# Research Papers — 2026-04 · bundle page 185 of 233 > Full-text markdown of papers #9201–9250 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/185 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/184.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/186.md) ### Selective Aggregation of Attention Maps Improves Diffusion-Based Visual Interpretation Authors: Jungwon Park, Jungmin Ko, Dongnam Byun, Wonjong Rhee Published: 2026-04-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.05906v1 Numerous studies on text-to-image (T2I) generative models have utilized cross-attention maps to boost application performance and interpret model behavior. However, the distinct characteristics of attention maps from different attention heads remain relatively underexplored. In this study, we show that selectively aggregating cross-attention maps from heads most relevant to a target concept can improve visual interpretability. Compared to the diffusion-based segmentation method DAAM, our approach achieves higher mean IoU scores. We also find that the most relevant heads capture concept-specific features more accurately than the least relevant ones, and that selective aggregation helps diagnose prompt misinterpretations. These findings suggest that attention head selection offers a promising direction for improving the interpretability and controllability of T2I generation. --- ### Neural Parameter Estimation of RC Thermal Building Models for Model Predictive Control Authors: Fabian Raisch, Timo Germann, Sang-Woo Ham, J. Nathan Kutz, Christoph Goebel, Benjamin Tischler Published: 2026-04-07 Categories: eess.SY, cs.LG arXiv: https://arxiv.org/abs/2604.05904v2 Gray-box RC models are widely used to enable energy-efficient model predictive control (MPC) in buildings. However, estimating RC parameters remains difficult, as conventional optimization-based algorithms are prone to local minima, rely heavily on good initial guesses, and incur high computational cost. To address these issues, we propose the Estimator from Scratch, a novel neural parameter estimation approach that embeds the physical equations into a neural network's training process to estimate RC parameters. To further improve estimation accuracy and eliminate dependence on an initial guess, we extend this approach by pretraining the neural network on data from multiple source buildings, the Pretrained Estimator. We benchmark both methods against a genetic-algorithm-based RC estimator and a fully black-box neural network. All methods are evaluated across eight simulated and three real-world buildings for two RC configurations, for both prediction accuracy and closed-loop MPC performance, the latter only for the simulated buildings. The Pretrained Estimator achieves the best prediction performance among all RC-based methods, particularly with little training data, and yields the lowest and least variable MPC costs across buildings and benchmarks. These results position the Pretrained Estimator as a robust, computationally efficient, and initial-guess-free alternative for RC parameter estimation, with potential to extend to other control-oriented dynamical systems. --- ### Transfer Learning for Neural Parameter Estimation applied to Building RC Models Authors: Fabian Raisch, Timo Germann, J. Nathan Kutz, Christoph Goebel, Benjamin Tischler Published: 2026-04-07 Categories: eess.SY, cs.LG arXiv: https://arxiv.org/abs/2604.05904v1 Parameter estimation for dynamical systems remains challenging due to non-convexity and sensitivity to initial parameter guesses. Recent deep learning approaches enable accurate and fast parameter estimation but do not exploit transferable knowledge across systems. To address this, we introduce a transfer-learning-based neural parameter estimation framework based on a pretraining-fine-tuning paradigm. This approach improves accuracy and eliminates the need for an initial parameter guess. We apply this framework to building RC thermal models, evaluating it against a Genetic Algorithm and a from-scratch neural baseline across eight simulated buildings, one real-world building, two RC model configurations, and four training data lengths. Results demonstrate an 18.6-24.0% performance improvement with only 12 days of training data and up to 49.4% with 72 days. Beyond buildings, the proposed method represents a new paradigm for parameter estimation in dynamical systems. --- ### AICA-Bench: Holistically Examining the Capabilities of VLMs in Affective Image Content Analysis Authors: Dong She, Xianrong Yao, Liqun Chen, Jinghe Yu, Yang Gao, Zhanpeng Jin Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05900v1 Vision-Language Models (VLMs) have demonstrated strong capabilities in perception, yet holistic Affective Image Content Analysis (AICA), which integrates perception, reasoning, and generation into a unified framework, remains underexplored. To address this gap, we introduce AICA-Bench, a comprehensive benchmark with three core tasks: Emotion Understanding (EU), Emotion Reasoning (ER), and Emotion-Guided Content Generation (EGCG). We evaluate 23 VLMs and identify two major limitations: weak intensity calibration and shallow open-ended descriptions. To address these issues, we propose Grounded Affective Tree (GAT) Prompting, a training-free framework that combines visual scaffolding with hierarchical reasoning. Experiments show that GAT reduces intensity errors and improves descriptive depth, providing a strong baseline for future research on affective multimodal understanding and generation. --- ### FRENCH-YMCA: A FRENCH Corpus meeting the language needs of Youth, froM Children to Adolescents Authors: Cherifa Ben Khelil, Jean-Yves Antoine, Anaïs Halftermeyer, Frédéric Rayar, Mathieu Thebaud Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05899v1 In this paper, we introduce the French-YMCA corpus, a new linguistic resource specifically tailored for children and adolescents. The motivation for building this corpus is clear: children have unique language requirements, as their language skills are in constant evolution and differ from those of adults. With an extensive collection of 39,200 text files, the French-YMCA corpus encompasses a total of 22,471,898 words. It distinguishes itself through its diverse sources, consistent grammar and spelling, and the commitment to providing open online accessibility for all. Such corpus can serve as the foundation for training language models that understand and anticipate youth's language, thereby enhancing the quality of digital interactions and ensuring that responses and suggestions are age-appropriate and adapted to the comprehension level of users of this age. --- ### Physics-Aware Video Instance Removal Benchmark Authors: Zirui Li, Xinghao Chen, Lingyu Jiang, Dengzhe Hou, Fangzhou Lin, Kazunori Yamada, Xiangbo Gao, Zhengzhong Tu Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05898v1 Video Instance Removal (VIR) requires removing target objects while maintaining background integrity and physical consistency, such as specular reflections and illumination interactions. Despite advancements in text-guided editing, current benchmarks primarily assess visual plausibility, often overlooking the physical causalities, such as lingering shadows, triggered by object removal. We introduce the Physics-Aware Video Instance Removal (PVIR) benchmark, featuring 95 high-quality videos annotated with instance-accurate masks and removal prompts. PVIR is partitioned into Simple and Hard subsets, the latter explicitly targeting complex physical interactions. We evaluate four representative methods, PISCO-Removal, UniVideo, DiffuEraser, and CoCoCo, using a decoupled human evaluation protocol across three dimensions to isolate semantic, visual, and spatial failures: instruction following, rendering quality, and edit exclusivity. Our results show that PISCO-Removal and UniVideo achieve state-of-the-art performance, while DiffuEraser frequently introduces blurring artifacts and CoCoCo struggles significantly with instruction following. The persistent performance drop on the Hard subset highlights the ongoing challenge of recovering complex physical side effects. --- ### A Tensor-Train Framework for Bayesian Inference in High-Dimensional Systems: Applications to MIMO Detection and Channel Decoding Authors: Luca Schmid, Dominik Sulz, Shrinivas Chimmalgi, Laurent Schmalen Published: 2026-04-07 Categories: cs.IT, cs.LG arXiv: https://arxiv.org/abs/2604.05890v1 Bayesian inference in high-dimensional discrete-input additive noise models is a fundamental challenge in communication systems, as the support of the required joint a posteriori probability (APP) mass function grows exponentially with the number of unknown variables. In this work, we propose a tensor-train (TT) framework for tractable, near-optimal Bayesian inference in discrete-input additive noise models. The central insight is that the joint log-APP mass function admits an exact low-rank representation in the TT format, enabling compact storage and efficient computations. To recover symbol-wise APP marginals, we develop a practical inference procedure that approximates the exponential of the log-posterior using a TT-cross algorithm initialized with a truncated Taylor-series. To demonstrate the generality of the approach, we derive explicit low-rank TT constructions for two canonical communication problems: the linear observation model under additive white Gaussian noise (AWGN), applied to multiple-input multiple-output (MIMO) detection, and soft-decision decoding of binary linear block error correcting codes over the binary-input AWGN channel. Numerical results show near-optimal error-rate performance across a wide range of signal-to-noise ratios while requiring only modest TT ranks. These results highlight the potential of tensor-network methods for efficient Bayesian inference in communication systems. --- ### HybridKV: Hybrid KV Cache Compression for Efficient