# Research Papers — 2026-04 · bundle page 18 of 233 > Full-text markdown of papers #851–900 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/18 [← Previous bundle](https://aipod.dev/papers/archive/2026-04/page/17.md) [Next bundle →](https://aipod.dev/papers/archive/2026-04/page/19.md) ### ViBE: Visual-to-M/EEG Brain Encoding via Spatio-Temporal VAE and Distribution-Aligned Projection Authors: Ganxi Xu, Zhao-Rong Lai, Yuting Tang, Yonghao Song, Shuyan Zhou, Guoxu Zhou, Boyu Wang, Jian Zhu, Jinyi Long Published: 2026-04-29 Categories: cs.CV arXiv: https://arxiv.org/abs/2604.26218v1 Brain encoding models not only serve to decipher how visual stimuli are transformed into neural responses, but also represent a critical step toward visual prostheses that restore vision for patients with severe vision disorders. Brain encoding involves two fundamental steps: achieving faithful reconstruction of neural responses and establishing cross-modal alignment between visual stimuli and neural responses. To this end, we propose ViBE, a novel brain encoding framework for generating magnetoencephalography (MEG) and electroencephalography (EEG) signals from visual stimuli. Specifically, we first design a spatio-temporal convolutional variational autoencoder (TSC-VAE) that captures the spatio-temporal characteristics of M/EEG signals for effective neural response reconstruction. To bridge the modality gap between visual features and neural representations, we employ Q-Former to map CLIP image embeddings to the TSC-VAE latent space, producing neural proxy embeddings. For comprehensive cross-modal alignment, we combine mean squared error (MSE) loss for point-wise feature matching with sliced Wasserstein distance (SWD) for probability distribution alignment between the neural proxy embeddings and TSC-VAE latent embeddings. We conduct extensive experiments on the THINGS-EEG2 and THINGS-MEG datasets, demonstrating the effectiveness of our approach in generating high-quality M/EEG signals from visual stimuli. --- ### Unsupervised Graph Modeling for Anomaly Detection in Accounting Subject Relationships Authors: Yuhan Wang, Ruobing Yan, Zhe Su, Hejing Chen, Ningjing Sang, Yunfei Nie Published: 2026-04-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.26216v1 This paper addresses the problem of anomaly detection in accounting subject association structures, proposing a structured modeling and unsupervised discriminant framework based on graph neural networks. This framework is used to mine stable correspondences between subjects and identify structural deviations from general ledger details and voucher entries. The method first abstracts accounting subjects as graph nodes, and the co-occurrence and debit/credit correspondence of subjects in the same business record are abstracted as weighted edges. The edge weights are characterized by statistical measures such as co-occurrence frequency or amount aggregation, thus forming a period-level accounting subject association graph. In the representation learning stage, a message passing mechanism is used to fuse the node's own attributes and neighborhood context to obtain node embeddings containing structural information. In the anomaly detection stage, the rationality of subject pair connections is estimated through a relation reconstruction decoder, and edge-level anomaly scores are defined based on the degree of deviation in reconstruction probabilities. These scores are then aggregated to obtain node-level risk ranking and local anomaly localization. This framework can simultaneously capture local substructure anomalies and cross-community anomaly connections without relying on anomaly labeling, outputting traceable subject pair risk clues. Comparative experiments demonstrate more stable comprehensive discriminant capabilities and higher top-ranking accuracy. --- ### Qvine: Vine Structured Quantum Circuits for Loading High Dimensional Distributions Authors: David Quiroga, Hannes Leipold, Bibhas Adhikari Published: 2026-04-29 Categories: quant-ph, cs.AI arXiv: https://arxiv.org/abs/2604.26213v1 Loading high dimensional distributions is an important task for utilizing quantum computers on applications ranging from machine learning to finance. The high dimensionality leads to a curse of dimensionality, representing a d-dimensional distribution with k resolution requires dk qubits and an unstructured parameterized circuit would express a unitary in an exponential operator space in the number of qubits, leading to vanishing gradients and poor convergence guarantees even at high depth. Vine copula decompositions are widely used to represent high dimensional distributions classically, showing high quality approximation in many important applications, such as financial modeling. We present Qvine, a vine structured ansatz for quantum circuits, that mirrors the vine decomposition to construct scalable quantum circuits with efficient trainability while achieving similarly high quality approximation for amplitude encoding distributions. For regular vines (R-vines), we show that the circuit depth scales at most quadratic in the dimension of the distribution, while for D-vines, as well as many practical R-vines, the circuit depth scales linear in the dimension. For 3-dimensional and 4-dimensional Gaussians and empirical joint stock price return distributions for selected stocks, our experiments show Qvines achieve high quality loading. --- ### OMEGA: Optimizing Machine Learning by Evaluating Generated Algorithms Authors: Jeremy Nixon, Annika Singh Published: 2026-04-29 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.26211v1 In order to automate AI research we introduce a full, end-to-end framework, OMEGA: Optimizing Machine learning by Evaluating Generated Algorithms, that starts at idea generation and ends with executable code. Our system combines structured meta-prompt engineering with executable code generation to create new ML classifiers. The OMEGA framework has been utilized to generate several novel algorithms that outperform scikit-learn baselines across a robust selection of 20 benchmark datasets (infinity-bench). You can access models discussed in this paper and more in the python package: pip install omega-models. --- ### Breaking the Autoregressive Chain: Hyper-Parallel Decoding for Efficient LLM-Based Attribute Value Extraction Authors: Theodore Glavas, Nikhita Vedula, Dushyanta Dhyani, Yilun Zhu, Shervin Malmasi Published: 2026-04-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.26209v1 Some text generation tasks, such as Attribute Value Extraction (AVE), require decoding multiple independent sequences from the same