# Research Papers — 2026-04 · bundle page 128 of 233 > Full-text markdown of papers #6351–6400 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/128 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/127.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/129.md) ### Masked by Consensus: Disentangling Privileged Knowledge in LLM Correctness Authors: Tomer Ashuach, Shai Gretz, Yoav Katz, Yonatan Belinkov, Liat Ein-Dor Published: 2026-04-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12373v5 Humans use introspection to evaluate their understanding through private internal states inaccessible to external observers. We investigate whether large language models possess similar privileged knowledge about answer correctness, information unavailable through external observation. We train correctness classifiers on question representations from both a model's own hidden states and external models, testing whether self-representations provide a performance advantage. On standard evaluation, we find no advantage: self-probes perform comparably to peer-model probes. We hypothesize this is due to high inter-model agreement of answer correctness. To isolate genuine privileged knowledge, we evaluate on disagreement subsets, where models produce conflicting predictions. Here, we discover domain-specific privileged knowledge: self-representations consistently outperform peer representations in factual knowledge tasks, but show no advantage in math reasoning. We further localize this domain asymmetry across model layers, finding that the factual advantage emerges progressively from early-to-mid layers onward, consistent with model-specific memory retrieval, while math reasoning shows no consistent advantage at any depth. --- ### Is Sliding Window All You Need? An Open Framework for Long-Sequence Recommendation Authors: Sayak Chakrabarty, Souradip Pal Published: 2026-04-14 Categories: cs.LG, cs.IR arXiv: https://arxiv.org/abs/2604.12372v1 Long interaction histories are central to modern recommender systems, yet training with long sequences is often dismissed as impractical under realistic memory and latency budgets. This work demonstrates that it is not only practical but also effective-at academic scale. We release a complete, end-to-end framework that implements industrial-style long-sequence training with sliding windows, including all data processing, training, and evaluation scripts. Beyond reproducing prior gains, we contribute two capabilities missing from earlier reports: (i) a runtime-aware ablation study that quantifies the accuracy-compute frontier across windowing regimes and strides, and (ii) a novel k-shift embedding layer that enables million-scale vocabularies on commodity GPUs with negligible accuracy loss. Our implementation trains reliably on modest university clusters while delivering competitive retrieval quality (e.g., up to +6.04% MRR and +6.34% Recall@10 on Retailrocket) with $\sim 4 \times $ training-time overheads. By packaging a robust pipeline, reporting training time costs, and introducing an embedding mechanism tailored for low-resource settings, we transform long-sequence training from a closed, industrial technique into a practical, open, and extensible methodology for the community. --- ### Reading Between the Pixels: Linking Text-Image Embedding Alignment to Typographic Attack Success on Vision-Language Models Authors: Ravikumar Balakrishnan, Sanket Mendapara, Ankit Garg Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12371v2 We study typographic prompt injection attacks on vision-language models (VLMs), where adversarial text is rendered as images to bypass safety mechanisms, posing a growing threat as VLMs serve as the perceptual backbone of autonomous agents, from browser automation and computer-use systems to camera-equipped embodied agents. In practice, the attack surface is heterogeneous: adversarial text appears at varying font sizes and under diverse visual conditions, while the growing ecosystem of VLMs exhibits substantial variation in vulnerability, complicating defensive approaches. Evaluating 1,000 prompts from SALAD-Bench across four VLMs, namely, GPT-4o, Claude Sonnet 4.5, Mistral-Large-3, and Qwen3-VL-4B-Instruct under varying font sizes (6--28px) and visual transformations (rotation, blur, noise, contrast changes), we find: (1) font size significantly affects attack success rate (ASR), with very small fonts (6px) yielding near-zero ASR while mid-range fonts achieve peak effectiveness; (2) text attacks are more effective than image attacks for GPT-4o (36% vs 8%) and Claude (47% vs 22%), while Qwen3-VL and Mistral show comparable ASR across modalities; (3) text-image embedding distance from two multimodal embedding models (JinaCLIP and Qwen3-VL-Embedding) shows strong negative correlation with ASR across all four models (r = -0.71 to -0.93, p < 0.01); (4) heavy degradations increase embedding distance by 10--12% and reduce ASR by 34--96%, while rotation asymmetrically affects models (Mistral drops 50%, GPT-4o unchanged). These findings highlight that model-specific robustness patterns preclude one-size-fits-all defenses and offer empirical guidance for practitioners selecting VLM backbones for agentic systems operating in adversarial environments. --- ### Adaptive Spiking Neurons for Vision and Language Modeling Authors: Chenlin Zhou, Sihang Guo, Jiaqi Wang, Dongyang Ma, Jin Cheng, Qingyan Meng, Zhengyu Ma, Yonghong Tian Published: 2026-04-14 Categories: cs.NE arXiv: https://arxiv.org/abs/2604.12365v1 Regarded as the third generation of neural networks, Spiking Neural Networks (SNNs) have garnered significant traction due to their biological plausibility and energy efficiency. Recent advancements in large models necessitate spiking neurons capable of high performance, adaptability, and training efficiency. In this work, we first propose a novel functional perspective that provides general guidance for designing the new generation of spiking neurons. Following the insightful guidelines, we propose the Adaptive Spiking Neuron (ASN), which incorporates trainable parameters to learn membrane potential dynamics and enable adaptive firing. ASN adopts an integer training and spike inference paradigm, facilitating efficient SNN training. To further enhance robustness, we propose a specialized variant of ASN, the Normalized Adaptive Spiking Neuron (NASN), which integrates normalization to stabilize training. We evaluate our neuron model on 19 datasets spanning five distinct tasks in both vision and language modalities, demonstrating the effectiveness and versatility of the ASN family. Our ASN family is expected to become the new generation of general-purpose spiking neurons. --- ### Cross-Domain Transfer with Particle Physics Foundation Models: From Jets to Neutrino Interactions Authors: Gregor Krzmanc, Vinicius Mikuni, Benjamin Nachman, Callum Wilkinson Published: 2026-04-14 Categories: hep-ex, cs.LG, hep-ph, physics.data-an arXiv: https://arxiv.org/abs/2604.12364v2 Future AI-based studies in particle physics will likely start from a foundation model to accelerate training and enhance sensitivity. As a step toward a general-purpose foundation model for particle physics, we investigate whether the OmniLearned and ParticleViT foundation models pretrained on diverse high-$Q^2$ simulated and real $pp$ and $ep$ collisions retain useful knowledge to a few-GeV fixed-target neutrino experiment. We process MINERvA neutrino--nucleus scattering events and evaluate pretrained models on two types of tasks: regression of available energy and binary classification of charged-current pion final states ($\mathrm{CC1π^{\pm}}$, $\mathrm{CCNπ^{\pm}}$, and $\mathrm{CC1π^{0}}$). Pretrained OmniLearned and ParticleViT models outperform similarly sized models trained from scratch at the same compute budget, with the largest gains for OmniLearned on regression and for ParticleViT on classification. When the same transformer architecture is instead initialized from unrelated text pretraining (BERT), this advantage appears only marginally for classification in terms of compute efficiency and not in any way for regression. These results suggest that particle-level foundation models acquire inductive biases that generalize across large differences in energy scale, detector technology, and underlying physics processes, pointing toward detector-agnostic inference in particle physics. --- ### Cross-Domain Transfer with Particle Physics Foundation Models: From Jets to Neutrino Interactions Authors: Gregor Krzmanc, Vinicius Mikuni, Benjamin Nachman, Callum Wilkinson Published: 2026-04-14 Categories: hep-ex, cs.LG, hep-ph, physics.data-an arXiv: https://arxiv.org/abs/2604.12364v1 Future AI-based studies in particle physics will likely start from a foundation model to accelerate training and enhance sensitivity. As a step towards a general-purpose foundation model for particle physics, we investigate whether the OmniLearned foundation model pre-trained on diverse high-$Q^2$ simulated and real $pp$ and $ep$ collisions can be effectively transferred to a few-GeV fixed-target neutrino experiment. We process MINERvA neutrino--nucleus scattering events and evaluate pre-trained models on two types of tasks: regression of available energy and binary classification of charged-current pion final states ($\mathrm{CC1π^{\pm}}$, $\mathrm{CCNπ^{\pm}}$, and $\mathrm{CC1π^{0}}$). Pre-trained OmniLearned models consistently outperform similarly sized models trained from scratch, achieving better overall performance at the same compute budget, as well as achieving better performance at the same number of training steps. These results suggest that particle-level foundation models acquire inductive biases that generalize across large differences in energy scale, detector technology, and underlying physics processes, pointing toward a paradigm of detector-agnostic inference in particle physics. --- ### Compiling Activation Steering into Weights via Null-Space Constraints for Stealthy Backdoors Authors: Rui Yin, Tianxu Han, Naen Xu, Changjiang Li, Ping He, Chunyi Zhou, Jun Wang, Zhihui Fu, Tianyu Du, Jinbao Li, Shouling Ji Published: 2026-04-14 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2604.12359v1 Safety-aligned large language models (LLMs) are increasingly deployed in real-world pipelines, yet this deployment also enlarges the supply-chain attack surface: adversaries can distribute backdoored checkpoints that behave normally under standard evaluation but jailbreak when a hidden trigger is present. Recent post-hoc weight-editing methods offer an efficient approach to injecting