Multimodal Large Language Model Inference Authors: Bowen Zeng, Feiyang Ren, Jun Zhang, Xiaoling Gu, Ke Chen, Lidan Shou, Huan Li Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05887v1 Multimodal Large Language Models (MLLMs) have advanced unified reasoning over text, images, and videos, but their inference is hindered by the rapid growth of key-value (KV) caches. Each visual input expands into thousands of tokens, causing caches to scale linearly with context length and remain resident in GPU memory throughout decoding, which leads to prohibitive memory overhead and latency even on high-end GPUs. A common solution is to compress caches under a fixed allocated budget at different granularities: token-level uniformly discards less important tokens, layer-level varies retention across layers, and head-level redistributes budgets across heads. Yet these approaches stop at allocation and overlook the heterogeneous behaviors of attention heads that require distinct compression strategies. We propose HybridKV, a hybrid KV cache compression framework that integrates complementary strategies in three stages: heads are first classified into static or dynamic types using text-centric attention; then a top-down budget allocation scheme hierarchically assigns KV budgets; finally, static heads are compressed by text-prior pruning and dynamic heads by chunk-wise retrieval. Experiments on 11 multimodal benchmarks with Qwen2.5-VL-7B show that HybridKV reduces KV cache memory by up to $7.9\times$ and achieves $1.52\times$ faster decoding, with almost no performance drop or even higher relative to the full-cache MLLM. --- ### Automatic dental superimposition of 3D intraorals and 2D photographs for human identification Authors: Antonio D. Villegas-Yeguas, Xavier Abreau-Freire, Guillermo R-García, Andrea Valsecchi, Teresa Pinho, Daniel Pérez-Mongiovi, Oscar Ibáñez, Oscar Cordón Published: 2026-04-07 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.05877v1 Dental comparison is considered a primary identification method, at the level of fingerprints and DNA profiling. One crucial but time-consuming step of this method is the morphological comparison. One of the main challenges to apply this method is the lack of ante-mortem medical records, specially on scenarios such as migrant death at the border and/or in countries where there is no universal healthcare. The availability of photos on social media where teeth are visible has led many odontologists to consider morphological comparison using them. However, state-of-the-art proposals have significant limitations, including the lack of proper modeling of perspective distortion and the absence of objective approaches that quantify morphological differences. Our proposal involves a 3D (post-mortem scan) - 2D (ante-mortem photos) approach. Using computer vision and optimization techniques, we replicate the ante-mortem image with the 3D model to perform the morphological comparison. Two automatic approaches have been developed: i) using paired landmarks and ii) using a segmentation of the teeth region to estimate camera parameters. Both are capable of obtaining very promising results over 20,164 cross comparisons from 142 samples, obtaining mean ranking values of 1.6 and 1.5, respectively. These results clearly outperform filtering capabilities of automatic dental chart comparison approaches, while providing an automatic, objective and quantitative score of the morphological correspondence, easily to interpret and analyze by visualizing superimposed images. --- ### Addressing the Reasoning Gap: Mechanistic Circuit-Based Knowledge Editing in Large Language Models Authors: Tianyi Zhao, Yinhan He, Wendy Zheng, Chen Chen Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05876v3 Deploying Large Language Models (LLMs) in real-world dynamic environments raises the challenge of updating their pre-trained knowledge. While existing knowledge editing methods can reliably patch isolated facts, they frequently suffer from a \textit{reasoning gap}, where the model recalls the edited fact but fails to utilize it in multi-step reasoning chains. Our analysis shows that multi-hop reasoning relies on sparse circuitry beyond direct factual recall, with functionally differentiated components contributing to intermediate processing and final-answer prediction. To bridge this gap, we introduce MCircKE (\underline{M}echanistic \underline{Circ}uit-based \underline{K}nowledge \underline{E}diting), a mechanistically grounded framework that enables a map-and-adapt editing procedure. MCircKE first identifies the causal circuits responsible for a specific reasoning task, capturing both the storage of the fact and the routing of its logical consequences. It then surgically update parameters exclusively within this mapped circuit. Extensive experiments on the MQuAKE-series benchmarks demonstrate the effectiveness of the proposed method for multi-hop reasoning in knowledge editing. --- ### Mechanistic Circuit-Based Knowledge Editing in Large Language Models Authors: Tianyi Zhao, Yinhan He, Wendy Zheng, Chen Chen Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05876v1 Deploying Large Language Models (LLMs) in real-world dynamic environments raises the challenge of updating their pre-trained knowledge. While existing knowledge editing methods can reliably patch isolated facts, they frequently suffer from a "Reasoning Gap", where the model recalls the edited fact but fails to utilize it in multi-step reasoning chains. To bridge this gap, we introduce MCircKE (\underline{M}echanistic \underline{Circ}uit-based \underline{K}nowledge \underline{E}diting), a novel framework that enables a precise "map-and-adapt" editing procedure. MCircKE first identifies the causal circuits responsible for a specific reasoning task, capturing both the storage of the fact and the routing of its logical consequences. It then surgically update parameters exclusively within this mapped circuit. Extensive experiments on the MQuAKE-3K benchmark demonstrate the effectiveness of the proposed method for multi-hop reasoning in knowledge editing. --- ### Joint Knowledge Base Completion and Question Answering by Combining Large Language Models and Small Language Models Authors: Yinan Liu, Dongying Lin, Sigang Luo, Xiaochun Yang, Bin Wang Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05875v1 Knowledge Bases (KBs) play a key role in various applications. As two representative KB-related tasks, knowledge base completion (KBC) and knowledge base question answering (KBQA) are closely related and inherently complementary with each other. Thus, it will be beneficial to solve the task of joint KBC and KBQA to make them reinforce each other. However, existing studies usually rely on the small language model (SLM) to enhance them jointly, and the large language model (LLM)'s strong reasoning ability is ignored. In this paper, by combining the strengths of the LLM with the SLM, we propose a novel framework JCQL, which can make these two tasks enhance each other in an iterative manner. To make KBC enhance KBQA, we augment the LLM agent-based KBQA model's reasoning paths by incorporating an SLM-trained KBC model as an action of the agent, alleviating the LLM's hallucination and high computational costs issue in KBQA. To make KBQA enhance KBC, we incrementally fine-tune the KBC model by leveraging KBQA's reasoning paths as its supplementary training data, improving the ability of the SLM in KBC. Extensive experiments over two public benchmark data sets demonstrate that JCQL surpasses all baselines for both KBC and KBQA tasks. --- ### Swiss-Bench 003: Evaluating LLM Reliability and Adversarial Security for Swiss Regulatory Contexts Authors: Fatih Uenal Published: 2026-04-07 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.05872v1 The deployment of large language models (LLMs) in Swiss financial and regulatory contexts demands empirical evidence of both production reliability and adversarial security, dimensions not jointly operationalized in existing Swiss-focused evaluation frameworks. This paper introduces Swiss-Bench 003 (SBP-003), extending the HAAS (Helvetic AI Assessment Score) from six to eight dimensions by adding D7 (Self-Graded Reliability Proxy) and D8 (Adversarial Security). I evaluate ten frontier models across 808 Swiss-specific items in four languages (German, French, Italian, English), comprising seven Swiss-adapted benchmarks (Swiss TruthfulQA, Swiss IFEval, Swiss SimpleQA, Swiss NIAH, Swiss PII-Scope, System Prompt Leakage, and Swiss German Comprehension) targeting FINMA Guidance 08/2024, the revised Federal Act on Data Protection (nDSG), and OWASP Top 10 for LLMs. Self-graded D7 scores (73-94%) exceed externally judged D8 security scores (20-61%) by a wide margin, though these dimensions use non-comparable scoring regimes. System prompt leakage resistance ranges from 24.8% to 88.2%, while PII extraction defense remains weak (14-42%) across all models. Qwen 3.5 Plus achieves the highest self-graded D7 score (94.4%), while GPT-oss 120B achieves the highest D8 score (60.7%) despite being the lowest-cost model evaluated. All evaluations are zero-shot under provider default settings; D7 is self-graded and does not constitute independently validated accuracy. I provide conceptual mapping tables relating benchmark dimensions to FINMA model validation requirements, nDSG data protection obligations, and OWASP LLM risk categories. --- ### Understanding Performance Gap Between Parallel and Sequential Sampling in Large Reasoning Models Authors: Xiangming Gu, Soham De, Larisa Markeeva, Petar Veličković, Razvan Pascanu Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05868v1 Large Reasoning Models (LRMs) have shown remarkable performance on challenging questions, such as math and coding. However, to obtain a high quality solution, one may need to sample more than once. In principal, there are two sampling strategies that can be composed to form