard detection. In contrast, small vision-language models (sVLMs) with fewer than 4 billion parameters offer improved efficiency but often suffer from reduced accuracy and hallucination when analyzing complex construction scenes. To address this trade-off, this study proposes a detection-guided sVLM framework that integrates object detection with multimodal reasoning for contextual hazard identification. The framework first employs a YOLOv11n detector to localize workers and construction machinery within the scene. The detected entities are then embedded into structured prompts to guide the reasoning process of sVLMs, enabling spatially grounded hazard assessment. Within this framework, six sVLMs (Gemma-3 4B, Qwen-3-VL 2B/4B, InternVL-3 1B/2B, and SmolVLM-2B) were evaluated in zero-shot settings on a curated dataset of construction site images with hazard annotations and explanatory rationales. The proposed approach consistently improved hazard detection performance across all models. The best-performing model, Gemma-3 4B, achieved an F1-score of 50.6%, compared to 34.5% in the baseline configuration. Explanation quality also improved significantly, with BERTScore F1 increasing from 0.61 to 0.82. Despite incorporating object detection, the framework introduces minimal overhead, adding only 2.5 ms per image during inference. These results demonstrate that integrating lightweight object detection with small VLM reasoning provides an effective and efficient solution for context-aware construction safety hazard detection. --- ### Vehicle-as-Prompt: A Unified Deep Reinforcement Learning Framework for Heterogeneous Fleet Vehicle Routing Problem Authors: Shihong Huang, Shengjie Wang, Lei Gao, Hong Ma, Zhanluo Zhang, Feng Zhang, Weihua Zhou Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05195v1 Unlike traditional homogeneous routing problems, the Heterogeneous Fleet Vehicle Routing Problem (HFVRP) involves heterogeneous fixed costs, variable travel costs, and capacity constraints, rendering solution quality highly sensitive to vehicle selection. Furthermore, real-world logistics applications often impose additional complex constraints, markedly increasing computational complexity. However, most existing Deep Reinforcement Learning (DRL)-based methods are restricted to homogeneous scenarios, leading to suboptimal performance when applied to HFVRP and its complex variants. To bridge this gap, we investigate HFVRP under complex constraints and develop a unified DRL framework capable of solving the problem across various variant settings. We introduce the Vehicle-as-Prompt (VaP) mechanism, which formulates the problem as a single-stage autoregressive decision process. Building on this, we propose VaP-CSMV, a framework featuring a cross-semantic encoder and a multi-view decoder that effectively addresses various problem variants and captures the complex mapping relationships between vehicle heterogeneity and customer node attributes. Extensive experimental results demonstrate that VaP-CSMV significantly outperforms existing state-of-the-art DRL-based neural solvers and achieves competitive solution quality compared to traditional heuristic solvers, while reducing inference time to mere seconds. Furthermore, the framework exhibits strong zero-shot generalization capabilities on large-scale and previously unseen problem variants, while ablation studies validate the vital contribution of each component. --- ### Faster Superword Tokenization Authors: Craig W. Schmidt, Chris Tanner, Yuval Pinter Published: 2026-04-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05192v2 Byte Pair Encoding (BPE) is a widely used tokenization algorithm, whose tokens cannot extend across pre-tokenization boundaries, functionally limiting it to representing at most full words. The BoundlessBPE and SuperBPE algorithms extend and improve BPE by relaxing this limitation and allowing the formation of superwords, which are combinations of pretokens that form phrases. However, previous implementations were impractical to train: for example, BoundlessBPE took 4.7 CPU days to train on 1GB of data. We show that supermerge candidates, two or more consecutive pretokens eligible to form a supermerge, can be aggregated by frequency much like regular pretokens. This avoids keeping full documents in memory, as the original implementations of BoundlessBPE and SuperBPE required, leading to a significant training speedup. We present a two-phase formulation of BoundlessBPE that separates first-phase learning of regular merges from second-phase learning of supermerges, producing identical results