such backdoors by directly modifying model weights to map a trigger to an attacker-specified response. However, existing methods typically optimize a token-level mapping that forces an affirmative prefix (e.g., ``Sure''), which does not guarantee sustained harmful output -- the model may begin with apparent agreement yet revert to safety-aligned refusal within a few decoding steps. We address this reliability gap by shifting the backdoor objective from surface tokens to internal representations. We extract a steering vector that captures the difference between compliant and refusal behaviors, and compile it into a persistent weight modification that activates only when the trigger is present. To preserve stealthiness and benign utility, we impose a null-space constraint so that the injected edit remains dormant on clean inputs. The method is efficient, requiring only a small set of examples and admitting a closed-form solution. Across multiple safety-aligned LLMs and jailbreak benchmarks, our method achieves high triggered attack success while maintaining non-triggered safety and general utility. --- ### Why and When Visual Token Pruning Fails? A Study on Relevant Visual Information Shift in MLLMs Decoding Authors: Jiwan Kim, Kibum Kim, Wonjoong Kim, Byung-Kwan Lee, Chanyoung Park Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12358v2 Recently, visual token pruning has been studied to handle the vast number of visual tokens in Multimodal Large Language Models. However, we observe that while existing pruning methods perform reliably on simple visual understanding, they struggle to effectively generalize to complex visual reasoning tasks, a critical gap underexplored in previous studies. Through a systematic analysis, we identify Relevant Visual Information Shift (RVIS) during decoding as the primary failure driver. To address this, we propose Decoding-stage Shift-aware Token Pruning (DSTP), a training-free add-on framework that enables existing pruning methods to align visual tokens with shifting reasoning requirements during the decoding stage. Extensive experiments demonstrate that DSTP significantly mitigates performance degradation of pruning methods in complex reasoning tasks, while consistently yielding performance gains even across visual understanding benchmarks. Furthermore, DSTP demonstrates effectiveness across diverse state-of-the-art architectures, highlighting its generalizability and efficiency with minimal computational overhead. --- ### ReflectCAP: Detailed Image Captioning with Reflective Memory Authors: Kyungmin Min, Minbeom Kim, Kang-il Lee, Seunghyun Yoon, Kyomin Jung Published: 2026-04-14 Categories: cs.AI, cs.CV arXiv: https://arxiv.org/abs/2604.12357v1 Detailed image captioning demands both factual grounding and fine-grained coverage, yet existing methods have struggled to achieve them simultaneously. We address this tension with Reflective Note-Guided Captioning (ReflectCAP), where a multi-agent pipeline analyzes what the target large vision-language model (LVLM) consistently hallucinates and what it systematically overlooks, distilling these patterns into reusable guidelines called Structured Reflection Notes. At inference time, these notes steer the captioning model along both axes -- what to avoid and what to attend to -- yielding detailed captions that jointly improve factuality and coverage. Applying this method to 8 LVLMs spanning the GPT-4.1 family, Qwen series, and InternVL variants, ReflectCAP reaches the Pareto frontier of the trade-off between factuality and coverage, and delivers substantial gains on CapArena-Auto, where generated captions are judged head-to-head against strong reference models. Moreover, ReflectCAP offers a more favorable trade-off between caption quality and compute cost than model scaling or existing multi-agent pipelines, which incur 21--36\% greater overhead. This makes high-quality detailed captioning viable under real-world cost and latency constraints. --- ### OmniFood8K: Single-Image Nutrition Estimation via Hierarchical Frequency-Aligned Fusion Authors: Dongjian Yu, Weiqing Min, Qian Jiang, Xing Lin, Xin Jin, Shuqiang Jiang Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12356v1 Accurate estimation of food nutrition plays a vital role in promoting healthy dietary habits and personalized diet management. Most existing food datasets primarily focus on Western cuisines and lack sufficient coverage of Chinese dishes, which restricts accurate nutritional estimation for Chinese meals. Moreover, many state-of-the-art nutrition prediction methods rely on depth sensors, restricting their applicability in daily scenarios. To address these limitations, we introduce OmniFood8K, a comprehensive multimodal dataset comprising 8,036 food samples, each with detailed nutritional annotations and multi-view images. In addition, to enhance models' capability in nutritional prediction, we construct NutritionSynth-115K, a large-scale synthetic dataset that introduces compositional variations while preserving precise nutritional labels. Moreover, we propose an end-to-end framework for nutritional prediction from a single RGB image. First, we predict a depth map from a single RGB image and design the Scale-Shift Residual Adapter (SSRA) to refine it for global scale consistency and local structural preservation. Second, we propose the Frequency-Aligned Fusion Module (FAFM) to hierarchically align and fuse RGB and depth features in the frequency domain. Finally, we design a Mask-based Prediction Head (MPH) to emphasize key ingredient regions via dynamic channel selection for more accurate prediction. Extensive experiments on multiple datasets demonstrate the superiority of our method over existing approaches. Project homepage: https://yudongjian.github.io/OmniFood8K-food/ --- ### Combating Pattern and Content Bias: Adversarial Feature Learning for Generalized AI-Generated Image Detection Authors: Haifeng Zhang, Qinghui He, Xiuli Bi, Bo Liu, Chi-Man Pun, Bin Xiao Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12353v1 In recent years, the rapid development of generative artificial intelligence technology has significantly lowered the barrier to creating high-quality fake images, posing a serious challenge to information authenticity and credibility. Existing generated image detection methods typically enhance generalization through model architecture or network design. However, their generalization performance remains susceptible to data bias, as the training data may drive models to fit specific generative patterns and content rather than the common features shared by images from different generative models (asymmetric bias learning). To address this issue, we propose a Multi-dimensional Adversarial Feature Learning (MAFL) framework. The framework adopts a pretrained multimodal image encoder as the feature extraction backbone, constructs a real-fake feature learning network, and designs an adversarial bias-learning branch equipped with a multi-dimensional adversarial loss, forming an adversarial training mechanism between authenticity-discriminative feature learning and bias feature learning. By suppressing generation-pattern and content biases, MAFL guides the model to focus on the generative features shared across different generative models, thereby effectively capturing the fundamental differences between real and generated images, enhancing cross-model generalization, and substantially reducing the reliance on large-scale training data. Through extensive experimental validation, our method outperforms existing state-of-the-art approaches by 10.89% in accuracy and 8.57% in Average Precision (AP). Notably, even when trained with only 320 images, it can still achieve over 80% detection accuracy on public datasets. --- ### MultiDocFusion: Hierarchical and Multimodal Chunking Pipeline for Enhanced RAG on Long Industrial Documents Authors: Joongmin Shin, Chanjun Park, Jeongbae Park, Jaehyung Seo, Heuiseok Lim Published: 2026-04-14 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12352v1 RAG-based QA has emerged as a powerful method for processing long industrial documents. However, conventional text chunking approaches often neglect complex and long industrial document structures, causing information loss and reduced answer quality. To address this, we introduce MultiDocFusion, a multimodal chunking pipeline that integrates: (i) detection of document regions using vision-based document parsing, (ii) text extraction from these regions via OCR, (iii) reconstruction of document structure into a hierarchical tree using large language model (LLM)-based document section hierarchical parsing (DSHP-LLM), and (iv) construction of hierarchical chunks through DFS-based grouping. Extensive experiments across industrial benchmarks demonstrate that MultiDocFusion improves retrieval precision by 8-15% and ANLS QA scores by 2-3% compared to baselines, emphasizing the critical role of explicitly leveraging document hierarchy for multimodal document-based QA. These significant performance gains underscore the necessity of structure-aware chunking in enhancing the fidelity of RAG-based QA systems. --- ### Fundus Image-based Glaucoma Screening via Retinal Knowledge-Oriented Dynamic Multi-Level Feature Integration Authors: Chi Liu, Yuzhuo Zhou, Sheng Shen, Zongyuan Ge, Fengshi Jing, Shiran Zhang, Yu Jiang, Anli Wang, Wenjian Liu, Feilong Yang, Tianqing Zhu, Xiaotong Han Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12351v2 While deep learning has advanced automated glaucoma screening via color fundus photography, existing purely data-driven models often overfit to confounding imaging artifacts and struggle to capture unpredictable pathological cues located beyond predefined anatomical boundaries. To address these limitations, we propose a retinal knowledge-oriented framework that synergizes dynamic multi-scale feature learning with domain-specific anatomical priors. The proposed architecture adopts a tri-branch structure to jointly model the global retinal context, the structural characteristics of the optic cup and disc, and dynamically cropped pathological regions. Specifically, we devise a Dynamic Window Mechanism (DWM) that adaptively discovers diagnostically informative patches via image-level supervision. Furthermore, we introduce a Knowledge-Enhanced Convolutional Block Attention Module (KE-CBAM) that explicitly incorporates retinal priors from a pre-trained foundation model RETFound to guide spatial attention, preventing the network from assigning spurious weights to irrelevant background noise. Extensive evaluations on the