more complex processes: sequential sampling and parallel sampling. In this paper, we first compare these two approaches with rigor, and observe, aligned with previous works, that parallel sampling seems to outperform sequential sampling even though the latter should have more representation power. To understand the underline reasons, we make three hypothesis on the reason behind this behavior: (i) parallel sampling outperforms due to the aggregator operator; (ii) sequential sampling is harmed by needing to use longer contexts; (iii) sequential sampling leads to less exploration due to conditioning on previous answers. The empirical evidence on various model families and sizes (Qwen3, DeepSeek-R1 distilled models, Gemini 2.5) and question domains (math and coding) suggests that the aggregation and context length do not seem to be the main culprit behind the performance gap. In contrast, the lack of exploration seems to play a considerably larger role, and we argue that this is one main cause for the performance gap. --- ### Beyond Paper-to-Paper: Structured Profiling and Rubric Scoring for Paper-Reviewer Matching Authors: Yicheng Pan, Zhiyuan Ning, Ludi Wang, Yi Du Published: 2026-04-07 Categories: cs.IR, cs.CL, cs.DL arXiv: https://arxiv.org/abs/2604.05866v1 As conference submission volumes continue to grow, accurately recommending suitable reviewers has become a challenge. Most existing methods follow a ``Paper-to-Paper'' matching paradigm, implicitly representing a reviewer by their publication history. However, effective reviewer matching requires capturing multi-dimensional expertise, and textual similarity to past papers alone is often insufficient. To address this gap, we propose P2R, a training-free framework that shifts from implicit paper-to-paper matching to explicit profile-based matching. P2R uses general-purpose LLMs to construct structured profiles for both submissions and reviewers, disentangling them into Topics, Methodologies, and Applications. Building on these profiles, P2R adopts a coarse-to-fine pipeline to balance efficiency and depth. It first performs hybrid retrieval that combines semantic and aspect-level signals to form a high-recall candidate pool, and then applies an LLM-based committee to evaluate candidates under strict rubrics, integrating both multi-dimensional expert views and a holistic Area Chair perspective. Experiments on NeurIPS, SIGIR, and SciRepEval show that P2R consistently outperforms state-of-the-art baselines. Ablation studies further verify the necessity of each component. Overall, P2R highlights the value of explicit, structured expertise modeling and offers practical guidance for applying LLMs to reviewer matching. --- ### JTON: A Token-Efficient JSON Superset with Zen Grid Tabular Encoding for Large Language Models Authors: Gowthamkumar Nandakishore Published: 2026-04-07 Categories: cs.AI, cs.PL arXiv: https://arxiv.org/abs/2604.05865v1 When LLMs process structured data, the serialization format directly affects cost and context utilization. Standard JSON wastes tokens repeating key names in every row of a tabular array--overhead that scales linearly with row count. This paper presents JTON (JSON Tabular Object Notation), a strict JSON superset whose main idea, Zen Grid, factors column headers into a single row and encodes values with semicolons, preserving JSON's type system while cutting redundancy. Across seven real-world domains, Zen Grid reduces token counts by 15-60% versus JSON compact (28.5% average; 32% with bare_strings). Comprehension tests on 10 LLMs show a net +0.3 pp accuracy gain over JSON: four models improve, three hold steady, and three dip slightly. Generation tests on 12 LLMs yield 100% syntactic validity in both few-shot and zero-shot settings. A Rust/PyO3 reference implementation adds SIMD-accelerated parsing at 1.4x the speed of Python's json module. Code, a 683-vector test suite, and all experimental data are publicly available. --- ### LoRM: Learning the Language of Rotating Machinery for Self-Supervised Condition Monitoring Authors: Xiao Qin, Xingyi Song, Tong Liu, Hatim Laalej, Zepeng Liu, Yunpeng Zhu, Ligang He Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05863v1 We present LoRM (Language of Rotating Machinery), a self-supervised framework for multi-modal rotating-machinery signal understanding and real-time condition monitoring. LoRM is built on the idea that rotating-machinery signals can be viewed as a machine language: local signals can be tokenised into discrete symbolic units, and their future evolution can be predicted from observed multi-sensor context. Unlike conventional signal-processing methods that rely on hand-crafted transforms and features, LoRM reformulates multi-modal sensor data as a token-based sequence-prediction problem. For each data window, the observed context segment is retained in continuous form, while the future target segment of each sensing channel is quantised into a discrete token. Then, efficient knowledge transfer is achieved by partially fine-tuning a general-purpose pre-trained language model on industrial signals, avoiding the need to train a large model from scratch. Finally, condition monitoring is performed by tracking token-prediction errors as a health indicator, where increasing errors indicate degradation. In-situ tool condition monitoring (TCM) experiments demonstrate stable real-time tracking and strong cross-tool generalisation, showing that LoRM provides a practical bridge between language modelling and industrial signal analysis. The source code is publicly available at https://github.com/Q159753258/LormPHM. --- ### When Do We Need LLMs? A Diagnostic for Language-Driven Bandits Authors: Uljad Berdica, Fernando Acero, Anton Ipsen, Parisa Zehtabi, Michael Cashmore, Manuela Veloso Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05859v2 We study Contextual Multi-Armed Bandits (CMABs) for non-episodic decision-making problems where the context includes both textual and numerical information (e.g., recommendation systems, dynamic portfolio adjustments, offer selection; all frequent problems in finance). While Large Language Models (LLMs) are increasingly applied to these settings, utilizing LLMs for reasoning at every decision step is computationally expensive, and uncertainty estimates are difficult to obtain. To address this, we introduce LLMP-UCB, a bandit algorithm that derives uncertainty estimates from LLMs via repeated inference. However, our experiments demonstrate that lightweight numerical bandits operating on text embeddings (dense or Matryoshka) match or exceed the accuracy of LLM-based solutions at a fraction of their cost. We further show that embedding dimensionality is a practical lever on the exploration-exploitation balance, enabling cost-performance tradeoffs without prompt complexity. Finally, to guide practitioners, we propose a geometric diagnostic based on the arms' embeddings to decide when to use LLM-driven reasoning versus a lightweight numerical bandit. Our results provide a principled deployment framework for cost-effective, uncertainty-aware decision systems with broad applicability across AI use cases. --- ### Weight-Informed Self-Explaining Clustering for Mixed-Type Tabular Data Authors: Lehao Li, Qiang Huang, Yihao Ang, Bryan Kian Hsiang Low, Anthony K. H. Tung, Xiaokui Xiao Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05857v1 Clustering mixed-type tabular data is fundamental for exploratory analysis, yet remains challenging due to misaligned numerical-categorical representations, uneven and context-dependent feature relevance, and disconnected and post-hoc explanation from the clustering process. We propose WISE, a Weight-Informed Self-Explaining framework that unifies representation, feature weighting, clustering, and interpretation in a fully unsupervised and transparent pipeline. WISE introduces Binary Encoding with Padding (BEP) to align heterogeneous features in a unified sparse space, a Leave-One-Feature-Out (LOFO) strategy to sense multiple high-quality and diverse feature-weighting views, and a two-stage weight-aware clustering procedure to aggregate alternative semantic partitions. To ensure intrinsic interpretability, we further develop Discriminative FreqItems (DFI), which yields feature-level explanations that are consistent from instances to clusters with an additive decomposition guarantee. Extensive experiments on six real-world datasets demonstrate that WISE consistently outperforms classical and neural baselines in clustering quality while remaining efficient, and produces faithful, human-interpretable explanations grounded in the same primitives that drive clustering. --- ### Neural Network Pruning via QUBO Optimization Authors: Osama Orabi, Artur Zagitov, Hadi Salloum, Viktor A. Lobachev, Kasymkhan Khubiev, Yaroslav Kholodov Published: 2026-04-07 Categories: cs.CV, cs.AI, cs.LG, cs.NE arXiv: https://arxiv.org/abs/2604.05856v1 Neural network pruning can be formulated as a combinatorial optimization problem, yet most existing approaches rely on greedy heuristics that ignore complex interactions between filters. Formal optimization methods such as Quadratic Unconstrained Binary Optimization (QUBO) provide a principled alternative but have so far underperformed due to oversimplified objective formulations based on metrics like the L1-norm. In this work, we propose a unified Hybrid QUBO framework that bridges heuristic importance estimation with global combinatorial optimization. Our formulation integrates gradient-aware sensitivity metrics - specifically first-order Taylor and second-order Fisher information - into the linear term, while utilizing data-driven activation similarity in the quadratic term. This allows the QUBO objective to jointly capture individual filter relevance and inter-filter functional redundancy. We further introduce a dynamic capacity-driven search to strictly enforce target sparsity without distorting the