document context. While standard autoregressive decoding is slow due to its sequential nature, the independence between output sequences offers an opportunity for parallelism. We present Hyper-Parallel Decoding, a novel decoding algorithm that accelerates offline decoding by leveraging both shared memory and computation across batches. HPD enables out-of-order token generation through position ID manipulation, significantly improving efficiency. Experiments on AVE show that attribute-value pairs are conditionally independent, enabling us to parallelize value generation within each prompt. By further stacking multiple documents within a single prompt, we can decode in parallel up to 96 tokens per prompt. HPD works with all LLMs, and reduces both inference costs and total inference time by up to 13.8X without compromising output quality, potentially saving hundreds of thousands of dollars on industry AVE tasks. Although designed for attribute extraction, HPD makes no assumptions unique to the AVE domain and can in theory be applied to other scenarios with independent output structures. --- ### Option-Order Randomisation Reveals a Distributional Position Attractor in Prompted Sandbagging Authors: Jon-Paul Cacioli Published: 2026-04-29 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.26206v1 A predecessor pilot (Cacioli, 2026) found that Llama-3-8B implements prompted sandbagging as positional collapse rather than answer avoidance. However, fixed option ordering in MMLU-Pro left open whether this reflected a model-level position-dominant policy or dataset-level distractor structure. This pre-registered follow-up (3 models, 2,000 MMLU-Pro items, 4 conditions, 24,000 primary trials) added cyclic option-order randomisation as the critical control. The pre-registered item-level same-letter diagnostic did not confirm deterministic position-tracking (same-letter rate 37.3%, below the 50% threshold). However, pre-specified supporting analyses revealed that the response-position distribution under sandbagging was highly stable under complete content rotation (Pearson r = 0.9994; Jensen-Shannon divergence = 0.027, compared to 0.386 between honest and sandbagging conditions). Accuracy spiked to 72.1% when the correct answer coincidentally occupied the preferred position E, and fell to 4.3% at position A. The data provide strong evidence for a soft distributional attractor: under sandbagging instruction, the model enters a low-entropy response-position basin centred on E/F/G that is highly stable and largely content-invariant at the aggregate level. Qwen-2.5-7B served as a negative control (non-compliant, no distributional shift). These results provide evidence, at the 7-9 billion parameter scale, that response-position entropy is a promising black-box behavioural signature of this sandbagging mode. --- ### Hierarchical Long-Term Semantic Memory for LinkedIn's Hiring Agent Authors: Zhentao Xu, Shangjin Zhang, Emir Poyraz, Yvonne Li, Ye Jin, Xie Lu, Xiaoyang Gu, Karthik Ramgopal, Praveen Kumar Bodigutla, Xiaofeng Wang Published: 2026-04-29 Categories: cs.IR, cs.LG arXiv: https://arxiv.org/abs/2604.26197v3 Large Language Model (LLM) agents are increasingly used in real-world products, where personalized and context-aware user interactions are essential. A central enabler of such capabilities is the agent's long-term semantic memory system, which extracts implicit and explicit signals from noisy longitudinal behavioral data, stores them in a structured form, and supports low-latency retrieval. Building industrial-grade long-term memory for LLM agents raises five challenges: scalability, low-latency retrieval, privacy constraints, adaptability, and observability. We introduce the Hierarchical Long-Term Semantic Memory (HLTM) framework, which organizes textual data into a schema-aligned memory tree that captures semantic knowledge at multiple levels of granularity, enabling scalable ingestion, privacy-aware storage, low-latency retrieval, and transparent provenance; HLTM further incorporates an adaptation mechanism to generalize across diverse use cases. Extensive evaluations on LinkedIn's Hiring Assistant show that HLTM improves answer correctness by more than 5% and retrieval F1 by more than 10%, while significantly advancing the Pareto frontier between query and indexing latency. HLTM has been fully deployed in LinkedIn's Hiring Assistant to power core personalization features in production hiring workflows. --- ### Flashback: A Reversible Bilateral Run-Peeling Decomposition of Strings Authors: Thomas Konstantinovsky, Gur Yaari Published: 2026-04-29 Categories: cs.DS, cs.CL arXiv: https://arxiv.org/abs/2604.26190v1 We introduce Flashback, a reversible string decomposition that repeatedly peels the maximal leading and trailing character runs from a sentinel-wrapped input, recording each pair as one bilateral token. Decomposition and reconstruction both run in O(n) time and space. Our central result is a run-pairing theorem: Flashback is equivalent to pairing the first run of the string with the last, the second with the second-to-last, and so on. This gives an exact token count of 1+(r/2) for a string with r maximal runs, and matches a lower bound that holds for any admissible bilateral run-peeling scheme. From the run-pairing theorem the main structural properties follow as corollaries: the irreducible peeling kernel uses at most two symbols; palindromes are precisely the strings whose run-length encoding is symmetric with an odd number of runs; the image of the decomposition admits an explicit finite-state characterisation; and changing one run length rewrites exactly one content token. --- ### Efficient and Interpretable Transformer for Counterfactual Fairness Authors: Panyi Dong, Zhiyu Quan Published: 2026-04-29 Categories: cs.LG arXiv: https://arxiv.org/abs/2604.26188v1 The growing reliance of machine learning models in high-stakes, highly regulated domains such as finance and insurance has created a growing tension between predictive performance, interpretability, and regulatory fairness requirements. In these settings, models are expected not only to deliver reliable predictions but also to provide transparent decision rationales and comply with strict fairness requirements. Attention-based transformers offer powerful mechanisms for modeling complex data relationships as demonstrated in various language tasks, yet their attention mechanisms alone do not ensure counterfactually fair predictions, even when combined with fairness-aware techniques. To address these limitations, we propose the Feature Correlation Transformer (FCorrTransformer), an attention-light