to the original implementation. We also show a near-equivalence between two-phase BoundlessBPE and SuperBPE, with the difference being that a manually selected hyperparameter used in SuperBPE can be automatically determined in the second phase of BoundlessBPE. These changes enable a much faster implementation, allowing training on that same 1GB of data in 603 and 593 seconds for BoundlessBPE and SuperBPE, respectively, a more than 600x increase in speed. For each of BoundlessBPE, SuperBPE, and BPE, we open-source both a reference Python implementation and a fast Rust implementation. --- ### Faster Superword Tokenization Authors: Craig W. Schmidt, Chris Tanner, Yuval Pinter Published: 2026-04-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05192v1 Byte Pair Encoding (BPE) is a widely used tokenization algorithm, whose tokens cannot extend across pre-tokenization boundaries, functionally limiting it to representing at most full words. The BoundlessBPE and SuperBPE algorithms extend and improve BPE by relaxing this limitation and allowing the formation of superwords, which are combinations of pretokens that form phrases. However, previous implementations were impractical to train: for example, BoundlessBPE took 4.7 CPU days to train on 1GB of data. We show that supermerge candidates, two or more consecutive pretokens eligible to form a supermerge, can be aggregated by frequency much like regular pretokens. This avoids keeping full documents in memory, as the original implementations of BoundlessBPE and SuperBPE required, leading to a significant training speedup. We present a two-phase formulation of BoundlessBPE that separates first-phase learning of regular merges from second-phase learning of supermerges, producing identical results to the original implementation. We also show a near-equivalence between two-phase BoundlessBPE and SuperBPE, with the difference being that a manually selected hyperparameter used in SuperBPE can be automatically determined in the second phase of BoundlessBPE. These changes enable a much faster implementation, allowing training on that same 1GB of data in 603 and 593 seconds for BoundlessBPE and SuperBPE, respectively, a more than 600x increase in speed. For each of BoundlessBPE, SuperBPE, and BPE, we open-source both a reference Python implementation and a fast Rust implementation. --- ### Retrieval-Augmented LLMs for Evidence Localization in Clinical Trial Recruitment from Longitudinal EHR Narratives Authors: Ziyi Chen, Mengxian Lyu, Cheng Peng, Yonghui Wu Published: 2026-04-06 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2604.05190v2 Screening patients for enrollment is a well-known, labor-intensive bottleneck that leads to under-enrollment and, ultimately, trial failures. Recent breakthroughs in large language models (LLMs) offer a promising opportunity to use artificial intelligence to improve screening. This study systematically explored both encoder- and decoder-based generative LLMs for screening clinical narratives to facilitate clinical trial recruitment. We examined both general-purpose LLMs and medical-adapted LLMs and explored three strategies to alleviate the "Lost in the Middle" issue when handling long documents, including 1) Original long-context: using the default context windows of LLMs, 2) NER-based extractive summarization: converting the long document into summarizations using named entity recognition, 3) RAG: dynamic evidence retrieval based on eligibility criteria. The 2018 N2C2 Track 1 benchmark dataset is used for evaluation. Our experimental results show that the MedGemma model with the RAG strategy achieved the best micro-F1 score of 89.05%, outperforming other models. Generative LLMs have remarkably improved trial criteria that require long-term reasoning across long documents, whereas trial criteria that span a short piece of context (e.g., lab tests) show incremental improvements. The real-world adoption of LLMs for trial recruitment must consider specific criteria for selecting among rule-based queries, encoder-based LLMs, and generative LLMs to maximize efficiency within reasonable computing costs. --- ### FNO$^{\angle θ}$: Extended Fourier neural operator for learning state and optimal control of distributed parameter systems Authors: Zhexian Li, Ketan Savla Published: 2026-04-06 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.05187v1 We propose an extended Fourier neural operator (FNO) architecture for learning state and linear quadratic additive optimal control of systems governed by partial differential equations. Using the Ehrenpreis-Palamodov fundamental