large-scale AIROGS dataset demonstrate that our method achieves a state-of-the-art AUC of $98.5\%$ and an accuracy of $94.6\%$. More importantly, the integration of anatomical priors effectively mitigates the inherent class imbalance, significantly improving the detection of referable glaucoma cases. Additional validations on the SMDG-19 benchmark further confirm its superior cross-domain generalization, indicating that our contribution provides a robust, interpretable, and scalable solution for real-world clinical glaucoma diagnosis. Our code is available at https://github.com/magiczhuo/Glaucoma-Detection --- ### Scaffold-Conditioned Preference Triplets for Controllable Molecular Optimization with Large Language Models Authors: Yi Xiong, Liang Xiong, Xiaohong Ji, Sen Yang, Zhifeng Gao, Huaimin Wang, Kele Xu Published: 2026-04-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.12350v1 Molecular property optimization is central to drug discovery, yet many deep learning methods rely on black-box scoring and offer limited control over scaffold preservation, often producing unstable or biologically implausible edits. While large language models (LLMs) are promising molecular generators, optimization remains constrained by the lack of chemistry-grounded preference supervision and principled data curation. We introduce \textbf{Scaffold-Conditioned Preference Triplets (SCPT)}, a pipeline that constructs similarity-constrained triplets $\langle\text{scaffold}, \text{better}, \text{worse}\rangle$ via scaffold alignment and chemistry-driven filters for validity, synthesizability, and meaningful property gains. Using these preferences, we align a pretrained molecular LLM as a conditional editor, enabling property-improving edits that retain the scaffold. Across single- and multi-objective benchmarks, SCPT improves optimization success and property gains while maintaining higher scaffold similarity than competitive baselines. Compared with representative non-LLM molecular optimization methods, SCPT-trained LLMs are better suited to scaffold-constrained and multi-objective optimization. In addition, models trained on single-property and two-property supervision generalize effectively to three-property tasks, indicating promising extrapolative generalization under limited higher-order supervision. SCPT also provides controllable data-construction knobs that yield a predictable similarity-gain frontier, enabling systematic adaptation to diverse optimization regimes. --- ### PrivEraserVerify: Efficient, Private, and Verifiable Federated Unlearning Authors: Parthaw Goswami, Md Khairul Islam, Ashfak Yeafi Published: 2026-04-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.12348v1 Federated learning (FL) enables collaborative model training without sharing raw data, offering a promising path toward privacy preserving artificial intelligence. However, FL models may still memorize sensitive information from participants, conflicting with the right to be forgotten (RTBF). To meet these requirements, federated unlearning has emerged as a mechanism to remove the contribution of departing clients. Existing solutions only partially address this challenge: FedEraser improves efficiency but lacks privacy protection, FedRecovery ensures differential privacy (DP) but degrades accuracy, and VeriFi enables verifiability but introduces overhead without efficiency or privacy guarantees. We present PrivEraserVerify (PEV), a unified framework that integrates efficiency, privacy, and verifiability into federated unlearning. PEV employs (i) adaptive checkpointing to retain critical historical updates for fast reconstruction, (ii) layer adaptive differentially private calibration to selectively remove client influence while minimizing accuracy loss, and (iii) fingerprint based verification, enabling participants to confirm unlearning in a decentralized and noninvasive manner. Experiments on image, handwritten character, and medical datasets show that PEV achieves up to 2 to 3 times faster unlearning than retraining, provides formal indistinguishability guarantees with reduced performance degradation, and supports scalable verification. To the best of our knowledge, PEV is the first framework to simultaneously deliver efficiency, privacy, and verifiability for federated unlearning, moving FL closer to practical and regulation compliant deployment. --- ### Unlocking the Potential of Grounding DINO in Videos: Parameter-Efficient Adaptation for Limited-Data Spatial-Temporal Localization Authors: Zanyi Wang, Fan Li, Dengyang Jiang, Liuzhuozheng Li, Yunhua Zhong, Guang Dai, Mengmeng Wang Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12346v1 Spatio-temporal video grounding (STVG) aims to localize queried objects within dynamic video segments. Prevailing fully-trained approaches are notoriously data-hungry. However, gathering large-scale STVG data is exceptionally challenging: dense frame-level bounding boxes and complex temporal language alignments are prohibitively expensive to annotate, especially for specialized video domains. Consequently, conventional models suffer from severe overfitting on these inherently limited datasets, while zero-shot foundational models lack the task-specific temporal awareness needed for precise localization. To resolve this small-data challenge, we introduce ST-GD, a data-efficient framework that adapts pre-trained 2D visual-language models (e.g., Grounding DINO) to video tasks. To avoid destroying pre-trained priors on small datasets, ST-GD keeps the base model frozen and strategically injects lightweight adapters (~10M trainable parameters) to instill spatio-temporal awareness, alongside a novel temporal decoder for boundary prediction. This design naturally counters data scarcity. Consequently, ST-GD excels in data-scarce scenarios, achieving highly competitive performance on the limited-scale HC-STVG v1/v2 benchmarks, while maintaining robust generalization on the VidSTG dataset. This validates ST-GD as a powerful paradigm for complex video understanding under strict small-data constraints. --- ### FRTSearch: Unified Detection and Parameter Inference of Fast Radio Transients using Instance Segmentation Authors: Bin Zhang, Yabiao Wang, Xiaoyao Xie, Shanping You, Xuhong Yu, Qiuhua Li, Hongwei Li, Shaowen Du, Chenchen Miao, Dengke Zhou, Jianhua Fang, Jiafu Wu et al. Published: 2026-04-14 Categories: astro-ph.IM, cs.AI arXiv: https://arxiv.org/abs/2604.12344v1 The exponential growth of data from modern radio telescopes presents a significant challenge to traditional single-pulse search algorithms, which are computationally intensive and prone to high false-positive rates due to Radio Frequency Interference (RFI). In this work, we introduce FRTSearch, an end-to-end framework unifying the detection and physical characterization of Fast Radio Transients (FRTs). Leveraging the morphological universality of dispersive trajectories in time-frequency dynamic spectra, we reframe FRT detection as a pattern recognition problem governed by the cold plasma dispersion relation. To facilitate this, we constructed CRAFTS-FRT, a pixel-level annotated dataset derived from the Commensal Radio Astronomy FAST Survey (CRAFTS), comprising 2{,}392 instances across diverse source classes. This dataset enables the training of a Mask R-CNN model for precise trajectory segmentation. Coupled with our physics-driven IMPIC algorithm, the framework maps the geometric coordinates of segmented trajectories to directly infer the Dispersion Measure (DM) and Time of Arrival (ToA). Benchmarking on the FAST-FREX dataset shows that FRTSearch achieves a 98.0\% recall, competitive with exhaustive search methods, while reducing false positives by over 99.9\% compared to PRESTO and delivering a processing speedup of up to $13.9\times$. Furthermore, the framework demonstrates robust cross-facility generalization, detecting all 19 tested FRBs from the ASKAP survey without retraining. By shifting the paradigm from ``search-then-identify'' to ``detect-and-infer,'' FRTSearch provides a scalable, high-precision solution for real-time discovery in the era of petabyte-scale radio astronomy. --- ### Detecting Precise Hand Touch Moments in Egocentric Video Authors: Huy Anh Nguyen, Feras Dayoub, Minh Hoai Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12343v1 We address the challenging task of detecting the precise moment when hands make contact with objects in egocentric videos. This frame-level detection is crucial for augmented reality, human-computer interaction, assistive technologies, and robot learning applications, where contact onset signals action initiation or completion. Temporally precise detection is particularly challenging due to subtle hand motion variations near contact, frequent occlusions, fine-grained manipulation patterns, and the inherent motion dynamics of first-person perspectives. To tackle these challenges, we propose a Hand-informed Context Enhanced module (HiCE; pronounced `high-see') that leverages spatiotemporal features from hand regions and their surrounding context through cross-attention mechanisms, learning to identify potential contact patterns. Our approach is further refined with a grasp-aware loss and soft label that emphasizes hand pose patterns and movement dynamics characteristic of touch events, enabling the model to distinguish between near-contact and actual contact frames. We also introduce TouchMoment, an egocentric dataset containing 4,021 videos and 8,456 annotated contact moments spanning over one million frames. Experiments on TouchMoment show that, under a strict evaluation criterion that counts a prediction as correct only if it falls within a two-frame tolerance of the ground-truth moment, our method achieves substantial gains and outperforms state-of-the-art event-spotting baselines by 16.91% average precision. --- ### CoLA: A Choice Leakage Attack Framework to Expose Privacy Risks in Subset Training Authors: Qi Li, Cheng-Long Wang, Yinzhi Cao, Di Wang Published: 2026-04-14 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2604.12342v1 Training models on a carefully chosen portion of data rather than the full dataset is now a standard preprocess for modern ML. From vision coreset selection to large-scale filtering in language models, it enables scalability with minimal utility loss. A common intuition is that training on fewer samples should also reduce privacy risks. In this paper, we challenge this assumption. We show that subset training is not privacy free: the very choices of which data are included or excluded can introduce new privacy surface and leak more