architecture tailored for tabular data. In this design, the attention matrix admits a direct statistical interpretation as pairwise feature dependencies, enhancing both interpretability and efficiency. Leveraging this structure, we introduce Counterfactual Attention Regularization (CAR), a framework that enforces group-invariant fair representations of sensitive features at the attention level, promoting counterfactually fair predictions without relying on explicit causal assumptions. Empirical evaluations on imbalanced classification and regression benchmarks demonstrate that FCorrTransformer combined with CAR achieves strong counterfactual fairness while maintaining competitive predictive performance and substantially reducing model complexity compared with standard transformer-based baselines. Overall, this work bridges a critical gap between fairness theory and machine learning models, offering a practical framework for responsible AI in regulatory-sensitive domains. --- ### FASH-iCNN: Making Editorial Fashion Identity Inspectable Through Multimodal CNN Probing Authors: Morayo Danielle Adeyemi, Ryan A. Rossi, Franck Dernoncourt Published: 2026-04-29 Categories: cs.CV, cs.HC, cs.IR, cs.MM arXiv: https://arxiv.org/abs/2604.26186v1 Fashion AI systems routinely encode the aesthetic logic of specific houses, editors, and historical moments without disclosing it. We present FASH-iCNN, a multimodal system trained on 87,547 Vogue runway images across 15 fashion houses spanning 1991-2024 that makes this cultural logic inspectable. Given a photograph of a garment, the system recovers which house produced it, which era it belongs to, and which color tradition it reflects. A clothing-only model identifies the fashion house at 78.2% top-1 across 14 houses, the decade at 88.6% top-1, and the specific year at 58.3% top-1 across 34 years with a mean error of just 2.2 years. Probing which visual channels carry this signal reveals a sharp dissociation: removing color costs only 10.6pp of house identity accuracy, while removing texture costs 37.6pp, establishing texture and luminance as the primary carriers of editorial identity. FASH-iCNN treats editorial culture as the signal rather than background noise, identifying which houses, eras, and color traditions shaped each output so that users can see not just what the system predicts but which houses, editors, and historical moments are encoded in that prediction. --- ### Privacy-Preserving Clothing Classification using Vision Transformer for Thermal Comfort Estimation Authors: Tatsuya Chuman, Yousuke Udagawa, Hitoshi Kiya Published: 2026-04-29 Categories: cs.CV, cs.CR arXiv: https://arxiv.org/abs/2604.26184v1 A privacy-preserving clothing classification scheme is presented to enable secure occupant-centric control (OCC) systems. Although the utilization of camera images for HVAC control has been widely studied to optimize thermal comfort, privacy protection of occupant images has not been considered in prior works. While various privacy-preserving methods have been proposed for image classification, applying conventional schemes results in severe accuracy degradation. In this paper, we introduce a privacy-preserving classification method using Vision Transformer (ViT) applied to clothing insulation estimation. In an experiment using the DeepFashion dataset categorized by clothing insulation, while the conventional pixel-based method suffers a severe accuracy drop, our scheme maintains a high accuracy on encrypted images, showing no degradation from plain images across all categories. --- ### From Agent Failure Paths to Quantified Residual Risk: A Compositional Framework for Resilient Agentic AI Authors: Hassan Karim, Sai Sitharaman, Deepti Gupta, Danda B. Rawat Published: 2026-04-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2607.18243v1 Agentic AI is crossing trust boundaries faster than current risk models can represent. Existing approaches provide one of two partial views. They either describe failure mechanisms without producing a transferable residual-risk estimate, or they produce a risk estimate while treating the internal failure path as a black box. We couple those two views by proposing CPSAINT, a seven-layer integrity decomposition over Physical state, Sensors, Data, Compute, Actuators, Environment, and Time, paired with FRIESA-K, a residual-risk functional that maps each failure path to a quantified risk instance. FRIESA-K grounds the resistance term K in a controlled absorbing Markov model so that control effectiveness is derived from state dynamics rather than assigned as an informal score. The result is a concise mechanism-to magnitude pipeline for resilient agentic and embodied AI. We report governance observability through a separate additive penalty instead of inserting governance as a new variable in the resistance functional. We formalize structural composability linking valid failure paths to well-defined risk instances and show the framework on two contrasting scenarios a hard real-time warehouse robot and a governance-instrumented financial-services agent. Across both cases, the same layer grammar, variable semantics, and dynamic-resistance construction remain intact. Thus, we obtain a compact kernel that supports cross-domain reasoning, explicit assumptions, and quantitatively grounded formalism of composable trust. --- ### Investigating Linguistic Steering: An Analysis of Adjectival Effects Across Large Language Model Architectures Authors: Lars Malmqvist Published: 2026-04-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.20572v1 Achieving reliable control of Large Language Models (LLMs) requires a precise, scalable understanding of how they interpret linguistic cues. We introduce a rigorous framework using Shapley values to quantify the steering effect of individual adjectives on model performance, moving beyond anecdotal heuristics to principled attribution. Applying this method to 100 adjectives across a diverse suite of models (including o3, gpt-4o-mini, phi-3, llama-3-70b, and deepseek-r1) on the MMLU benchmark, we uncover several critical findings for AI alignment. First, we find that a small subset of adjectives act as disproportionately powerful "levers," yet their effects are not universal. Cross-model analysis reveals a "family effect": models of a shared lineage exhibit correlated sensitivity profiles, while architecturally distinct models react in a largely uncorrelated manner, challenging the notion of a one-size-fits-all prompting strategy. Second, focused follow-up studies demonstrate that the steering direction of these powerful adjectives is not intrinsic but is highly