principle, we show that any state and optimal control of linear PDEs with constant coefficients can be represented as an integral in the complex domain. The integrand of this representation involves the same exponential term as in the inverse Fourier transform, where the latter is used to represent the convolution operator in FNO layer. Motivated by this observation, we modify the FNO layer by extending the frequency variable in the inverse Fourier transform from the real to complex domain to capture the integral representation from the fundamental principle. We illustrate the performance of FNO in learning state and optimal control for the nonlinear Burgers' equation, showing order of magnitude improvements in training errors and more accurate predictions of non-periodic boundary values over FNO. --- ### Cross-fitted Proximal Learning for Model-Based Reinforcement Learning Authors: Nishanth Venkatesh, Andreas A. Malikopoulos Published: 2026-04-06 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.05185v1 Model-based reinforcement learning is attractive for sequential decision-making because it explicitly estimates reward and transition models and then supports planning through simulated rollouts. In offline settings with hidden confounding, however, models learned directly from observational data may be biased. This challenge is especially pronounced in partially observable systems, where latent factors may jointly affect actions, rewards, and future observations. Recent work has shown that policy evaluation in such confounded partially observable Markov decision processes (POMDPs) can be reduced to estimating reward-emission and observation-transition bridge functions satisfying conditional moment restrictions (CMRs). In this paper, we study the statistical estimation of these bridge functions. We formulate bridge learning as a CMR problem with nuisance objects given by a conditional mean embedding and a conditional density. We then develop a $K$-fold cross-fitted extension of the existing two-stage bridge estimator. The proposed procedure preserves the original bridge-based identification strategy while using the available data more efficiently than a single sample split. We also derive an oracle-comparator bound for the cross-fitted estimator and decompose the resulting error into a Stage I term induced by nuisance estimation and a Stage II term induced by empirical averaging. --- ### OrthoFuse: Training-free Riemannian Fusion of Orthogonal Style-Concept Adapters for Diffusion Models Authors: Ali Aliev, Kamil Garifullin, Nikolay Yudin, Vera Soboleva, Alexander Molozhavenko, Ivan Oseledets, Aibek Alanov, Maxim Rakhuba Published: 2026-04-06 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05183v1 In a rapidly growing field of model training there is a constant practical interest in parameter-efficient fine-tuning and various techniques that use a small amount of training data to adapt the model to a narrow task. However, there is an open question: how to combine several adapters tuned for different tasks into one which is able to yield adequate results on both tasks? Specifically, merging subject and style adapters for generative models remains unresolved. In this paper we seek to show that in the case of orthogonal fine-tuning (OFT), we can use structured orthogonal parametrization and its geometric properties to get the formulas for training-free adapter merging. In particular, we derive the structure of the manifold formed by the recently proposed Group-and-Shuffle ($\mathcal{GS}$) orthogonal matrices, and obtain efficient formulas for the geodesics approximation between two points. Additionally, we propose a $\text{spectra restoration}$ transform that restores spectral properties of the merged adapter for higher-quality fusion. We conduct experiments in subject-driven generation tasks showing that our technique to merge two $\mathcal{GS}$ orthogonal matrices is capable of uniting concept and style features of different adapters. To the best of our knowledge, this is the first training-free method for merging multiplicative orthogonal adapters. Code is available via the $\href{https://github.com/ControlGenAI/OrthoFuse}{link}$. --- ### LSRM: High-Fidelity Object-Centric Reconstruction via Scaled Context Windows Authors: Zhengqin Li, Cheng Zhang, Jakob Engel, Zhao Dong Published: 2026-04-06 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.05182v3 We introduce the Large Sparse Reconstruction Model to study how scaling transformer context windows affects feed-forward 3D reconstruction. Although recent object-centric feed-forward methods