sensitive information. Such information can be captured by adversaries either through side-channel metadata from the subset selection process or via the outputs of the target model. To systematically study this phenomenon, we propose CoLA (Choice Leakage Attack), a unified framework for analyzing privacy leakage in subset selection. In CoLA, depending on the adversary's knowledge of the side-channel information, we define two practical attack scenarios: Subset-aware Side-channel Attacks and Black-box Attacks. Under both scenarios, we investigate two privacy surfaces unique to subset training: (1) Training-membership MIA (TM-MIA), which concerns only the privacy of training data membership, and (2) Selection-participation MIA (SP-MIA), which concerns the privacy of all samples that participated in the subset selection process. Notably, SP-MIA enlarges the notion of membership from model training to the entire data-model supply chain. Experiments on vision and language models show that existing threat models underestimate subset-training privacy risks: the expanded privacy surface leaks both training and selection membership, extending risks from individual models to the broader ML ecosystem. --- ### Bridging the Micro--Macro Gap: Frequency-Aware Semantic Alignment for Image Manipulation Localization Authors: Xiaojie Liang, Zhimin Chen, Ziqi Sheng, Wei Lu Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12341v2 As generative image editing advances, image manipulation localization (IML) must handle both traditional manipulations with conspicuous forensic artifacts and diffusion-generated edits that appear locally realistic. Existing methods typically rely on either low-level forensic cues or high-level semantics alone, leading to a fundamental micro--macro gap. To bridge this gap, we propose FASA, a unified framework for localizing both traditional and diffusion-generated manipulations. Specifically, we extract manipulation-sensitive frequency cues through an adaptive dual-band DCT module and learn manipulation-aware semantic priors via patch-level contrastive alignment on frozen CLIP representations. We then inject these priors into a hierarchical frequency pathway through a semantic-frequency side adapter for multi-scale feature interaction, and employ a prototype-guided, frequency-gated mask decoder to integrate semantic consistency with boundary-aware localization for tampered region prediction. Extensive experiments on OpenSDI and multiple traditional manipulation benchmarks demonstrate state-of-the-art localization performance, strong cross-generator and cross-dataset generalization, and robust performance under common image degradations. --- ### Information-Geometric Decomposition of Generalization Error in Unsupervised Learning Authors: Gilhan Kim Published: 2026-04-14 Categories: stat.ML, cond-mat.stat-mech, cs.IT, cs.LG, math.ST arXiv: https://arxiv.org/abs/2604.12340v1 We decompose the Kullback--Leibler generalization error (GE) -- the expected KL divergence from the data distribution to the trained model -- of unsupervised learning into three non-negative components: model error, data bias, and variance. The decomposition is exact for any e-flat model class and follows from two identities of information geometry: the generalized Pythagorean theorem and a dual e-mixture variance identity. As an analytically tractable demonstration, we apply the framework to $ε$-PCA, a regularized principal component analysis in which the empirical covariance is truncated at rank $N_K$ and discarded directions are pinned at a fixed noise floor $ε$. Although rank-constrained $ε$-PCA is not itself e-flat, it admits a technical reformulation with the same total GE on isotropic Gaussian data, under which each component of the decomposition takes closed form. The optimal rank emerges as the cutoff $λ_{\mathrm{cut}}^{*} = ε$ -- the model retains exactly those empirical eigenvalues exceeding the noise floor -- with the cutoff reflecting a marginal-rate balance between model-error gain and data-bias cost. A boundary comparison further yields a three-regime phase diagram -- retain-all, interior, and collapse -- separated by the lower Marchenko--Pastur edge and an analytically computable collapse threshold $ε_{*}(α)$, where $α$ is the dimension-to-sample-size ratio. All claims are verified numerically. --- ### Identifying and Mitigating Gender Cues in Academic Recommendation Letters: An Interpretability Case Study Authors: Charlotte S. Alexander, Shane Storks, Souradip Pal, Sayak Chakrabarty, Arushi Sharma, Mlen-Too Wesley, Bailey Russo Published: 2026-04-14 Categories: cs.LG, cs.CY arXiv: https://arxiv.org/abs/2604.12337v1 Letters of recommendation (LoRs) can carry patterns of implicitly gendered language that can inadvertently influence downstream decisions, e.g. in hiring and admissions. In this work, we investigate the extent to which Transformer-based encoder models as well as Large Language Models (LLMs) can infer the gender of applicants in academic LoRs submitted to an U.S. medical-residency program after explicit identifiers like names and pronouns are de-gendered. While using three models (DistilBERT, RoBERTa, and Llama 2) to classify the gender of anonymized and de-gendered LoRs, significant gender leakage was observed as evident from up to 68% classification accuracy. Text interpretation methods, like TF-IDF and SHAP, demonstrate that certain linguistic patterns are strong proxies for gender, e.g. "emotional'' and "humanitarian'' are commonly associated with LoRs from female applicants. As an experiment in creating truly gender-neutral LoRs, these implicit gender cues were remove resulting in a drop of up to 5.5% accuracy and 2.7% macro $F_1$ score on re-training the classifiers. However, applicant gender prediction still remains better than chance. In this case study, our findings highlight that 1) LoRs contain gender-identifying cues that are hard to remove and may activate bias in decision-making and 2) while our technical framework may be a concrete step toward fairer academic and professional evaluations, future work is needed to interrogate the role that gender plays in LoR review. Taken together, our findings motivate upstream auditing of evaluative text in real-world academic letters of recommendation as a necessary complement to model-level fairness interventions. --- ### GeM-EA: A Generative and Meta-learning Enhanced Evolutionary Algorithm for Streaming Data-Driven Optimization Authors: Yue Wu, Yuan-Ting Zhong, Ze-Yuan Ma, Yue-Jiao Gong Published: 2026-04-14 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2604.12336v1 Streaming Data-Driven Optimization (SDDO) problems arise in many applications where data arrive continuously and the optimization environment evolves over time. Concept drift produces non-stationary landscapes, making optimization methods challenging due to outdated models. Existing approaches often rely on simple surrogate combinations or directly injecting solutions, which may cause negative transfer under sudden environmental changes. We propose GeM-EA, a Generative and Meta-learning Enhanced Evolutionary Algorithm for SDDO that unifies meta-learned surrogate adaptation with generative replay for effective evolutionary search. Upon detecting concept drift, a bi-level meta-learning strategy rapidly initializes the surrogate using environment-relevant priors, while a linear residual component captures global trends. A multi-island evolutionary strategy further leverages historical knowledge via generative replay to accelerate optimization. Experimental results on benchmark SDDO problems demonstrate that GeM-EA achieves faster adaptation and improved robustness compared with state-of-the-art methods. --- ### SynMulti: Synthetic-to-Real Learning for Multimodal Video Understanding Authors: Tanzila Rahman, Renjie Liao, Leonid Sigal Published: 2026-04-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.12335v2 Training multimodal large language models (MLLMs) for video understanding requires large-scale annotated data spanning diverse tasks such as object counting, question answering, and segmentation. However, collecting and annotating multimodal video data in real-world is costly, slow, and inherently limited in diversity and coverage. To address this challenge, we propose~\textbf{SynMulti} dataset, along with a unified synthetic data generation pipeline capable of automatically producing unlimited multimodal video data with rich and diverse supervision. Our framework supports multiple task formats within a single pipeline, enabling scalable and consistent data creation across tasks. To further enhance reasoning ability, we introduce a VQA-based fine-tuning strategy that trains models to answer structured questions about visual content rather than relying solely on captions or simple instructions. This formulation encourages deeper visual grounding and reasoning. We evaluate our approach in three challenging tasks: video object counting, video-based visual question answering, and video object segmentation. Experimental results demonstrate that models trained predominantly on synthetic data generalize effectively to real-world datasets, often outperforming traditionally trained counterparts. Our findings highlight the potential of unified synthetic data pipelines as a scalable alternative to expensive real-world annotation for multimodal video understanding. --- ### All in One: A Unified Synthetic Data Pipeline for Multimodal Video Understanding Authors: Tanzila Rahman, Renjie Liao, Leonid Sigal Published: 2026-04-14 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.12335v1 Training multimodal large language models (MLLMs) for video understanding requires large-scale annotated data spanning diverse tasks such as object counting, question answering, and segmentation. However, collecting and annotating multimodal video data in real-world is costly, slow, and inherently limited in diversity and coverage. To address this challenge, we propose a unified synthetic data generation pipeline capable of automatically producing unlimited multimodal video data with rich and diverse supervision. Our framework supports multiple task formats within a single pipeline, enabling scalable and consistent data creation across tasks. To further enhance reasoning ability, we introduce a VQA-based fine-tuning strategy that trains models to answer structured questions about visual content rather than relying solely on captions or simple instructions. This formulation encourages deeper visual grounding and reasoning. We evaluate our approach in three challenging tasks: video object counting, video-based visual question answering, and video object segmentation. Experimental results demonstrate that models trained predominantly on synthetic data generalize effectively to real-world datasets, often outperforming traditionally trained counterparts. Our findings highlight the potential of unified synthetic data pipelines as a scalable alternative to expensive real-world annotation for multimodal video understanding. --- ### HyperLiDAR: Adaptive Post-Deployment LiDAR Segmentation via Hyperdimensional Computing Authors: Ivannia Gomez Moreno, Yi Yao, Ye Tian, Xiaofan Yu, Flavio Ponzina, Michael Sullivan, Jingyi Zhang, Mingyu Yang, Hun Seok Kim, Tajana Rosing Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12331v1 LiDAR semantic segmentation plays a pivotal role in 3D scene understanding for edge applications such as autonomous driving. However, significant challenges remain for real-world deployments, particularly for on-device post-deployment adaptation. Real-world environments can shift as the system navigates through different locations, leading to substantial performance degradation without effective and timely model adaptation. Furthermore, edge systems operate under strict computational and energy constraints, making it infeasible to adapt conventional segmentation models (based on large neural networks) directly on-device. To address the above challenges, we introduce HyperLiDAR, the first lightweight, post-deployment LiDAR segmentation framework based on Hyperdimensional Computing (HDC). The design of HyperLiDAR fully leverages the fast learning and high efficiency of HDC, inspired by how the human brain processes information. To further improve the adaptation efficiency, we identify the high data volume per scan as a key bottleneck and introduce a buffer selection strategy that focuses learning on the most informative points. We conduct extensive evaluations on two state-of-the-art LiDAR segmentation benchmarks and two representative devices. Our results show that HyperLiDAR outperforms or achieves comparable adaptation performance to state-of-the-art segmentation methods, while achieving up to a 13.8x speedup in retraining. --- ### Black-Box Optimization From Small Offline Datasets via Meta Learning with Synthetic Tasks Authors: Azza Fadhel, The Hung Tran, Trong Nghia Hoang, Jana Doppa Published: 2026-04-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.12325v3 We consider the problem of offline black-box optimization, where the goal is to discover optimal designs (e.g., molecules or materials) from past experimental data. A key challenge in this setting is data scarcity: in many scientific applications, only small or poor-quality datasets are available, which severely limits the effectiveness of existing algorithms. Prior work has theoretically and empirically shown that performance of offline optimization algorithms depends on how well the surrogate model captures the optimization bias (i.e., ability to rank input designs correctly), which is challenging to accomplish with limited experimental data. This paper proposes Surrogate Learning with Optimization Bias via Synthetic Task Generation (OptBias), a meta-learning framework that directly tackles data scarcity. OptBias learns a reusable optimization bias by training on synthetic tasks generated from a Gaussian process, and then fine-tunes the surrogate model on the small data for the target task. Across diverse continuous and discrete offline optimization benchmarks, OptBias consistently outperforms state-of-the-art baselines in small data regimes. These results highlight OptBias as a robust and practical solution for offline optimization in realistic small data settings. --- ### Self-Adversarial One Step Generation via Condition Shifting Authors: Deyuan Liu, Peng Sun, Yansen Han, Zhenglin Cheng, Chuyan Chen, Tao Lin Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12322v1 The push for efficient text to image synthesis has moved the field toward one step sampling, yet existing methods still face a three way tradeoff among fidelity, inference speed, and training efficiency. Approaches that rely on external discriminators can sharpen one step performance, but they often introduce training instability, high GPU memory overhead, and slow convergence, which complicates scaling and parameter efficient tuning. In contrast, regression based distillation and consistency objectives are easier to optimize, but they typically lose fine details when constrained to a single step. We present APEX, built on a key theoretical insight: adversarial correction signals can be extracted endogenously from a flow model through condition shifting. Using a transformation creates a shifted condition branch whose velocity field serves as an independent estimator of the model's current generation distribution, yielding a gradient that is provably GAN aligned, replacing the sample dependent discriminator terms that cause gradient vanishing. This discriminator free design is architecture preserving, making APEX a plug and play framework compatible with both full parameter and LoRA based tuning. Empirically, our 0.6B model surpasses FLUX-Schnell 12B (20$\times$ more parameters) in one step quality. With LoRA tuning on Qwen-Image 20B, APEX reaches a GenEval score of 0.89 at NFE=1 in 6 hours, surpassing the original 50-step teacher (0.87) and providing a 15.33$\times$ inference speedup. Code is available https://github.com/LINs-lab/APEX. --- ### ToxiTrace: Gradient-Aligned Training for Explainable Chinese Toxicity Detection Authors: Boyang Li, Hongzhe Shou, Yuanyuan Liang, Jingbin Zhang, Fang Zhou Published: 2026-04-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12321v1 Existing Chinese toxic content detection methods mainly target sentence-level classification but often fail to provide readable and contiguous toxic evidence spans. We propose \textbf{ToxiTrace}, an explainability-oriented method for BERT-style encoders with three components: (1) \textbf{CuSA}, which refines encoder-derived saliency cues into fine-grained toxic spans with lightweight LLM guidance; (2) \textbf{GCLoss}, a gradient-constrained objective that concentrates token-level saliency on toxic evidence while suppressing irrelevant activations; and (3) \textbf{ARCL}, which constructs sample-specific contrastive reasoning pairs to sharpen the semantic boundary between toxic and non-toxic content. Experiments show that ToxiTrace improves classification accuracy and toxic span extraction while preserving efficient encoder-based inference and producing more coherent, human-readable explanations. We have released the model at https://huggingface.co/ArdLi/ToxiTrace. --- ### EgoEsportsQA: An Egocentric Video Benchmark for Perception and Reasoning in Esports Authors: Jianzhe Ma, Zhonghao Cao, Shangkui Chen, Yichen Xu, Wenxuan Wang, Qin Jin Published: 2026-04-14 Categories: cs.CV, cs.AI, cs.MM arXiv: https://arxiv.org/abs/2604.12320v2 While video large language models (Video-LLMs) excel in understanding slow-paced, real-world egocentric videos, their capabilities in high-velocity, information-dense virtual environments remain under-explored. Existing benchmarks focus on daily activities, yet lack a rigorous testbed for evaluating fast, rule-bound reasoning in virtual scenarios. To fill this gap, we introduce EgoEsportsQA, a pioneering video question-answering (QA) benchmark for grounding perception and reasoning in expert esports knowledge. We curate 1,745 high-quality QA pairs from professional matches across 3 first-person shooter games via a scalable six-stage pipeline. These questions are structured into a two-dimensional decoupled taxonomy: 11 sub-tasks in the cognitive capability dimension (covering perception and reasoning levels) and 6 sub-tasks in the esports knowledge dimension. Comprehensive evaluations of state-of-the-art Video-LLMs reveal that current models still fail to achieve satisfactory performance, with the best model only 71.58%. The results expose notable gaps across both axes: models exhibit stronger capabilities in basic visual perception than in deep tactical reasoning, and they grasp overall macro-progression better than fine-grained micro-operations. Extensive ablation experiments demonstrate the intrinsic weaknesses of current Video-LLM architectures. Further analysis suggests that our dataset not only reveals the connections between real-world and virtual egocentric domains, but also offers guidance for optimizing downstream esports applications, thereby fostering the future advancement of Video-LLMs in various egocentric environments. --- ### RSGMamba: Reliability-Aware Self-Gated State Space Model for Multimodal Semantic Segmentation Authors: Guoan Xu, Yang Xiao, Guangwei Gao, Dongchen Zhu, Guo-Jun Qi, Wenjing Jia Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12319v2 Multimodal semantic segmentation has emerged as a powerful paradigm for enhancing scene understanding by leveraging complementary information from multiple sensing modalities (e.g., RGB, depth, and thermal). However, existing cross-modal fusion methods often implicitly assume that all modalities are equally reliable, which can lead to feature degradation when auxiliary modalities are noisy, misaligned, or incomplete. In this paper, we revisit cross-modal fusion from the perspective of modality reliability and propose a novel framework termed the Reliability-aware Self-Gated State Space Model (RSGMamba). At the core of our method is the Reliability-aware Self-Gated Mamba Block (RSGMB), which explicitly models modality reliability and dynamically regulates cross-modal interactions through a self-gating mechanism. Unlike conventional fusion strategies that indiscriminately exchange information across modalities, RSGMB enables reliability-aware feature selection and enhancing informative feature aggregation. In addition, a lightweight Local Cross-Gated Modulation (LCGM) is incorporated to refine fine-grained spatial details, complementing the global modeling capability of RSGMB. Extensive experiments demonstrate that RSGMamba achieves state-of-the-art performance on both RGB-D and RGB-T semantic segmentation benchmarks, resulting 58.8% / 54.0% mIoU on NYUDepth V2 and SUN-RGBD (+0.4% / +0.7% over prior best), and 61.1% / 88.9% mIoU on MFNet and PST900 (up to +1.6%), with only 48.6M parameters, thereby validating the effectiveness and superiority of the proposed approach. --- ### Cell Instance Segmentation via Multi-Task Image-to-Image Schrödinger Bridge Authors: Hayato Inoue, Shota Harada, Shumpei Takezaki, Ryoma