contingent on their syntactic role and position within the prompt. For larger models like gpt-4o-mini, we provide the first quantitative evidence of strong, non-additive interaction effects where adjectives can synergistically amplify, antagonistically dampen, or even reverse each other's impact. In contrast, smaller models like phi-3 exhibit a more literal and less compositional response. These results suggest that as models scale, their interpretation of prompts becomes more sophisticated but also less predictable, posing a significant challenge for robustly steering model behavior and highlighting the need for compositional and model-specific alignment techniques. --- ### Granularity-Regulated Adaptive Computational Efficiency for Optimal Verification in Test-Time Scaling Authors: Ardit Krasniqi, Luan Vejsiu, Elira Dervishi Published: 2026-04-28 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.19354v1 Test-time scaling (TTS) has emerged as a powerful paradigm for improving the reasoning performance of large language models (LLMs) by investing additional compute at inference time. A central component of TTS is the \emph{verifier}, which selects or scores candidate solutions to guide the search process. While prior work has explored the benefit of verification, a fundamental question remains underexplored: \emph{what is the optimal granularity of verification under a given compute budget?} Coarse-grained outcome reward models (ORMs) and fine-grained process reward models (PRMs) represent two extremes, yet neither alone achieves compute-optimality across all regimes. In this paper, we establish a unified theoretical framework, called \textbf{GRACE} (\underline{G}ranularity-\underline{R}egulated \underline{A}daptive \underline{C}omputational \underline{E}fficiency), that characterizes the optimal verification granularity as an explicit function of problem difficulty, verifier accuracy, and compute budget. We prove that there exists a phase transition: fine-grained verification dominates when either the compute budget is large or the problem is hard, whereas coarse-grained verification is preferred in the low-budget, easy-problem regime. Our theory unifies Best-of-$N$, beam search, and step-level MCTS within a single Pareto-optimality framework, and motivates an adaptive granularity strategy that provably achieves the compute-performance Pareto frontier. Empirical results on MATH-500, GSM8K, and AIME benchmarks corroborate all four theoretical claims, with our adaptive strategy outperforming fixed-granularity baselines by up to 3.1\% accuracy at matched compute. --- ### Quantifying Aleatoric Uncertainty of In-Context Learning for Robust Measure of LLM Prediction Confidence Authors: Jinseok Chung, Minkyoung Song, Hyunji Jung, Namhoon Lee Published: 2026-04-28 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.19353v1 In-Context Learning (ICL) allows LLMs to adapt to new tasks from a few demonstrations, but its reliability remains a concern: predictions are highly sensitive to both prompt design and the model's ability to understand the context, obscuring whether failures arise from data properties or model limitations. Uncertainty decomposition-separating aleatoric from epistemic sources-is particularly crucial in this setting, yet existing methods, designed for standard generation tasks, fail to capture the unique dynamics of ICL. To address this, we introduce a concept of self-function vectors, built upon Bayesian views and the mechanistic interpretability of ICL. These vectors leverage internal model representations to model the latent concept learned during in-context prompting, thereby enabling a direct estimation of aleatoric uncertainty within a Bayesian framework and circumventing the reliance on brittle input or decoding manipulations. Given the lack of established benchmarks and suitable evaluation protocols, we also propose the first and rigorous evaluation protocol, in which data is manipulated in controlled ways so as to quantify aleatoric uncertainty precisely and separately from epistemic uncertainty. With this new evaluation framework, initially grounded in synthetic tasks for conceptual development and subsequently extended to real-world datasets, we show that our proposed methodology can measure uncertainty of LLM predictions made under ICL more reliably than existing alternative methods. Moreover, we show it can be used as a practical tool for trustworthy-related applications, such as hallucination detection. Our findings pave a new direction for connecting the quantitative view of uncertainty with the mechanistic understanding of model behavior. --- ### Sign-Language Datasets at Scale: A Comprehensive Survey on Resources, Benchmarks, and Annotation Standards Authors: Yiming Ni, Zhi-Qi Cheng, Jiayu Li, Wei Cheng Published: 2026-04-28 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2606.19352v1 Sign languages are expressive visual languages used by Deaf and Hard-of-Hearing (DHH) communities. Despite substantial progress in sign-language recognition, translation, and production, advances remain constrained by fragmented datasets, inconsistent annotations, and limited linguistic coverage. Existing benchmarks often fail to reflect real-world communication needs, and systematic analyses of these limitations remain limited. In this survey, we present a comprehensive index of sign-language datasets, covering 120 resources across 35 sign languages. We analyze key challenges such as modality imbalance, annotation granularity, and signer bias, and outline considerations for future dataset design. We also introduce a 24-field Sign-Language Datasheet and release a public GitHub repository (https://github.com/Ginqwerty/Open-Sign-Language) to support standardized documentation and reproducible evaluation. Overall, our work provides a unified and practical foundation for developing inclusive, robust, and scalable sign-language technologies in real-world applications. --- ### Density Field State Space Models: 1-Bit Distillation, Efficient Inference, and Knowledge Organization in Mamba-2 Authors: Chirag Shinde Published: 2026-04-28 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2606.10932v1 We present Density Field State Space Models (DF-SSM), a framework for compressing SSMs to a 1-bit scaffold with int8 low-rank correction. Applied to Mamba-2 1.3B, we achieve a 278 MB model (9.7x smaller than the 2.7 GB FP16 teacher) that runs at 21.4x faster inference on GPU (batch=1, relative to the mamba-ssm reference implementation) while maintaining downstream task performance within 2-4 percentage points of BitMamba-2, a 1.58-bit model trained