produce robust, high-quality reconstructions, they still lag behind dense-view optimization in recovering fine-grained texture and appearance. We show that expanding the context window -- by substantially increasing the number of active object and image tokens -- narrows this gap and enables high-fidelity 3D object reconstruction and inverse rendering. To scale effectively, we adapt native sparse attention for 3D reconstruction with three key contributions: (1) an efficient coarse-to-fine pipeline that focuses computation on informative regions by predicting sparse high-resolution residuals; (2) a 3D-aware spatial routing mechanism that establishes accurate 2D-3D correspondences using explicit geometric distances rather than standard attention scores; and (3) a custom block-aware sequence-parallel strategy with an All-gather-KV protocol to balance dynamic, sparse workloads across GPUs. As a result, LSRM handles 20x more object tokens and >2x more image tokens than prior state-of-the-art (SOTA) methods. Extensive evaluations on standard novel-view synthesis benchmarks show substantial gains over the current SOTA, yielding >2.4dB higher PSNR and >40% lower LPIPS. Furthermore, when extending LSRM to inverse rendering, qualitative and quantitative evaluations on widely used benchmarks demonstrate consistent improvements in texture and geometry details, achieving an LPIPS that matches or exceeds that of SOTA dense-view optimization methods. Code and model weights are available on our project page. --- ### General Multimodal Protein Design Enables DNA-Encoding of Chemistry Authors: Jarrid Rector-Brooks, Théophile Lambert, Marta Skreta, Daniel Roth, Yueming Long, Zi-Qi Li, Xi Zhang, Miruna Cretu, Francesca-Zhoufan Li, Tanvi Ganapathy, Emily Jin, Avishek Joey Bose et al. Published: 2026-04-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05181v1 Evolution is an extraordinary engine for enzymatic diversity, yet the chemistry it has explored remains a narrow slice of what DNA can encode. Deep generative models can design new proteins that bind ligands, but none have created enzymes without pre-specifying catalytic residues. We introduce DISCO (DIffusion for Sequence-structure CO-design), a multimodal model that co-designs protein sequence and 3D structure around arbitrary biomolecules, as well as inference-time scaling methods that optimize objectives across both modalities. Conditioned solely on reactive intermediates, DISCO designs diverse heme enzymes with novel active-site geometries. These enzymes catalyze new-to-nature carbene-transfer reactions, including alkene cyclopropanation, spirocyclopropanation, B-H, and C(sp$^3$)-H insertions, with high activities exceeding those of engineered enzymes. Random mutagenesis of a selected design further confirmed that enzyme activity can be improved through directed evolution. By providing a scalable route to evolvable enzymes, DISCO broadens the potential scope of genetically encodable transformations. Code is available at https://github.com/DISCO-design/DISCO. --- ### MIRAGE: Benchmarking and Aligning Multi-Instance Image Editing Authors: Ziqian Liu, Stephan Alaniz Published: 2026-04-06 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05180v1 Instruction-guided image editing has seen remarkable progress with models like FLUX.2 and Qwen-Image-Edit, yet they still struggle with complex scenarios with multiple similar instances each requiring individual edits. We observe that state-of-the-art models suffer from severe over-editing and spatial misalignment when faced with multiple identical instances and composite instructions. To this end, we introduce a comprehensive benchmark specifically designed to evaluate fine-grained consistency in multi-instance and multi-instruction settings. To address the failures of existing methods observed in our benchmark, we propose Multi-Instance Regional Alignment via Guided Editing (MIRAGE), a training-free framework that enables precise, localized editing. By leveraging a vision-language model to parse complex instructions into regional subsets, MIRAGE employs a multi-branch parallel denoising strategy. This approach injects latent representations of target regions into the global representation space while maintaining background integrity through a reference trajectory. Extensive evaluations on MIRA-Bench and RefEdit-Bench demonstrate that our framework significantly outperforms existing methods in achieving precise instance-level modifications while preserving background consistency. Our benchmark and code are available at https://github.com/ZiqianLiu666/MIRAGE. ---