Bise Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12318v1 Existing cell instance segmentation pipelines typically combine deterministic predictions with post-processing, which imposes limited explicit constraints on the global structure of instance masks. In this work, we propose a multi-task image-to-image Schrödinger Bridge framework that formulates instance segmentation as a distribution-based image-to-image generation problem. Boundary-aware supervision is integrated through a reverse distance map, and deterministic inference is employed to produce stable predictions. Experimental results on the PanNuke dataset demonstrate that the proposed method achieves competitive or superior performance without relying on SAM pre-training or additional post-processing. Additional results on the MoNuSeg dataset show robustness under limited training data. These findings indicate that Schrödinger Bridge-based image-to-image generation provides an effective framework for cell instance segmentation. --- ### GTPBD-MM: A Global Terraced Parcel and Boundary Dataset with Multi-Modality Authors: Zhiwei Zhang, Xingyuan Zeng, Xinkai Kong, Kunquan Zhang, Haoyuan Liang, Bohan Shi, Juepeng Zheng, Jianxi Huang, Yutong Lu, Haohuan Fu Published: 2026-04-14 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2604.12315v1 Agricultural parcel extraction plays an important role in remote sensing-based agricultural monitoring, supporting parcel surveying, precision management, and ecological assessment. However, existing public benchmarks mainly focus on regular and relatively flat farmland scenes. In contrast, terraced parcels in mountainous regions exhibit stepped terrain, pronounced elevation variation, irregular boundaries, and strong cross-regional heterogeneity, making parcel extraction a more challenging problem that jointly requires visual recognition, semantic discrimination, and terrain-aware geometric understanding. Although recent studies have advanced visual parcel benchmarks and image-text farmland understanding, a unified benchmark for complex terraced parcel extraction under aligned image-text-DEM settings remains absent. To fill this gap, we present GTPBD-MM, the first multimodal benchmark for global terraced parcel extraction. Built upon GTPBD, GTPBD-MM integrates high-resolution optical imagery, structured text descriptions, and DEM data, and supports systematic evaluation under Image-only, Image+Text, and Image+Text+DEM settings. We further propose Elevation and Text guided Terraced parcel network (ETTerra), a multimodal baseline for terraced parcel delineation. Extensive experiments demonstrate that textual semantics and terrain geometry provide complementary cues beyond visual appearance alone, yielding more accurate, coherent, and structurally consistent delineation results in complex terraced scenes. --- ### CompliBench: Benchmarking LLM Judges for Compliance Violation Detection in Dialogue Systems Authors: Jingbo Yang, Guanyu Yao, Bairu Hou, Xinghan Yang, Nikolai Glushnev, Iwona Bialynicka-Birula, Duo Ding, Shiyu Chang Published: 2026-04-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12312v1 As Large Language Models (LLMs) are increasingly deployed as task-oriented agents in enterprise environments, ensuring their strict adherence to complex, domain-specific operational guidelines is critical. While utilizing an LLM-as-a-Judge is a promising solution for scalable evaluation, the reliability of these judges in detecting specific policy violations remains largely unexplored. This gap is primarily due to the lack of a systematic data generation method, which has been hindered by the extensive cost of fine-grained human annotation and the difficulty of synthesizing realistic agent violations. In this paper, we introduce CompliBench, a novel benchmark designed to evaluate the ability of LLM judges to detect and localize guideline violations in multi-turn dialogues. To overcome data scarcity, we develop a scalable, automated data generation pipeline that simulates user-agent interactions. Our controllable flaw injection process automatically yields precise ground-truth labels for the violated guideline and the exact conversation turn, while an adversarial search method ensures these introduced perturbations are highly challenging. Our comprehensive evaluation reveals that current state-of-the-art proprietary LLMs struggle significantly with this task. In addition, we demonstrate that a small-scale judge model fine-tuned on our synthesized data outperforms leading LLMs and generalizes well to unseen business domains, highlighting our pipeline as an effective foundation for training robust generative reward models. --- ### Is Vibe Coding the Future? An Empirical Assessment of LLM Generated Codes for Construction Safety Authors: S M Jamil Uddin Published: 2026-04-14 Categories: cs.SE, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2604.12311v2 The emergence of vibe coding, a paradigm where non-technical users instruct Large Language Models (LLMs) to generate executable codes via natural language, presents both significant opportunities and severe risks for the construction industry. While empowering construction personnel such as the safety managers, foremen, and workers to develop tools and software, the probabilistic nature of LLMs introduces the threat of silent failures, wherein generated code compiles perfectly but executes flawed mathematical safety logic. This study empirically evaluates the reliability, software architecture, and domain-specific safety fidelity of 450 vibe-coded Python scripts generated by three frontier models, Claude 3.5 Haiku, GPT-4o-Mini, and Gemini 2.5 Flash. Utilizing a persona-driven prompt dataset (n=150) and a bifurcated evaluation pipeline comprising isolated dynamic sandboxing and an LLM-as-a-Judge, the research quantifies the severe limits of zero-shot vibe codes for construction safety. The findings reveal a highly significant relationship between user persona and data hallucination, demonstrating that less formal prompts drastically increase the AI's propensity to invent missing safety variables. Furthermore, while the models demonstrated high foundational execution viability (~85%), this syntactic reliability actively masked logic deficits and a severe lack of defensive programming. Among successfully executed scripts, the study identified an alarming ~45% overall Silent Failure Rate, with GPT-4o-Mini generating mathematically inaccurate outputs in ~56% of its functional code. The results demonstrate that current LLMs lack the deterministic rigor required for standalone safety engineering, necessitating the adoption of deterministic AI wrappers and strict governance for cyber-physical deployments. --- ### Towards Realistic and Consistent Orbital Video Generation via 3D Foundation Priors Authors: Rong Wang, Ruyi Zha, Ziang Cheng, Jiayu Yang, Pulak Purkait, Hongdong Li Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12309v1 We present a novel method for generating geometrically realistic and consistent orbital videos from a single image of an object. Existing video generation works mostly rely on pixel-wise attention to enforce view consistency across frames. However, such mechanism does not impose sufficient constraints for long-range extrapolation, e.g. rear-view synthesis, in which pixel correspondences to the input image are limited. Consequently, these works often fail to produce results with a plausible and coherent structure. To tackle this issue, we propose to leverage rich shape priors from a 3D foundational generative model as an auxiliary constraint, motivated by its capability of modeling realistic object shape distributions learned from large 3D asset corpora. Specifically, we prompt the video generation with two scales of latent features encoded by the 3D foundation model: (i) a denoised global latent vector as an overall structural guidance, and (ii) a set of latent images projected from volumetric features to provide view-dependent and fine-grained geometry details. In contrast to commonly used 2.5D representations such as depth or normal maps, these compact features can model complete object shapes, and help to improve inference efficiency by avoiding explicit mesh extraction. To achieve effective shape conditioning, we introduce a multi-scale 3D adapter to inject feature tokens to the base video model via cross-attention, which retains its capabilities from general video pretraining and enables a simple and model-agonistic fine-tuning process. Extensive experiments on multiple benchmarks show that our method achieves superior visual quality, shape realism and multi-view consistency compared to state-of-the-art methods, and robustly generalizes to complex camera trajectories and in-the-wild images. --- ### ContextLens: Modeling Imperfect Privacy and Safety Context for Legal Compliance Authors: Haoran Li, Yulin Chen, Huihao Jing, Wenbin Hu, Tsz Ho Li, Chanhou Lou, Hong Ting Tsang, Sirui Han, Yangqiu Song Published: 2026-04-14 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.12308v1 Individuals' concerns about data privacy and AI safety are highly contextualized and extend beyond sensitive patterns. Addressing these issues requires reasoning about the context to identify and mitigate potential risks. Though researchers have widely explored using large language models (LLMs) as evaluators for contextualized safety and privacy assessments, these efforts typically assume the availability of complete and clear context, whereas real-world contexts tend to be ambiguous and incomplete. In this paper, we propose ContextLens, a semi-rule-based framework that leverages LLMs to ground the input context in the legal domain and explicitly identify both known and unknown factors for legal compliance. Instead of directly assessing safety outcomes, our ContextLens instructs LLMs to answer a set of crafted questions that span over applicability, general principles and detailed provisions to assess compliance with pre-defined priorities and rules. We conduct extensive experiments on existing compliance benchmarks that cover the General Data Protection Regulation (GDPR) and the EU AI Act. The results suggest that our ContextLens can significantly improve LLMs' compliance assessment and surpass existing baselines without any training. Additionally, our ContextLens can further identify the ambiguous and missing factors. --- ### Boosting Robust AIGI Detection with LoRA-based Pairwise Training Authors: Ruiyang Xia, Qi Zhang, Yaowen Xu, Zhaofan Zou, Hao Sun, Zhongjiang He, Xuelong Li Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12307v1 The proliferation