from scratch on 150B tokens. The distillation itself requires only 32M tokens and 6 hours on a single A100 GPU, though it presupposes a pretrained FP16 teacher. We develop an optimized inference pipeline combining cuBLAS INT8 tensor cores for the scaffold matmul, custom CUDA kernels for stateful SSM and convolution operations, and an AVX-512 CPU backend for efficient deployment on both GPU and CPU. Beyond compression, we investigate the internal knowledge organization of the resulting model, discovering three distinct processing phases: intent classification (layers 0-3, operating in an abstract space with no vocabulary alignment), knowledge retrieval (layers 25-35, where factual associations localize to a 5-layer window), and output formatting (layers 36-47, where category structure dissolves). Through systematic analysis of 445 factual prompts across 19 categories, we find that early-layer classification is syntactic (driven by template structure) rather than semantic, and that the model exhibits well-organized knowledge representations despite weak factual recall--suggesting that representational structure may precede factual strength. --- ### Beware of GeeksBearing Gifts: Building True EU Frontier AI Sovereignty Authors: Nick Moës, Toni Lorente, Amin Oueslati, Jonathan Smith, Robin Staes-Polet, Radina Kraeva Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.07536v1 Frontier artificial intelligence is reshaping all aspects of society, from economic output or military capability to democratic institutions. The EU is entering this transformation from a position of structural dependence: frontier models originate almost exclusively from the United States or China, the US holds approximately sixteen times the EU's AI supercomputing capacity, and only 15% of global hyperscale data centre capacity resides within EU borders. Although the European Commission has accelerated its policy response, existing initiatives remain fragmented and lack a cohesive vision for securing strategic autonomy across the full frontier AI value chain. Here we propose a unified framework connecting five sovereignty pillars (economic competitiveness, resilience, security and defence, European values, and foreign relations) to a decomposition of the frontier AI stack comprising five layers, 26 components, and 29 sub-components. This framework allows the identification of critical gaps, redundancies, and inter-pillar trade-offs that current EU policy leaves implicit. Our analysis of the AI Gigafactory Initiative illustrates how a sovereignty-centred lens reveals conflicts that narrowly economic framings obscure. Moreover, this framework offers policymakers a structured basis for designing, evaluating, and prioritising frontier AI interventions across multiple dimensions of European strategic autonomy across the 92 initiatives from four major Commission communications we. identify, and beyond. --- ### Assessing the Geographic Diversity of AI's Platial Representations in Image Generation Authors: Zilong Liu, Krzysztof Janowicz, Mina Karimi Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.05188v1 (Gen)AI diversity is not merely an ethical issue. From the perspective of geographic information science (GIScience), it could be interpreted as a function of uncertainty and as a form of cognitive bias, embedded in AI outputs. Recent work has sought to develop information-theoretic diversity measures and apply them to evaluate AI-chatbot outputs in a geographic context. As the AI ecosystem to which we are exposed on a daily basis becomes rapidly multimodal, we believe it is important to examine geographic diversity across various modalities. Focusing on images, this paper aims to fill this research gap. First, we select the GPT and DALL-E models as state-of-the-art examples and point out how assessing their geographic diversity involves various stages, including prompt revision and image generation. Then, taking inspiration from species diversity measures in ecological research, we incorporate similarity weighting into the measurement of geographic diversity. Next, we demonstrate how to evaluate geographic diversity in image generation through a case study. Our analysis reveals several counterintuitive findings. For instance, older models can exhibit greater geographic diversity despite producing lower-quality images, and prompt revision yields greater geographic diversity than image generation. At the same time, we observe explicit model homogeneity underlying the lack of geographic diversity, as the selected models consistently depict the same prototypical geo-specific feature or similar features. This is concerning, as it risks producing stereotypical representations of places. --- ### Geographic Bias and Diversity in AI Evaluation Authors: Zilong Liu, Krzysztof Janowicz, Gengchen Mai, Song Gao, Rui Zhu Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.05187v1 Among the many challenges hindering the responsible development and deployment of AI, arguably none has faced more intense scrutiny than bias in its various forms. This underscores the widespread concerns across AI researchers that model outputs, e.g., from generative AI, may encode structural distributional imbalances (stemming from training data or model design) that may amplify social inequality or introduce systemic distortions across application domains ranging from biodiversity to disaster mitigation. Yet, relatively little work has investigated the geographical nature of bias or developed measurable benchmarks for what it means for (generative) AI to be unbiased. In this chapter, we investigate this issue through a literature review. As foundation models are reshaping the landscape of bias research, we examine work spanning both the pre-generative AI and generative AI periods. First, we identify a range of geographic biases. These biases span from representation bias in the training data and regional disparities in the factual recall of language models to the tendency of generative AI to over-proportionally favor prototypical places (called defaults). Then, we showcase how recent studies address the latter bias by evaluating geographic diversity in the outputs of generative AI across various cognitive levels, parameter settings, and output modalities. --- ### Improving Hospital Process Management through Process Mining: A Case Study on COVID-19 Clinical Pathways Authors: Pasquale Ardimento, Mario Luca Bernardi, Marta Cimitile, Samuele Latorre Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.00041v1 This study analyzes COVID-19 care pathways using the COVID Data for Shared Learning dataset. We build a transparent, reproducible