optimization landscape. Finally, we employ a two-stage pipeline featuring a Tensor-Train (TT) Refinement stage - a gradient-free optimizer that fine-tunes the QUBO-derived solution directly against the true evaluation metric. Experiments on the SIDD image denoising dataset demonstrate that the proposed Hybrid QUBO significantly outperforms both greedy Taylor pruning and traditional L1-based QUBO, with TT Refinement providing further consistent gains at appropriate combinatorial scales. This highlights the potential of hybrid combinatorial formulations for robust, scalable, and interpretable neural network compression. --- ### Deep Researcher Agent: An Autonomous Framework for 24/7 Deep Learning Experimentation with Zero-Cost Monitoring Authors: Xiangyue Zhang Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05854v1 We present \textbf{Deep Researcher Agent}, an open-source framework that enables large language model (LLM) agents to autonomously conduct deep learning experiments around the clock. Unlike existing AI research assistants that focus on paper writing or code generation, our system addresses the full experiment lifecycle: hypothesis formation, code implementation, training execution, result analysis, and iterative refinement. The framework introduces three key innovations: (1) \textbf{Zero-Cost Monitoring} -- a monitoring paradigm that incurs zero LLM API costs during model training by relying solely on process-level checks and log file reads; (2) \textbf{Two-Tier Constant-Size Memory} -- a memory architecture capped at $\sim$5K characters regardless of runtime duration, preventing the unbounded context growth that plagues long-running agents; and (3) \textbf{Minimal-Toolset Leader-Worker Architecture} -- a multi-agent design where each worker agent is equipped with only 3--5 tools, reducing per-call token overhead by up to 73\%. In sustained deployments spanning 30+ days, the framework autonomously completed 500+ experiment cycles across four concurrent research projects, achieving a 52\% improvement over baseline metrics in one project through 200+ automated experiments -- all at an average LLM cost of \$0.08 per 24-hour cycle. Code is available at https://github.com/Xiangyue-Zhang/auto-deep-researcher-24x7. --- ### Reading Between the Pixels: An Inscriptive Jailbreak Attack on Text-to-Image Models Authors: Zonghao Ying, Haowen Dai, Lianyu Hu, Zonglei Jing, Quanchen Zou, Yaodong Yang, Aishan Liu, Xianglong Liu Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05853v3 Modern text-to-image (T2I) models can now render legible, paragraph-length text, enabling a fundamentally new class of misuse. We identify and formalize the inscriptive jailbreak, where an adversary coerces a T2I system into generating images containing harmful textual payloads (e.g., fraudulent documents) embedded within visually benign scenes. Unlike traditional depictive jailbreaks that elicit visually objectionable imagery, inscriptive attacks weaponize the text-rendering capability itself. Because existing jailbreak techniques are designed for coarse visual manipulation, they struggle to bypass multi-stage safety filters while maintaining character-level fidelity. To expose this vulnerability, we propose Etch, a black-box attack framework that decomposes the adversarial prompt into three functionally orthogonal layers: semantic camouflage, visual-spatial anchoring, and typographic encoding. This decomposition reduces joint optimization over the full prompt space to tractable sub-problems, which are iteratively refined through a zero-order loop. In this process, a vision-language model critiques each generated image, localizes failures to specific layers, and prescribes targeted revisions. Extensive evaluations across 7 models on the 2 benchmarks demonstrate that Etch achieves an average attack success rate of 65.57% (peaking at 91.00%), significantly outperforming existing baselines. Our results reveal a critical blind spot in current T2I safety alignments and underscore the urgent need for typography-aware defense multimodal mechanisms. --- ### Reading Between the Pixels: An Inscriptive Jailbreak Attack on Text-to-Image Models Authors: Zonghao Ying, Haowen Dai, Lianyu Hu, Zonglei Jing, Quanchen Zou, Yaodong Yang, Aishan Liu, Xianglong Liu Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05853v2 Modern text-to-image (T2I) models can now render legible, paragraph-length text, enabling a fundamentally new class of misuse. We identify and formalize the inscriptive jailbreak, where an adversary coerces a T2I system into generating images containing harmful textual payloads (e.g., fraudulent documents) embedded within visually benign scenes. Unlike traditional depictive jailbreaks that elicit visually objectionable imagery, inscriptive attacks weaponize the text-rendering capability itself. Because existing jailbreak techniques are designed for coarse visual manipulation, they struggle to bypass multi-stage safety filters while maintaining character-level fidelity. To expose this vulnerability, we propose Etch, a black-box attack framework that decomposes the adversarial prompt into three functionally orthogonal layers: semantic camouflage, visual-spatial anchoring, and typographic encoding. This decomposition reduces joint optimization over the full prompt space to tractable sub-problems, which are iteratively refined through a zero-order loop. In this process, a vision-language model critiques each generated image, localizes failures to specific layers, and prescribes targeted revisions. Extensive evaluations across 7 models on the 2 benchmarks demonstrate that Etch achieves an average attack success rate of 65.57% (peaking at 91.00%), significantly outperforming existing baselines. Our results reveal a critical blind spot in current T2I safety alignments and underscore the urgent need for typography-aware defense multimodal mechanisms. --- ### Evaluating Learner Representations for Differentiation Prior to Instructional Outcomes Authors: Junsoo Park, Youssef Medhat, Htet Phyo Wai, Ploy Thajchayapong, Ashok K. Goel Published: 2026-04-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05848v1 Learner representations play a central role in educational AI systems, yet it is often unclear whether they preserve meaningful differences between students when instructional outcomes are unavailable or highly context-dependent. This work examines how to evaluate learner representations based on whether they retain separation between learners under a shared comparison rule. We introduce distinctiveness, a representation-level measure that evaluates how each learner differs from others in the cohort using pairwise distances, without requiring clustering, labels, or task-specific evaluation. Using student-authored questions collected through a conversational AI agent in an online learning environment, we compare representations based on individual questions with representations that aggregate patterns across a student's interactions over time. Results show that learner-level representations yield higher separation, stronger clustering structure, and more reliable pairwise discrimination than interaction-level representations. These findings demonstrate that learner representations can be evaluated independently of instructional outcomes and provide a practical pre-deployment criterion using distinctiveness as a diagnostic metric for assessing whether a representation supports differentiated modeling or personalization. --- ### AgentGL: Towards Agentic Graph Learning with LLMs via Reinforcement Learning Authors: Yuanfu Sun, Kang Li, Dongzhe Fan, Jiajin Liu, Qiaoyu Tan Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05846v2 Large Language Models (LLMs) increasingly rely on agentic capabilities-iterative retrieval, tool use, and decision-making-to overcome the limits of static, parametric knowledge. Yet existing agentic frameworks treat external information as unstructured text and fail to leverage the topological dependencies inherent in real-world data. To bridge this gap, we introduce Agentic Graph Learning (AGL), a paradigm that reframes graph learning as an interleaved process of topology-aware navigation and LLM-based inference. Specifically, we propose AgentGL, the first reinforcement learning (RL)-driven framework for AGL. AgentGL equips an LLM agent with graph-native tools for multi-scale exploration, regulates tool usage via search-constrained thinking to balance accuracy and efficiency, and employs a graph-conditioned curriculum RL strategy to stabilize long-horizon policy learning without step-wise supervision. Across diverse Text-Attributed Graph (TAG) benchmarks and multiple LLM backbones, AgentGL substantially outperforms strong GraphLLMs and GraphRAG baselines, achieving absolute improvements of up to 17.5% in node classification and 28.4% in link prediction. These results demonstrate that AGL is a promising frontier for enabling LLMs to autonomously navigate and reason over complex relational environments. The code is publicly available at https://github.com/sunyuanfu/AgentGL. --- ### JD-BP: A Joint-Decision Generative Framework for Auto-Bidding and Pricing Authors: Linghui Meng, Chun Gan, Shengsheng Niu, Chengcheng Zhang, Chenchen Li, Chuan Yang, Yi Mao, Xin Zhu, Jie He, Zhangang Lin, Ching Law Published: 2026-04-07 Categories: cs.GT, cs.LG arXiv: https://arxiv.org/abs/2604.05845v2 Auto-bidding services optimize real-time bidding strategies for advertisers under key performance indicator (KPI) constraints such as target return on investment and budget. However, uncertainties such as model prediction errors and feedback latency can cause bidding strategies to deviate from ex-post optimality, leading to inefficient allocation. To address this issue, we propose JD-BP, a Joint generative Decision framework for Bidding and Pricing. Unlike prior methods, JD-BP jointly outputs a bid value and a pricing correction term that acts additively with