of highly realistic AI-Generated Image (AIGI) has necessitated the development of practical detection methods. While current AIGI detectors perform admirably on clean datasets, their detection performance frequently decreases when deployed "in the wild", where images are subjected to unpredictable, complex distortions. To resolve the critical vulnerability, we propose a novel LoRA-based Pairwise Training (LPT) strategy designed specifically to achieve robust detection for AIGI under severe distortions. The core of our strategy involves the targeted finetuning of a visual foundation model, the deliberate simulation of data distribution during the training phase, and a unique pairwise training process. Specifically, we introduce distortion and size simulations to better fit the distribution from the validation and test sets. Based on the strong visual representation capability of the visual foundation model, we finetune the model to achieve AIGI detection. The pairwise training is utilized to improve the detection via decoupling the generalization and robustness optimization. Experiments show that our approach secured the 3th placement in the NTIRE Robust AI-Generated Image Detection in the Wild challenge --- ### GCA Framework: A GCC Countries-Grounded Dataset and Agentic Pipeline for Climate Decision Support Authors: Muhammad Umer Sheikh, Khawar Shehzad, Salman Khan, Fahad Shahbaz Khan, Muhammad Haris Khan Published: 2026-04-14 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.12306v3 Climate decision-making in the GCC states increasingly demands systems that can translate heterogeneous scientific and policy evidence into actionable guidance, yet general-purpose large language models (LLMs) remain weak both in region-specific climate knowledge and grounded interaction with geospatial and forecasting tools. We present the GCA framework, which unifies (i) GCA-DS, a curated multimodal dataset grounded in the GCC states, and (ii) Gulf Climate Agent (GCA), a tool-augmented agent for climate analysis. GCA-DS comprises 200k question--answer pairs spanning governmental policies and adaptation plans, NGO and international frameworks, academic literature, and event-driven reporting on heatwaves, dust storms, and floods, complemented with remote-sensing inputs that couple imagery with textual evidence. Building on this foundation, the GCA agent orchestrates a modular tool pipeline grounded in real-time and historical signals and geospatial processing that produces derived indices and interpretable visualizations. Finally, we benchmark open and proprietary LLMs on climate tasks in the GCC states and show that domain fine-tuning and tool integration substantially improve reliability over general-purpose baselines. --- ### CBAM-Enhanced DenseNet121 for Multi-Class Chest X-Ray Classification with Grad-CAM Explainability Authors: Utsho Kumar Dey Published: 2026-04-14 Categories: eess.IV, cs.CV arXiv: https://arxiv.org/abs/2604.12305v1 Pneumonia remains a leading cause of childhood mortality worldwide, with a heavy burden in low-resource settings such as Bangladesh where radiologist availability is limited. Most existing deep learning approaches treat pneumonia detection as a binary problem, overlooking the clinically critical distinction between bacterial and viral aetiology. This paper proposes CBAM-DenseNet121, a transfer-learning framework that integrates the Convolutional Block Attention Module (CBAM) into DenseNet121 for three-class chest X-ray classification: Normal, Bacterial Pneumonia, and Viral Pneumonia. We also conduct a systematic binary-task baseline study revealing that EfficientNetB3 (73.88%) underperforms even the custom CNN baseline (78.53%) -- a practically important negative finding for medical imaging model selection. To ensure statistical reliability, all experiments were repeated three times with independent random seeds (42, 7, 123), and results are reported as mean +/- standard deviation. CBAM-DenseNet121 achieves 84.29% +/- 1.14% test accuracy with per-class AUC scores of 0.9565 +/- 0.0010, 0.9610 +/- 0.0014, and 0.9187 +/- 0.0037 for bacterial pneumonia, normal, and viral pneumonia respectively. Grad-CAM visualizations confirm that the model attends to anatomically plausible pulmonary regions for each class, supporting interpretable deployment in resource-constrained clinical environments. --- ### Beyond Weather Correlation: A Comparative Study of Static and Temporal Neural Architectures for Fine-Grained Residential Energy Consumption Forecasting in Melbourne, Australia Authors: Prasad Nimantha Madusanka Ukwatta Hewage, Hao Wu Published: 2026-04-14 Categories: cs.LG, eess.SY arXiv: https://arxiv.org/abs/2604.12304v1 Accurate short-term residential energy consumption forecasting at sub-hourly resolution is critical for smart grid management, demand response programmes, and renewable energy integration. While weather variables are widely acknowledged as key drivers of residential electricity demand, the relative merit of incorporating temporal autocorrelation - the sequential memory of past consumption; over static meteorological features alone remains underexplored at fine-grained (5-minute) temporal resolution for Australian households. This paper presents a rigorous empirical comparison of a Multilayer Perceptron (MLP) and a Long Short-Term Memory (LSTM) recurrent network applied to two real-world Melbourne households: House 3 (a standard grid-connected dwelling) and House 4 (a rooftop solar photovoltaic-integrated household). Both models are trained on 14 months of 5-minute interval smart meter data (March 2023-April 2024) merged with official Bureau of Meteorology (BOM) daily weather observations, yielding over 117,000 samples per household. The LSTM, operating on 24-step (2-hour) sliding consumption windows, achieves coefficients of determination of R^2 = 0.883 (House 3) and R^2 = 0.865 (House 4), compared to R^2 = -0.055 and R^2 = 0.410 for the corresponding weather-driven MLPs - differences of 93.8 and 45.5 percentage points. These results establish that temporal autocorrelation in the consumption sequence dominates meteorological information for short-term forecasting at 5-minute granularity. Additionally, we demonstrate an asymmetry introduced by solar generation: for the PV-integrated household, the MLP achieves R^2 = 0.410, revealing implicit solar forecasting from weather-time correlations. A persistence baseline analysis and seasonal stratification contextualise model performance. We propose a hybrid weather-augmented LSTM and federated learning extensions as directions for future work. --- ### Labeled TrustSet Guided: Batch Active Learning with Reinforcement Learning Authors: Guofeng Cui, Yang Liu, Pichao Wang, Hankai Hsu, Xiaohang Sun, Xiang Hao, Zhu Liu Published: 2026-04-14 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.12303v1 Batch active learning (BAL) is a crucial technique for reducing labeling costs and improving data efficiency in training large-scale deep learning models. Traditional BAL methods often rely on metrics like Mahalanobis Distance to balance uncertainty and diversity when selecting data for annotation. However, these methods predominantly focus on the distribution of unlabeled data and fail to leverage feedback from labeled data or the model's performance. To address these limitations, we introduce TrustSet, a novel approach that selects the most informative data from the labeled dataset, ensuring a balanced class distribution to mitigate the long-tail problem. Unlike CoreSet, which focuses on maintaining the overall data distribution, TrustSet optimizes the model's performance by pruning redundant data and using label information to refine the selection process. To extend the benefits of TrustSet to the unlabeled pool, we propose a reinforcement learning (RL)-based sampling policy that approximates the selection of high-quality TrustSet candidates from the unlabeled data. Combining TrustSet and RL, we introduce the Batch Reinforcement Active Learning with TrustSet (BRAL-T) framework. BRAL-T achieves state-of-the-art results across 10 image classification benchmarks and 2 active fine-tuning tasks, demonstrating its effectiveness and efficiency in various domains. --- ### Local-Splitter: A Measurement Study of Seven Tactics for Reducing Cloud LLM Token Usage on Coding-Agent Workloads Authors: Justice Owusu Agyemang, Jerry John Kponyo, Elliot Amponsah, Godfred Manu Addo Boakye, Kwame Opuni-Boachie Obour Agyekum Published: 2026-04-14 Categories: cs.DC, cs.AI, cs.SE arXiv: https://arxiv.org/abs/2604.12301v1 We present a systematic measurement study of seven tactics for reducing cloud LLM token usage when a small local model can act as a triage layer in front of a frontier cloud model. The tactics are: (1) local routing, (2) prompt compression, (3) semantic caching, (4) local drafting with cloud review, (5) minimal-diff edits, (6) structured intent extraction, and (7) batching with vendor prompt caching. We implement all seven in an open-source shim that speaks both MCP and the OpenAI-compatible HTTP surface, supporting any local model via Ollama and any cloud model via an OpenAI-compatible endpoint. We evaluate each tactic individually, in pairs, and in a greedy-additive subset across four coding-agent workload classes (edit-heavy, explanation-heavy, general chat, RAG-heavy). We measure tokens saved, dollar cost, latency, and routing accuracy. Our headline finding is that T1 (local routing) combined with T2 (prompt compression) achieves 45-79% cloud token savings on edit-heavy and explanation-heavy workloads, while on RAG-heavy workloads the full tactic set including T4 (draft-review) achieves 51% savings. We observe that the optimal tactic subset is workload-dependent, which we believe is the most actionable finding for practitioners deploying coding agents today. --- ### CoSyncDiT: Cognitive Synchronous Diffusion Transformer for Movie Dubbing Authors: Gaoxiang Cong, Liang Li, Jiaxin Ye, Zhedong Zhang, Hongming Shan, Yuankai Qi, Qingming Huang Published: 2026-04-14 Categories: cs.SD, cs.CV, cs.MM arXiv: https://arxiv.org/abs/2604.12292v1 Movie dubbing aims to synthesize speech that preserves the vocal identity of a reference audio while synchronizing with the lip movements in a target video. Existing methods fail to achieve precise lip-sync and lack naturalness due to explicit alignment at the duration level. While implicit alignment solutions have emerged, they remain susceptible to interference from the reference audio, triggering timbre and pronunciation degradation in in-the-wild scenarios. In this paper, we