pipeline that transforms heterogeneous clinical tables into a process-mining-ready event log and applies discovery, declarative conformance checking, and outcome analysis. The reconstructed pathways highlight the monitoring backbone of inpatient care, variability at the Emergency department-admission interface, and outcome differences driven by age and exposure to intensive care units. These insights support triage standardization, capacity planning, and step-down coordination from intensive care units to lower-acuity wards, showing how process mining can inform evidence-based hospital governance. --- ### Tracing GenAI Literacy: Uncovering Student-AI Interaction Patterns in Academic Writing through Epistemic Network Analysis Authors: Angxuan Chen, Jiyou Jia Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.00040v1 As Generative AI (GenAI) becomes integral to education, fostering GenAI literacy is critical. However, current assessments largely rely on self-reported scales, lacking insights into how literacy manifests in actual learning processes. This study leverages Learning Analytics (LA) to bridge this gap. We collected interaction logs from 162 university students engaged in a GenAI-assisted abstract writing task. Using Epistemic Network Analysis (ENA), we modeled and compared the questioning strategies of students with varying GenAI literacy levels. Preliminary results reveal distinct interaction signatures: high-literacy students engage in iterative refinement and strategic questioning, while low-literacy students rely on direct generation commands. This work contributes to the workshop by demonstrating how process data can characterize GenAI literacy, paving the way for data-driven literacy assessment and real-time interventions. --- ### Beyond Categories of Caste: Examining Caste Bias and Morality in Text-to-Image AI Models Authors: Divyanshu Kumar Singh, Dipto Das, Deepika Rama Subramanian, Koustuv Saha, Stephen Voida, Bryan Semaan Published: 2026-04-28 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2606.00039v1 Text-to-Image (T2I) models have shown promising utility across various domains. However, such models are also amplifying harmful societal biases in their outputs. In the context of South Asia, recent work has shown caste biases and stereotypes are being perpetuated through Generative AI (GenAI) systems. While this research offers extremely relevant insight into invisibilized narratives of caste discrimination through the GenAI system, they often treat caste as an identity category. Therefore, in this work we shift our ontology to focus on the relational aspect of caste. This enables us to develop a more nuanced understanding of the mechanics of caste discrimination by and through T2I models. Combining an algorithmic audit with critical discourse analysis, we draw on a conceptual frame challenging Brahminical Normativity to show how caste biases are perpetuated beyond the simple binaries of upper vs lower-caste categories. Our contributions are two-fold. Beyond challenging the categorical understanding of caste as a category, we propose an anti-caste approach to tackle the issue of caste bias and fairness in AI systems. --- ### Beyond Tool Adoption: A Practical Five-Stage Developmental Continuum for AI Literacy in Higher Education Authors: J. Paul Liu, Rachel Levy Published: 2026-04-28 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2606.00038v5 Artificial intelligence (AI) literacy is increasingly recognized as a foundational competency for all university graduates. Yet students' engagement with AI tools often clusters at two extremes: avoidance driven by fear, mistrust, ethical concern, or lack of access, and uncritical reliance that produces fluent output while masking misunderstanding. Existing AI literacy frameworks provide valuable competency definitions, but most offer limited guidance for diagnosing where learners begin and how they progress toward responsible, critical engagement. This paper proposes a five-stage AI Literacy Continuum: 0) Not Yet Engaged, 1) Uncritical Use, 2) Informed Use, 3) Critical Evaluation, and 4) Improvement --that describes developmental orientations toward AI use in higher education. The continuum complements dimensional frameworks by providing educators with a practical diagnostic and instructional pathway aligned with international frameworks, including UNESCO and OECD. We present a design-based implementation case from North Carolina State University, where credit-bearing courses and intensive hands-on workshops engaged more than 330 participants between Fall 2024 and Spring 2026. Because the implementation did not use a validated pre/post instrument or comparison group, we frame the findings as observational and practice-based: participants exhibited behaviors consistent with movement from non-engagement or uncritical use toward informed engagement, while sustained and discipline-embedded experiences produced stronger evidence of critical evaluation and improvement-oriented practice. We discuss curricular pathways, opportunity considerations, assessment strategies, and argue that AI literacy should be understood not as tool adoption alone but as a developmental capacity to understand, evaluate, and responsibly apply AI systems in disciplinary and societal contexts. --- ### QASM-Eval: A Dataset to Train and Evaluate LLMs on OpenQASM-3 Beyond Quantum Circuits Authors: Zhenxiao Fu, Lei Jiang, Fan Chen Published: 2026-04-28 Categories: cs.LG, quant-ph arXiv: https://arxiv.org/abs/2605.30358v2 Quantum computing remains in the Noisy Intermediate-Scale Quantum (NISQ) era, with performance constrained by noise. Addressing this limitation requires hardware-facing capabilities beyond gate sequences: mid-circuit measurement and classical feedback for quantum error correction (QEC), precise timing for dynamical decoupling (DD), and pulse-level waveform access for calibration. OpenQASM 3 exposes these capabilities through a hardware-level programming interface. Despite rapid progress in large language models (LLMs) for code generation, datasets targeting these advanced features remain lacking. We introduce QASM-Eval, the first comprehensive dataset to train and evaluate LLMs on OpenQASM 3, targeting completion of hardware-facing constructs within a supplied program context. QASM-Eval comprises an expert-curated test set of 1,200 tasks and a training set of over 4,000 tasks, covering classical logic, timing scheduling, pulse control, and complex tasks inspired by quantum-control workflows. An extended verifier automatically checks syntax, quantum measurement distributions, and program