the payment rule such as GSP. To mitigate adverse effects of historical constraint violations, we design a memory-less Return-to-Go that encourages future value maximizing of bidding actions while the cumulated bias is handled by the pricing correction. Moreover, a trajectory augmentation algorithm is proposed to generate joint bidding-pricing trajectories from a (possibly arbitrary) base bidding policy, enabling efficient plug-and-play deployment of our algorithm from existing RL/generative bidding models. Finally, we employ an Energy-Based Direct Preference Optimization method in conjunction with a cross-attention module to enhance the joint learning performance of bidding and pricing correction. Offline experiments on the AuctionNet dataset demonstrate that JD-BP achieves state-of-the-art performance. Online A/B tests at JD.com confirm its practical effectiveness, showing a 4.70% increase in ad revenue and a 6.48% improvement in target cost. --- ### Modeling Patient Care Trajectories with Transformer Hawkes Processes Authors: Saumya Pandey, Varun Chandola Published: 2026-04-07 Categories: cs.LG, q-bio.QM arXiv: https://arxiv.org/abs/2604.05844v1 Patient healthcare utilization consists of irregularly time-stamped events, such as outpatient visits, inpatient admissions, and emergency encounters, forming individualized care trajectories. Modeling these trajectories is crucial for understanding utilization patterns and predicting future care needs, but is challenging due to temporal irregularity and severe class imbalance. In this work, we build on the Transformer Hawkes Process framework to model patient trajectories in continuous time. By combining Transformer-based history encoding with Hawkes process dynamics, the model captures event dependencies and jointly predicts event type and time-to-event. To address extreme imbalance, we introduce an imbalance-aware training strategy using inverse square-root class weighting. This improves sensitivity to rare but clinically important events without altering the data distribution. Experiments on real-world data demonstrate improved performance and provide clinically meaningful insights for identifying high-risk patient populations. --- ### EEG-MFTNet: An Enhanced EEGNet Architecture with Multi-Scale Temporal Convolutions and Transformer Fusion for Cross-Session Motor Imagery Decoding Authors: Panagiotis Andrikopoulos, Siamak Mehrkanoon Published: 2026-04-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.05843v1 Brain-computer interfaces (BCIs) enable direct communication between the brain and external devices, providing critical support for individuals with motor impairments. However, accurate motor imagery (MI) decoding from electroencephalography (EEG) remains challenging due to noise and cross-session variability. This study introduces EEG-MFTNet, a novel deep learning model based on the EEGNet architecture, enhanced with multi-scale temporal convolutions and a Transformer encoder stream. These components are designed to capture both short and long-range temporal dependencies in EEG signals. The model is evaluated on the SHU dataset using a subject-dependent cross-session setup, outperforming baseline models, including EEGNet and its recent derivatives. EEG-MFTNet achieves an average classification accuracy of 58.9% while maintaining low computational complexity and inference latency. The results highlight the model's potential for real-time BCI applications and underscore the importance of architectural innovations in improving MI decoding. This work contributes to the development of more robust and adaptive BCI systems, with implications for assistive technologies and neurorehabilitation. --- ### Expectation Maximization (EM) Converges for General Agnostic Mixtures Authors: Avishek Ghosh Published: 2026-04-07 Categories: cs.LG, cs.IT, stat.ML arXiv: https://arxiv.org/abs/2604.05842v1 Mixture of linear regression is well studied in statistics and machine learning, where the data points are generated probabilistically using $k$ linear models. Algorithms like Expectation Maximization (EM) may be used to recover the ground truth regressors for this problem. Recently, in \cite{pal2022learning,ghosh_agnostic} the mixed linear regression problem is studied in the agnostic setting, where no generative model on data is assumed. Rather, given a set of data points, the objective is \emph{fit} $k$ lines by minimizing a suitable loss function. It is shown that a modification of EM, namely gradient EM converges exponentially to appropriately defined loss minimizer even in the agnostic setting. In this paper, we study the problem of \emph{fitting} $k$ parametric functions to given set of data points. We adhere to the agnostic setup. However, instead of fitting lines equipped with quadratic loss, we consider any arbitrary parametric function fitting equipped with a strongly convex and smooth loss. This framework encompasses a large class of problems including mixed linear regression (regularized), mixed linear classifiers (mixed logistic regression, mixed Support Vector Machines) and mixed generalized linear regression. We propose and analyze gradient EM for this problem and show that with proper initialization and separation condition, the iterates of gradient EM converge exponentially to appropriately defined population loss minimizers with high probability. This shows the effectiveness of EM type algorithm which converges to \emph{optimal} solution in the non-generative setup beyond mixture of linear regression. --- ### Vision-Guided Iterative Refinement for Frontend Code Generation Authors: Hannah Sansford, Derek H. C. Law, Wei Liu, Abhishek Tripathi, Niresh Agarwal, Gerrit J. J. van den Burg Published: 2026-04-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2604.05839v1 Code generation with large language models often relies on multi-stage human-in-the-loop refinement, which is effective but very costly - particularly in domains such as frontend web development where the solution quality depends on rendered visual output. We present a fully automated critic-in-the-loop framework in which a vision-language model serves as a visual critic that provides structured feedback on rendered webpages to guide iterative refinement of generated code. Across real-world user requests from the WebDev Arena dataset, this approach yields consistent improvements in solution quality, achieving up to 17.8% increase in performance over three refinement cycles. Next, we investigate parameter-efficient fine-tuning using LoRA to understand whether the improvements provided by the critic can be internalized by the code-generating LLM. Fine-tuning achieves 25% of the gains from the best critic-in-the-loop solution without a significant increase in token counts. Our findings indicate that automated, VLM-based critique of frontend code generation leads to significantly higher quality solutions than can be achieved through a single LLM inference pass, and highlight the importance of iterative refinement for the complex visual outputs associated with web development. --- ### Hidden in the Multiplicative Interaction: Uncovering Fragility in Multimodal Contrastive Learning Authors: Tillmann Rheude, Stefan Hegselmann, Roland Eils, Benjamin Wild Published: 2026-04-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.05834v2 Contrastive learning has become a standard approach for unsupervised learning from paired data, as demonstrated by CLIP for image-text matching. However, many domains involve more than two modalities and require objectives that capture higher-order dependencies beyond pairwise alignment. Symile extends CLIP to this setting by replacing the dot product with the multilinear inner product (MIP) over modality embeddings. In this work, we show that there is a fragility which ishidden in the multiplicative interaction: a single weakly informative, misaligned, or missing modality can propagate through the objective and distort cross-modal retrieval scores. We propose Gated Symile, a contrastive gating mechanism that adapts modality contributions on an attention-based, per-candidate basis. The gate suppresses unreliable inputs by interpolating embeddings toward learnable neutral directions with an explicit NULL option when reliable cross-modal alignment is unlikely. Across a controlled synthetic benchmark that uncovers this fragility and three real-world trimodal datasets, Gated Symile achieves higher top-1 retrieval accuracy than well-tuned state-of-the-art (sota) baselines. More broadly, our results highlight gating as a step toward robust multimodal contrastive learning beyond two modalities in the presence of noise, misalignment, or missing inputs. --- ### "OK Aura, Be Fair With Me": Demographics-Agnostic Training for Bias Mitigation in Wake-up Word Detection Authors: Fernando López, Paula Delgado-Santos, Pablo Gómez, David Solans, Jordi Luque Published: 2026-04-07 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.05830v1 Voice-based interfaces are widely used; however, achieving fair Wake-up Word detection across diverse speaker populations remains a critical challenge due to persistent demographic biases. This study evaluates the effectiveness of demographics-agnostic training techniques in mitigating performance disparities among speakers of varying sex, age, and accent. We utilize the OK Aura database for our experiments, employing a training methodology that excludes demographic labels, which are reserved for evaluation purposes. We explore (i) data augmentation techniques to enhance model generalization and (ii) knowledge distillation of pre-trained foundational speech models. The experimental results indicate that these demographics-agnostic training techniques markedly reduce demographic bias, leading to a more equitable performance profile across different speaker groups. Specifically, one of the evaluated techniques achieves a Predictive Disparity reduction of 39.94\% for