propose a novel flow matching-based movie dubbing framework driven by the Cognitive Synchronous Diffusion Transformer (CoSync-DiT), inspired by the cognitive process of professional actors. This architecture progressively guides the noise-to-speech generative trajectory by executing acoustic style adapting, fine-grained visual calibrating, and time-aware context aligning. Furthermore, we design the Joint Semantic and Alignment Regularization (JSAR) mechanism to simultaneously constrain frame-level temporal consistency on the contextual outputs and semantic consistency on the flow hidden states, ensuring robust alignment. Extensive experiments on both standard benchmarks and challenging in-the-wild dubbing benchmarks demonstrate that our method achieves the state-of-the-art performance across multiple metrics. --- ### Frontier-Eng: Benchmarking Self-Evolving Agents on Real-World Engineering Tasks with Generative Optimization Authors: Yizhe Chi, Deyao Hong, Dapeng Jiang, Tianwei Luo, Kaisen Yang, Boshi Zhang, Zhe Cao, Xiaoyan Fan, Bingxiang He, Han Hao, Weiyang Jin, Dianqiao Lei et al. Published: 2026-04-14 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.12290v2 Current LLM agent benchmarks, which predominantly focus on binary pass/fail tasks such as code generation or search-based question answering, often neglect the value of real-world engineering that is often captured through the iterative optimization of feasible designs. To this end, we introduce Frontier-Eng, a human-verified benchmark for generative optimization -- an iterative propose-execute-evaluate loop in which an agent generates candidate artifacts, receives executable verifier feedback, and revises them under a fixed interaction budget -- spanning $47$ tasks across five broad engineering categories. Unlike previous suites, Frontier-Eng tasks are grounded in industrial-grade simulators and verifiers that provide continuous reward signals and enforce hard feasibility constraints under constrained budgets. We evaluate eight frontier language models using representative search frameworks, finding that while GPT 5.4 achieves the most robust performance, the benchmark remains challenging for all models. Our analysis suggests a dual power-law decay in improvement frequency ($\sim$ 1/iteration) and magnitude ($\sim$ 1/improvement count). We further show that although width improves parallelism and diversity, depth remains crucial for hard-won improvements under a fixed budget. Frontier-Eng establishes a new standard for assessing the capacity of AI agents to integrate domain knowledge with executable feedback to solve complex, open-ended engineering problems. --- ### The Enforcement and Feasibility of Hate Speech Moderation Authors: Manuel Tonneau, Dylan Thurgood, Diyi Liu, Niyati Malhotra, Victor Orozco-Olvera, Ralph Schroeder, Scott A. Hale, Manoel Horta Ribeiro, Paul Röttger, Samuel P. Fraiberger Published: 2026-04-14 Categories: cs.CY, cs.CL arXiv: https://arxiv.org/abs/2604.12289v2 Online hate speech is associated with harms ranging from deteriorating mental health to violence, yet how consistently platforms moderate hate, and whether enforcement is feasible at scale, remain poorly understood. We audit hate speech moderation on Twitter (now X) using 540,000 tweets annotated by trained native speakers, representative of a full day on the platform. Five months after posting, 80% of hateful tweets, including violent ones, remained online. Removal was only marginally more likely than for non-hateful tweets, far below scams or adult content, and insensitive to severity and reach. Automated detection could not reliably classify hate but ranked it highly, enabling human triage. Simulating this workflow, current staffing curbed little exposure, yet substantial reductions proved financially feasible, far below applicable regulatory fines. Persistent hate reflects resource allocation, not technical limits. --- ### The Enforcement and Feasibility of Hate Speech Moderation on Twitter Authors: Manuel Tonneau, Dylan Thurgood, Diyi Liu, Niyati Malhotra, Victor Orozco-Olvera, Ralph Schroeder, Scott A. Hale, Manoel Horta Ribeiro, Paul Röttger, Samuel P. Fraiberger Published: 2026-04-14 Categories: cs.CY, cs.CL arXiv: https://arxiv.org/abs/2604.12289v1 Online hate speech is associated with substantial social harms, yet it remains unclear how consistently platforms enforce hate speech policies or whether enforcement is feasible at scale. We address these questions through a global audit of hate speech moderation on Twitter (now X). Using a complete 24-hour snapshot of public tweets, we construct representative samples comprising 540,000 tweets annotated for hate speech by trained annotators across eight major languages. Five months after posting, 80% of hateful tweets remain online, including explicitly violent hate speech. Such tweets are no more likely to be removed than non-hateful tweets, with neither severity nor visibility increasing the likelihood of removal. We then examine whether these enforcement gaps reflect technical limits of large-scale moderation systems. While fully automated detection systems cannot reliably identify hate speech without generating large numbers of false positives, they effectively prioritize likely violations for human review. Simulations of a human-AI moderation pipeline indicate that substantially reducing user exposure to hate speech is economically feasible at a cost below existing regulatory penalties. These results suggest that the persistence of online hate cannot be explained by technical constraints alone but also reflects institutional choices in the allocation of moderation resources. --- ### SMART Fine-tuning Factor Augmented Neural Lasso Authors: Jinhang Chai, Jianqing Fan, Cheng Gao, Qishuo Yin Published: 2026-04-14 Categories: stat.ML, cs.LG, stat.ME arXiv: https://arxiv.org/abs/2604.12288v2 Fine-tuning is a widely used strategy for adapting pre-trained models to new tasks, yet its methodology and theoretical properties in high-dimensional nonparametric settings with variable selection have not yet been developed. We propose a source-model-augmented residual tuning (SMART) framework, which incorporates the pre-trained source model as an augmented feature into the target learner and estimates only the residual target-specific component. The approach is widely applicable, from parametric and sparse models to neural networks and blackbox machine learning models. We focus on the development of fine-tuning factor-augmented neural Lasso, resulting in SMART-FAN-Lasso. This transfer-learning framework for high-dimensional nonparametric regression with variable selection simultaneously handles covariate and posterior shifts. We use a low-rank factor structure to manage high-dimensional dependent covariates and a residual tuning decomposition in which the target function is expressed as a function of source model and other target-specific variables, thereby reducing the effective complexity of the target task. We derive minimax-optimal excess risk bounds, characterizing the precise conditions, in terms of relative sample sizes and function complexities, under which fine-tuning yields statistical acceleration over single-task learning. Extensive numerical experiments across diverse covariate- and posterior-shift scenarios demonstrate that SMART-FAN-Lasso consistently outperforms standard baselines and achieves near-oracle performance even under severe target sample size constraints, empirically validating the derived rates. --- ### LiveMoments: Reselected Key Photo Restoration in Live Photos via Reference-guided Diffusion Authors: Clara Xue, Zizheng Yan, Zhenning Shi, Yuhang Yu, Jingyu Zhuang, Qi Zhang, Jinwei Chen, Qingnan Fan Published: 2026-04-14 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.12286v1 Live Photo captures both a high-quality key photo and a short video clip to preserve the precious dynamics around the captured moment. While users may choose alternative frames as the key photo to capture better expressions or timing, these frames often exhibit noticeable quality degradation, as the photo capture ISP pipeline delivers significantly higher image quality than the video pipeline. This quality gap highlights the need for dedicated restoration techniques to enhance the reselected key photo. To this end, we propose LiveMoments, a reference-guided image restoration framework tailored for the reselected key photo in Live Photos. Our method employs a two-branch neural network: a reference branch that extracts structural and textural information from the original high-quality key photo, and a main branch that restores the reselected frame using the guidance provided by the reference branch. Furthermore, we introduce a unified Motion Alignment module that incorporates motion guidance for spatial alignment at both the latent and image levels. Experiments on real and synthetic Live Photos demonstrate that LiveMoments significantly improves perceptual quality and fidelity over existing solutions, especially in scenes with fast motion or complex structures. Our code is available at https://github.com/OpenVeraTeam/LiveMoments. --- ### GAM: Hierarchical Graph-based Agentic Memory for LLM Agents Authors: Zhaofen Wu, Hanrong Zhang, Fulin Lin, Wujiang Xu, Xinran Xu, Yankai Chen, Henry Peng Zou, Shaowen Chen, Weizhi Zhang, Xue Liu, Philip S. Yu, Hongwei Wang Published: 2026-04-14 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.12285v1 To sustain coherent long-term interactions, Large Language Model (LLM) agents must navigate the tension between acquiring new information and retaining prior knowledge. Current unified stream-based memory systems facilitate context updates but remain vulnerable to interference from transient noise. Conversely, discrete structured memory architectures provide robust knowledge retention but often struggle to adapt to evolving narratives. To address this, we propose GAM, a hierarchical Graph-based Agentic Memory framework that explicitly decouples memory encoding from consolidation to effectively resolve the conflict between rapid context perception and stable knowledge retention. By isolating ongoing dialogue in an event progression graph and integrating it into a topic associative network only upon semantic shifts, our approach minimizes interference while preserving long-term consistency. Additionally, we introduce a graph-guided, multi-factor retrieval strategy to enhance context precision. Experiments on LoCoMo and LongDialQA indicate that our method consistently outperforms state-of-the-art baselines in both reasoning accuracy and efficiency. ---