timelines. Our evaluation identifies substantial headroom for OpenQASM 3 code generation and significant gains from targeted fine-tuning. Fine-tuned Llama-3-8B outperforms zero-shot GPT-5.6-Terra, while fine-tuned Llama-3-70B achieves 61.17% overall pass@1 and approaches few-shot-augmented GPT-5.6-Terra. QASM-Eval provides a benchmark and training resource for specification-guided code generation over the hardware-facing constructs of OpenQASM 3, supporting the development of LLM assistants for quantum programming. Data and code: https://github.com/fuzhenxiao/QASM-Eval --- ### QASM-Eval: A Dataset to Train and Evaluate LLMs on OpenQASM-3 Beyond Quantum Circuits Authors: Zhenxiao Fu, Lei Jiang, Fan Chen Published: 2026-04-28 Categories: cs.LG, quant-ph arXiv: https://arxiv.org/abs/2605.30358v1 Quantum computing remains in the Noisy Intermediate-Scale Quantum (NISQ) era, where the performance is highly constrained to noise. Addressing the limitation often requires hardware-facing capabilities beyond gate-sequence circuit specification, including mid-circuit measurement and classical feedback for quantum error correction (QEC), precise timing control for dynamical decoupling (DD), and pulse-level waveform access for calibration. OpenQASM-3 was introduced to expose exactly these capabilities, providing a hardware-level programming interface. However, despite the rapid progress of large language models in code generation, there is still no dataset specifically designed to train and evaluate LLMs on OpenQASM-3 programs that involve its advanced hardware-oriented features. To address this gap, we introduce QASM-Eval, the first comprehensive dataset designed to train and evaluate LLMs on OpenQASM-3. Rather than focusing on quantum algorithm design or reasoning, QASM-Eval explicitly targets the language's hardware-facing features. QASM-Eval comprises an expert-verified test set of 100 tasks and a training set of 4,000 tasks, systematically covering classical logic, timing scheduling, pulse control, and complex real-world workflows. To automatically validate generated programs, we check syntax, quantum states and program timeline using an extended verifier. Our evaluation reveals that while state-of-the-art LLMs struggle heavily in OpenQASM-3 coding tasks, targeted fine-tuning on QASM-Eval yields significant gains. QASM-Eval provides a crucial benchmark and training foundation to accelerate the development of reliable LLM assistants for hardware-facing quantum programming in NISQ era. Data and code: https://github.com/fuzhenxiao/QASM-Eval --- ### MEMOR-E: In-Context and Fine-Tuned LLM Personalization for Alzheimer's Assistive Robotics Authors: Maissa Abir Smaili, Eren Sadikoglu, Ransalu Senanayake Published: 2026-04-28 Categories: cs.AI, cs.RO arXiv: https://arxiv.org/abs/2605.23941v1 Alzheimer's disease is a neurodegenerative disorder marked by progressive declines in memory and language that reduce independence in daily life, motivating socially assistive robotic support. This paper presents MEMOR-E, a mobile quadruped robot with an interactive tablet interface that assists patients and caregivers through medication reminders, routine guidance, memory oriented interactions, and companionship. We evaluated the feasibility of fine tuning large language models (LLMs) to emulate stage consistent cognitive behavior and interpret responses across standard neuropsychological language tasks, using audio transcriptions from 235 Alzheimer's patients and synthetically generated healthy controls. We also report findings on using in context learning (ICL) in LLMs, where a second LLM produced domain and severity level cognitive error summaries. Our results show that MEMOR-E can generate stage aware, non diagnostic cognitive summaries that support personalized assistive interactions, while explainable AI mechanisms translate model outputs into transparent, human readable evidence to enable caregiver oversight and trustworthy human robot interaction. --- ### Residual Drift Dominates Contradiction in Multi-Turn Constraint Reasoning Authors: Sebastien Kawada Published: 2026-04-28 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.23940v1 How do multi-turn reasoning systems fail? The expected answer is logical contradiction, in which the system's maintained state becomes unsatisfiable. We show that the dominant mode is instead satisfiable drift, where the internal state stays consistent while the returned answer silently violates prior commitments. We build DRIFT-Bench (Decomposing Reasoning Into Failure Types), a solver-instrumented benchmark of 816 test problems across three constraint domains, and evaluate four methods on it across four open-weight models (8B-120B parameters). MUS-Repair, which feeds minimal unsatisfiable subsets back to the generator, is strongest in every setting (+1.8 to +15.0 pp over the best non-MUS baseline). But the central finding is what repair leaves behind. After structured feedback, models rarely contradict themselves. They forget. Residual errors are 98-100% satisfiable drift across all settings, while contradiction drops to near zero. Reliable multi-turn systems must separately validate that the returned answer respects the maintained state. Code is available at https://github.com/kaons-research/drift-bench. --- ### DRIVE: Modeling Skills at the Reasoning and Interaction Levels for Web Agents under Continual Learning Authors: Xirui Liu, Sihang Zhou, Yanning Hou, Rong Zhou, Haoyuan Chen, Maolin He, Siwei Wang, Hao Chen, Jian Huang Published: 2026-04-28 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.23939v1 Web agents require both high-level reasoning (for task decomposition) and low-level interactions (for page elements manipulation) to conduct different tasks. However, these knowledge types differ fundamentally: reasoning knowledge (e.g., booking a flight requires first searching for routes) is abstract and transferable across websites, while interaction knowledge (e.g., clicking the Search button at a specific coordinate on Site A) depends heavily on page-specific contexts. Existing methods store experiences uniformly. This creates a dilemma: abstract representations lose executability on concrete pages, while concrete representations fail to generalize across domains. This entanglement limits capability accumulation: on new websites, agents either fail to recognize reusable task logic due to surface-level differences or attempt infeasible actions from outdated page structures. To disentangle them, we propose DRIVE, a dual-level skill modeling framework separating historical experience into natural language reasoning skills, which capture