sex, 83.65\% for age, and 40.48\% for accent when compared to the baseline. This study highlights the effectiveness of label-agnostic methodologies in fostering fairness in Wake-up Word detection. --- ### Bivariate Causal Discovery Using Rate-Distortion MDL: An Information Dimension Approach Authors: Tiago Brogueira, Mário A. T. Figueiredo Published: 2026-04-07 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2604.05829v1 Approaches to bivariate causal discovery based on the minimum description length (MDL) principle approximate the (uncomputable) Kolmogorov complexity of the models in each causal direction, selecting the one with the lower total complexity. The premise is that nature's mechanisms are simpler in their true causal order. Inherently, the description length (complexity) in each direction includes the description of the cause variable and that of the causal mechanism. In this work, we argue that current state-of-the-art MDL-based methods do not correctly address the problem of estimating the description length of the cause variable, effectively leaving the decision to the description length of the causal mechanism. Based on rate-distortion theory, we propose a new way to measure the description length of the cause, corresponding to the minimum rate required to achieve a distortion level representative of the underlying distribution. This distortion level is deduced using rules from histogram-based density estimation, while the rate is computed using the related concept of information dimension, based on an asymptotic approximation. Combining it with a traditional approach for the causal mechanism, we introduce a new bivariate causal discovery method, termed rate-distortion MDL (RDMDL). We show experimentally that RDMDL achieves competitive performance on the Tübingen dataset. All the code and experiments are publicly available at github.com/tiagobrogueira/Causal-Discovery-In-Exchangeable-Data. --- ### Reciprocal Trust and Distrust in Artificial Intelligence Systems: The Hard Problem of Regulation Authors: Martino Maggetti Published: 2026-04-07 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2604.05826v1 Policy makers, scientists, and the public are increasingly confronted with thorny questions about the regulation of artificial intelligence (AI) systems. A key common thread concerns whether AI can be trusted and the factors that can make it more trustworthy in front of stakeholders and users. This is indeed crucial, as the trustworthiness of AI systems is fundamental for both democratic governance and for the development and deployment of AI. This article advances the discussion by arguing that AI systems should also be recognized, as least to some extent, as artifacts capable of exercising a form of agency, thereby enabling them to engage in relationships of trust or distrust with humans. It further examines the implications of these reciprocal trust dynamics for regulators tasked with overseeing AI systems. The article concludes by identifying key tensions and unresolved dilemmas that these dynamics pose for the future of AI regulation and governance. --- ### CLEAR: Cross-Lingual Enhancement in Alignment via Reverse-training Authors: Seungyoon Lee, Minhyuk Kim, Seongtae Hong, Youngjoon Jang, Dongsuk Oh, Heuiseok Lim Published: 2026-04-07 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2604.05821v2 Existing multilingual embedding models often encounter challenges in cross-lingual scenarios due to imbalanced linguistic resources and less consideration of cross-lingual alignment during training. Although standardized contrastive learning approaches for cross-lingual adaptation are widely adopted, they may struggle to capture fundamental alignment between languages and degrade performance in well-aligned languages such as English. To address these challenges, we propose Cross-Lingual Enhancement in Retrieval via Reverse-training (CLEAR), a novel loss function utilizing a reverse training scheme to improve retrieval performance across diverse cross-lingual retrieval scenarios. CLEAR leverages an English passage as a bridge to strengthen alignments between the target language and English, ensuring robust performance in the cross-lingual retrieval task. Our extensive experiments demonstrate that CLEAR achieves notable improvements in cross-lingual scenarios, with gains up to 15%, particularly in low-resource languages, while minimizing performance degradation in English. Furthermore, our findings highlight that CLEAR offers promising effectiveness even in multilingual training, suggesting its potential for broad application and scalability. We release the code at https://github.com/dltmddbs100/CLEAR. --- ### Learn to Rank: Visual Attribution by Learning Importance Ranking Authors: David Schinagl, Christian Fruhwirth-Reisinger, Alexander Prutsch, Samuel Schulter, Horst Possegger Published: 2026-04-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.05819v2 Interpreting the decisions of complex computer vision models is crucial to establish trust and accountability, especially in safety-critical domains. An established approach to interpretability is generating visual attribution maps that highlight regions of the input most relevant to the model's prediction. However, existing methods face a three-way trade-off. Propagation-based approaches are efficient, but they can be biased and architecture-specific. Meanwhile, perturbation-based methods are causally grounded, yet they are expensive and for vision transformers often yield coarse, patch-level explanations. Learning-based explainers are fast but usually optimize surrogate objectives or distill from heuristic teachers. We propose a learning scheme that instead optimizes deletion and insertion metrics directly. Since these metrics depend on non-differentiable sorting and ranking, we frame them as permutation learning and replace the hard sorting with a differentiable relaxation using Gumbel-Sinkhorn. This enables end-to-end training through attribution-guided perturbations of the target model. During inference, our method produces dense, pixel-level attributions in a single forward pass with optional, few-step gradient refinement. Our experiments demonstrate consistent quantitative improvements and sharper, boundary-aligned explanations, particularly for transformer-based vision models. Code and pretrained models are available at https://github.com/dschinagl/AHA. --- ### Learn to Rank: Visual Attribution by Learning Importance Ranking Authors: David Schinagl, Christian Fruhwirth-Reisinger, Alexander Prutsch, Samuel Schulter, Horst Possegger Published: 2026-04-07 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2604.05819v1 Interpreting the decisions of complex computer vision models is crucial to establish trust and accountability, especially in safety-critical domains. An established approach to interpretability is generating visual attribution maps that highlight regions of the input most relevant to the model's prediction. However, existing methods face a three-way trade-off. Propagation-based approaches are efficient, but they can be biased and architecture-specific. Meanwhile, perturbation-based methods are causally grounded, yet they are expensive and for vision transformers often yield coarse, patch-level explanations. Learning-based explainers are fast but usually optimize surrogate objectives or distill from heuristic teachers. We propose a learning scheme that instead optimizes deletion and insertion metrics directly. Since these metrics depend on non-differentiable sorting and ranking, we frame them as permutation learning and replace the hard sorting with a differentiable relaxation using Gumbel-Sinkhorn. This enables end-to-end training through attribution-guided perturbations of the target model. During inference, our method produces dense, pixel-level attributions in a single forward pass with optional, few-step gradient refinement. Our experiments demonstrate consistent quantitative improvements and sharper, boundary-aligned explanations, particularly for transformer-based vision models. --- ### WikiSeeker: Rethinking the Role of Vision-Language Models in Knowledge-Based Visual Question Answering Authors: Yingjian Zhu, Xinming Wang, Kun Ding, Ying Wang, Bin Fan, Shiming Xiang Published: 2026-04-07 Categories: cs.CV, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2604.05818v2 Multi-modal Retrieval-Augmented Generation (RAG) has emerged as a highly effective paradigm for Knowledge-Based Visual Question Answering (KB-VQA). Despite recent advancements, prevailing methods still primarily depend on images as the retrieval key, and often overlook or misplace the role of Vision-Language Models (VLMs), thereby failing to leverage their potential fully. In this paper, we introduce WikiSeeker, a novel multi-modal RAG framework that bridges these gaps by proposing a multi-modal retriever and redefining the role of VLMs. Rather than serving merely as answer generators, we assign VLMs two specialized agents: a Refiner and an Inspector. The Refiner utilizes the capability of VLMs to rewrite the textual query according to the input image, significantly improving the performance of the multimodal retriever. The Inspector facilitates a decoupled generation strategy by selectively routing reliable retrieved context to another LLM for answer generation, while relying on the VLM's internal knowledge when retrieval is unreliable. Extensive experiments on EVQA, InfoSeek, and M2KR demonstrate that WikiSeeker achieves state-of-the-art performance, with substantial improvements in both retrieval accuracy and answer quality. Our code will be released on https://github.com/zhuyjan/WikiSeeker. --- ### Adjustable Text-Guided Backdoor Attacks with Natural-Word Triggers on Multimodal Pretrained Models Authors: Yiyang Zhang, Chaojian Yu, Ziming Hong, Yuanjie Shao, Qinmu Peng, Tongliang Liu, Xinge You Published: 2026-04-07 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2604.05809v2 This paper presents Text-Guided Backdoor (TGB), an adjustable backdoor attack against multimodal pretrained models that uses natural-word triggers, namely words that can naturally occur in ordinary textual inputs. Most existing backdoor attacks require specific trigger conditions that are typically not satisfied by ordinary inference inputs, thereby limiting their activation in real-world deployments. TGB overcomes this limitation by exploiting naturally occurring words as triggers, enabling stealthy activation without requiring explicit trigger insertion at inference time. This property avoids conspicuous trigger patterns and improves the practicality of TGB. Furthermore, we introduce visual adversarial perturbations on poisoned samples to modulate the model's learning of natural-word triggers, thereby enabling flexible adjustment of TGB attack strength without modifying the poisoned data. Extensive experiments are conducted on downstream tasks built upon multimodal pretrained models, including Composed Image Retrieval (CIR) and Visual Question Answering (VQA). Results demonstrate the effectiveness of TGB across diverse poisoning settings and its ability to flexibly adjust attack success rates, which reveal critical security vulnerabilities in multimodal pretrained models. --- ### Stealthy and Adjustable Text-Guided Backdoor Attacks on Multimodal Pretrained Models Authors: Yiyang Zhang, Chaojian Yu, Ziming Hong, Yuanjie Shao, Qinmu Peng, Tongliang Liu, Xinge You Published: 2026-04-07 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2604.05809v1 Multimodal pretrained models are vulnerable to backdoor attacks, yet most existing methods rely on visual or multimodal triggers, which are impractical since visually embedded triggers rarely occur in real-world data. To overcome this limitation, we propose a novel Text-Guided Backdoor (TGB) attack on multimodal pretrained models, where commonly occurring words in textual descriptions serve as backdoor triggers, significantly improving stealthiness and practicality. Furthermore, we introduce visual adversarial perturbations on poisoned samples to modulate the model's learning of textual triggers, enabling a controllable and adjustable TGB attack. Extensive experiments on downstream tasks built upon multimodal pretrained models, including Composed Image Retrieval (CIR) and Visual Question Answering (VQA), demonstrate that TGB achieves practicality and stealthiness with adjustable attack success rates across diverse realistic settings, revealing critical security vulnerabilities in multimodal pretrained models. --- ### Hierarchical Reinforcement Learning with Augmented Step-Level Transitions for LLM Agents Authors: Shuai Zhen, Yanhua Yu, Ruopei Guo, Nan Cheng, Yang Deng Published: 2026-04-07 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.05808v2 Large language model (LLM) agents have demonstrated strong capabilities in complex interactive decision-making tasks. However, existing LLM agents typically rely on increasingly long interaction histories, resulting in high computational cost and limited scalability. In this paper, we propose STEP-HRL, a hierarchical reinforcement learning (HRL) framework that enables step-level learning by conditioning only on single-step transitions rather than full interaction histories. STEP-HRL structures tasks hierarchically, using completed subtasks to represent global progress of overall task. By introducing a local progress module, it also iteratively and selectively summarizes interaction history within each subtask to produce a compact summary of local progress. Together, these components yield augmented step-level transitions for both high-level and low-level policies. Experimental results on ScienceWorld and ALFWorld benchmarks consistently demonstrate that STEP-HRL substantially outperforms baselines in terms of performance and generalization while reducing token usage. Our code is available at https://github.com/TonyStark042/STEP-HRL. --- ### Constraint-Driven Warm-Freeze for Efficient Transfer Learning in Photovoltaic Systems Authors: Yasmeen Saeed, Ahmed Sharshar, Mohsen Guizani Published: 2026-04-07 Categories: cs.NE arXiv: https://arxiv.org/abs/2604.05807v1 Detecting cyberattacks in photovoltaic (PV) monitoring and MPPT control signals requires models that are robust to bias, drift, and transient spikes, yet lightweight enough for resource-constrained edge controllers. While deep learning outperforms traditional physics-based diagnostics and handcrafted features, standard fine-tuning is computationally prohibitive for edge devices. Furthermore, existing Parameter-Efficient Fine-Tuning (PEFT) methods typically apply uniform adaptation or rely on expensive architectural searches, lacking the flexibility to adhere to strict hardware budgets. To bridge this gap, we propose Constraint-Driven Warm-Freeze (CDWF), a budget-aware adaptation framework. CDWF leverages a brief warm-start phase to quantify gradient-based block importance, then solves a constrained optimization problem to dynamically allocate full trainability to high-impact blocks while efficiently adapting the remaining blocks via Low-Rank Adaptation (LoRA). We evaluate CDWF on standard vision benchmarks (CIFAR-10/100) and a novel PV cyberattack dataset, transferring from bias pretraining to drift and spike detection. The experiments demonstrate that CDWF retains 90 to 99% of full fine-tuning performance while reducing trainable parameters by up to 120x. These results establish CDWF as an effective, importance-guided solution for reliable transfer learning under tight edge constraints. --- ### Measuring What Matters!! Assessing Therapeutic Principles in Mental-Health Conversation Authors: Abdullah Mazhar, Het Riteshkumar Shah, Aseem Srivastava, Smriti Joshi, Md Shad Akhtar Published: 2026-04-07 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.05795v3 The increasing use of large language models in mental health applications calls for principled evaluation frameworks that assess alignment with psychotherapeutic best practices beyond surface-level fluency. While recent systems exhibit conversational competence, they lack structured mechanisms to evaluate adherence to core therapeutic principles. In this paper, we study the problem of evaluating AI-generated therapist-like responses for clinically grounded appropriateness and effectiveness. We assess each therapists utterance along six therapeutic principles: non-judgmental acceptance, warmth, respect for autonomy, active listening, reflective understanding, and situational appropriateness using a fine-grained ordinal scale. We introduce FAITH-M, a benchmark annotated with expert-assigned ordinal ratings, and propose CARE, a multi-stage evaluation framework that integrates intra-dialogue context, contrastive exemplar retrieval, and knowledge-distilled chain-of-thought reasoning. Experiments show that CARE achieves an F-1 score of 63.34 versus the strong baseline Qwen3 F-1 score of 38.56 which is a 64.26 improvement, which also serves as its backbone, indicating that gains arise from structured reasoning and contextual modeling rather than backbone capacity alone. Expert assessment and external dataset evaluations further demonstrate robustness under domain shift, while highlighting challenges in modelling implicit clinical nuance. Overall, CARE provides a clinically grounded framework for evaluating therapeutic fidelity in AI mental health systems. --- ### EfficientMonoHair: Fast Strand-Level Reconstruction from Monocular Video via Multi-View Direction Fusion Authors: Da Li, Dominik Engel, Deng Luo, Ivan Viola Published: 2026-04-07 Categories: cs.CV, cs.GR arXiv: https://arxiv.org/abs/2604.05794v1 Strand-level hair geometry reconstruction is a fundamental problem in virtual human modeling and the digitization of hairstyles. However, existing methods still suffer from a significant trade-off between accuracy and efficiency. Implicit neural representations can capture the global hair shape but often fail to preserve fine-grained strand details, while explicit optimization-based approaches achieve high-fidelity reconstructions at the cost of heavy computation and poor scalability. To address this issue, we propose EfficientMonoHair, a fast and accurate framework that combines the implicit neural network with multi-view geometric fusion for strand-level reconstruction from monocular video. Our method introduces a fusion-patch-based multi-view optimization that reduces the number of optimization iterations for point cloud direction, as well as a novel parallel hair-growing strategy that relaxes voxel occupancy constraints, allowing large-scale strand tracing to remain stable and robust even under inaccurate or noisy orientation fields. Extensive experiments on representative real-world hairstyles demonstrate that our method can robustly reconstruct high-fidelity strand geometries with accuracy. On synthetic benchmarks, our method achieves reconstruction quality comparable to state-of-the-art methods, while improving runtime efficiency by nearly an order of magnitude. --- ### BodhiPromptShield: Pre-Inference Prompt Mediation for Surface-Form Privacy Propagation in LLM Agent Pipelines Authors: Bo Ma, Jinsong Wu, Weiqi Yan Published: 2026-04-07 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2604.05793v2 In LLM agent pipelines, prompt privacy risk propagates beyond a single model call: raw user content enters retrieval queries, memory writes, tool arguments, OCR-derived text, and logs, and every downstream copy inherits what the first write contained. Existing de-identification pipelines protect document boundaries but not this cross-stage surface. We present BodhiPromptShield, a policy-aware mediation layer that detects sensitive spans before they propagate, replaces each with a typed placeholder, a semantic abstraction, or a secure symbolic token under a configured policy, and defers restoration to authorized execution boundaries. We evaluate it under one protocol against Presidio, Casper-style sanitization, an LLM sanitizer, and transformer and learned detectors, on 300 AI4Privacy documents, 493 PrivacyLens trajectories, 200 PrivacyLens tasks scored by that benchmark's own judge, and AgentDojo tasks under injection. Three findings result. Identifier propagation is controllable: residual exposure falls to 7.4% on AI4Privacy and 1.8% on PrivacyLens, and exact identifiers in an agent's final action fall from 13.7% to 2.1-3.1%. Restoration timing governs what every stage upstream of the authorized boundary sees: deferring it leaves 1.6% of protected values readable in the released context against 51.0%, and 2.7% against 4.8% in what the agent emits, for 0.11 helpfulness points. Measuring factual disclosure is harder: a word-overlap metric and an LLM judge both report that mediation leaves facts intact, and both disagree with blind human annotation (kappa = 0.25 and 0.09). The human labels reverse that: inferability falls from 100% to 24-53% under mediation, so semantic-leakage measures need human validation before they are trusted. These are systems results on English text with open-weight models, not formal guarantees. --- ### BodhiPromptShield: Pre-Inference Prompt Mediation for Suppressing Privacy Propagation in LLM/VLM Agents Authors: Bo Ma, Jinsong Wu, Weiqi Yan Published: 2026-04-07 Categories: cs.CR, cs.CV arXiv: https://arxiv.org/abs/2604.05793v1 In LLM/VLM agents, prompt privacy risk propagates beyond a single model call because raw user content can flow into retrieval queries, memory writes, tool calls, and logs. Existing de-identification pipelines address document boundaries but not this cross-stage propagation. We propose BodhiPromptShield, a policy-aware framework that detects sensitive spans, routes them via typed placeholders, semantic abstraction, or secure symbolic mapping, and delays restoration to authorized boundaries. Relative to enterprise redaction, this adds explicit propagation-aware mediation and restoration timing as a security variable. Under controlled evaluation on the Controlled Prompt-Privacy Benchmark (CPPB), stage-wise propagation suppresses from 10.7\% to 7.1\% across retrieval, memory, and tool stages; PER reaches 9.3\% with 0.94 AC and 0.92 TSR, outperforming generic de-identification. These are controlled systems results on CPPB rather than formal privacy guarantees or public-benchmark transfer claims. The project repository is available at https://github.com/mabo1215/BodhiPromptShield.git. --- ### Sparse Gain Radio Map Reconstruction With Geometry Priors and Uncertainty-Guided Measurement Selection Authors: Zhihan Zeng, Ning Wei, Kaihe Wang, Phee Lep Yeoh, Fei Xu, Yue Xiu, Zhongpei Zhang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05788v2 Radio maps are important for environment-aware wireless communication, network planning, and radio resource optimization. However, dense radio map construction remains challenging when only a limited number of measurements are available, especially in complex urban environments with strong blockages, irregular geometry, and restricted sensing accessibility. Existing methods have explored interpolation, low-rank cartography, deep completion, and channel knowledge map (CKM) construction, but many of these methods insufficiently exploit explicit geometric priors or overlook the value of predictive uncertainty for subsequent sensing. In this paper, we study sparse gain radio map reconstruction from a geometry-aware and active sensing perspective. We first construct \textbf{UrbanRT-RM}, a controllable ray-tracing benchmark with diverse urban layouts, multiple base-station deployments, and multiple sparse sampling modes. We then propose \textbf{GeoUQ-GFNet}, a lightweight network that jointly predicts a dense gain radio map and a spatial uncertainty map from sparse measurements and structured scene priors. The predicted uncertainty is further used to guide active measurement selection under limited sensing budgets. Extensive experiments show that our proposed GeoUQ-GFNet method achieves strong and consistent reconstruction performance across different scenes and transmitter placements generated using UrbanRT-RM. Moreover, uncertainty-guided querying provides more effective reconstruction improvement than non-adaptive sampling under the same additional measurement budget. These results demonstrate the effectiveness of combining geometry-aware learning, uncertainty estimation, and benchmark-driven evaluation for sparse radio map reconstruction in complex urban environments. --- ### Sparse Gain Radio Map Reconstruction With Geometry Priors and Uncertainty-Guided Measurement Selection Authors: Zhihan Zeng, Ning Wei, Muhammad Baqer Mollah, Kaihe Wang, Phee Lep Yeoh, Fei Xu, Yue Xiu, Zhongpei Zhang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05788v1 Radio maps are important for environment-aware wireless communication, network planning, and radio resource optimization. However, dense radio map construction remains challenging when only a limited number of measurements are available, especially in complex urban environments with strong blockages, irregular geometry, and restricted sensing accessibility. Existing methods have explored interpolation, low-rank cartography, deep completion, and channel knowledge map (CKM) construction, but many of these methods insufficiently exploit explicit geometric priors or overlook the value of predictive uncertainty for subsequent sensing. In this paper, we study sparse gain radio map reconstruction from a geometry-aware and active sensing perspective. We first construct \textbf{UrbanRT-RM}, a controllable ray-tracing benchmark with diverse urban layouts, multiple base-station deployments, and multiple sparse sampling modes. We then propose \textbf{GeoUQ-GFNet}, a lightweight network that jointly predicts a dense gain radio map and a spatial uncertainty map from sparse measurements and structured scene priors. The predicted uncertainty is further used to guide active measurement selection under limited sensing budgets. Extensive experiments show that our proposed GeoUQ-GFNet method achieves strong and consistent reconstruction performance across different scenes and transmitter placements generated using UrbanRT-RM. Moreover, uncertainty-guided querying provides more effective reconstruction improvement than non-adaptive sampling under the same additional measurement budget. These results demonstrate the effectiveness of combining geometry-aware learning, uncertainty estimation, and benchmark-driven evaluation for sparse radio map reconstruction in complex urban environments. --- ### RHVI-FDD: A Hierarchical Decoupling Framework for Low-Light Image Enhancement Authors: Junhao Yang, Bo Yang, Hongwei Ge, Yanchun Liang, Heow Pueh Lee, Chunguo Wu Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05781v1 Low-light images often suffer from severe noise, detail loss, and color distortion, which hinder downstream multimedia analysis and retrieval tasks. The degradation in low-light images is complex: luminance and chrominance are coupled, while within the chrominance, noise and details are deeply entangled, preventing existing methods from simultaneously correcting color distortion, suppressing noise, and preserving fine details. To tackle the above challenges, we propose a novel hierarchical decoupling framework (RHVI-FDD). At the macro level, we introduce the RHVI transform, which mitigates the estimation bias caused by input noise and enables robust luminance-chrominance decoupling. At the micro level, we design a Frequency-Domain Decoupling (FDD) module with three branches for further feature separation. Using the Discrete Cosine Transform, we decompose chrominance features into low, mid, and high-frequency bands that predominantly represent global tone, local details, and noise components, which are then processed by tailored expert networks in a divide-and-conquer manner and fused via an adaptive gating module for content-aware fusion. Extensive experiments on multiple low-light datasets demonstrate that our method consistently outperforms existing state-of-the-art approaches in both objective metrics and subjective visual quality. --- ### Sparsity-Aware Voxel Attention and Foreground Modulation for 3D Semantic Scene Completion Authors: Yu Xue, Longjun Gao, Yuanqi Su, HaoAng Lu, Xiaoning Zhang Published: 2026-04-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.05780v1 Monocular Semantic Scene Completion (SSC) aims to reconstruct complete 3D semantic scenes from a single RGB image, offering a cost-effective solution for autonomous driving and robotics. However, the inherently imbalanced nature of voxel distributions, where over 93% of voxels are empty and foreground classes are rare, poses significant challenges. Existing methods often suffer from redundant emphasis on uninformative voxels and poor generalization to long-tailed categories. To address these issues, we propose VoxSAMNet (Voxel Sparsity-Aware Modulation Network), a unified framework that explicitly models voxel sparsity and semantic imbalance. Our approach introduces: (1) a Dummy Shortcut for Feature Refinement (DSFR) module that bypasses empty voxels via a shared dummy node while refining occupied ones with deformable attention; and (2) a Foreground Modulation Strategy combining Foreground Dropout (FD) and Text-Guided Image Filter (TGIF) to alleviate overfitting and enhance class-relevant features. Extensive experiments on the public benchmarks SemanticKITTI and SSCBench-KITTI-360 demonstrate that VoxSAMNet achieves state-of-the-art performance, surpassing prior monocular and stereo baselines with mIoU scores of 18.2% and 20.2%, respectively. Our results highlight the importance of sparsity-aware and semantics-guided design for efficient and accurate 3D scene completion, offering a promising direction for future research. ---