transferable task logic, and programmatic interaction skills, grounding abstract actions to executable operations. A scene-aware coordination mechanism adaptively retrieves and invokes these dual-level skills based on task semantics. DRIVE also uses skill-level reflection to identify hierarchy-specific failure modes, enabling targeted skill library expansion and refinement. Experiments across five WebArena domains show DRIVE attains an average task success rate of 52.8%, exceeding the skill-free baseline by 7.3 percentage points. Further ablations show reasoning and interaction skills provide distinct, complementary benefits, supporting separation of transferable task logic from executable page-level operations. --- ### Authority Inversion in LLM-Mediated Ubiquitous Systems: When Models Trust Users Over Sensors Authors: Long Zhang, Zi-bo Qin, Wei-neng Chen Published: 2026-04-28 Categories: cs.AI, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2605.23938v1 Large language models (LLMs) increasingly fuse heterogeneous inputs in ubiquitous systems. Yet, how LLMs implicitly allocate authority when sensor measurements and user claims conflict remains unexamined, raising critical reliability concerns for deployments where physical sensing must retain priority. Unlike explicit traditional fusion, LLMs bury authority allocation within learned representations. We discover this allocation is severely format-dependent: numerical sensor data fails to integrate into answer-relevant model directions, allowing natural-language claims to dominate the final decision, a phenomenon we term \textbf{Authority Inversion}.To diagnose and mitigate this, we develop a geometric framework of context integration, introduce two computable audit metrics, specifically the Context Integration Ratio (CIR) and Authority Alignment Index (AAI), and propose Geometric Authority Calibration (GAC), an inference-time layer-level intervention to suppress misplaced user authority. Evaluating four models (4B to 35B parameters, three architectures) across four datasets totaling 576 conflict instances reveals extreme inversion: on numerical tasks, models exhibit near-zero sensor trust (AAI = -0.805, Cohen's d = -2.14), unaffected by model capacity. Validating our geometric framework, theory-guided causal injection flips 80.2\% of incorrect decisions (vs. <0.4\% for random controls). Practically, GAC improves HAR accuracy from 0 -- 1.6\% to 21.9 -- 27.5\%, outperforming prompting baselines. Ultimately, authority allocation in LLM-mediated systems must be explicitly audited and application-specifically configured rather than left implicit. --- ### Does Theory of Mind Improvement Really Benefit Human-AI Interactions? Empirical Findings from Interactive Evaluations Authors: Nanxu Gong, Zixin Chen, Haotian Li, Zishu Zhao, Jianxun Lian, Huamin Qu, Yanjie Fu, Xing Xie Published: 2026-04-28 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.15205v1 Improving the Theory of Mind (ToM) capability of Large Language Models (LLMs) is crucial for effective social interactions between these AI models and humans. However, the existing benchmarks often measure ToM capability improvement through story-reading, multiple-choice questions from a third-person perspective, while ignoring the first-person, dynamic, and open-ended nature of human-AI (HAI) interactions. To directly examine how ToM improvement techniques benefit HAI interactions, we first proposed the new paradigm of interactive ToM evaluation with both perspective and metric shifts. Next, following the paradigm, we conducted a systematic study of four representative ToM enhancement techniques using both four real-world datasets and a user study, covering both goal-oriented tasks (e.g., coding, math) and experience-oriented tasks (e.g., counseling). Our findings reveal that improvements on static benchmarks do not always translate to better performance in dynamic HAI interactions. This paper offers critical insights into ToM evaluation, showing the necessity of interaction-based assessments in developing next-generation, socially aware LLMs for HAI symbiosis. --- ### Block-Wise Differentiable Sinkhorn Attention: Tail-Refinement Gradients with a Gap-Aware Dustbin Bridge Authors: Dylan Forde Published: 2026-04-28 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2605.08123v3 We study long-context balanced entropic optimal transport (OT) attention on TPU hardware through a stopped-base, fixed-depth tail-refinement surrogate. After a stopped $T$-step Sinkhorn solve, we unroll a short refinement tail and differentiate that surrogate exactly. For the reported $R=2$ TPU path, the backward pass contains four staircase plan factors. We prove an exact one-reference-tile schedule: the $R=2$ score cotangent is a single reference plan tile times an explicit modifier field built from vector cotangents and dual differences. This yields block-wise cost $O((T+R)LW)$, $O(Ld)$ input storage, and $O(L)$ additional HBM usage for fixed head dimension $d$ and band width $W$ on the balanced fixed-support path. We also formalize the current \texttt{dustbin\_block} path as the same unit-target surrogate on an augmented support, so the adjoint schedule lifts to the single-active-dustbin path used in our TPU runs; this bridge is algebraic and does not claim a general KL-unbalanced or arbitrary-capacity gap model. We provide a local surrogate-bias bound, an a posteriori bias certificate, and a projective contraction certificate for strictly positive active blocks. On synthetic masked problems, the optimized kernel matches exact autodiff of the same centered surrogate to within $10^{-5}$--$10^{-10}$. On TPU v6e-8, a four-configuration Pfam screen completes end-to-end, and a promoted balanced $R=2$ run sustains roughly $8.5$ examples per second through a three-hour budget, reaching step $1437$. Held-out Pfam test shards improve reconstruction from $5.57$ to $2.05$ and sparse CE from $5.53$ to $5.30$ relative to step $0$, with CE logged diagnostically rather than optimized directly; target-barycenter alignment metrics do not materially improve, and a deterministic diagonal reference remains stronger on those metrics. --- ### Block-Wise Differentiable Sinkhorn Attention: Tail-Refinement Gradients with a Gap-Aware Dustbin Bridge Authors: Dylan Forde Published: 2026-04-28 Categories: cs.LG, cs.CL arXiv: https://arxiv.org/abs/2605.08123v2 We study long-context balanced entropic optimal transport (OT) attention on TPU hardware through a stopped-base, fixed-depth tail-refinement surrogate. After a stopped $T$-step Sinkhorn solve, we unroll a short refinement tail and differentiate that surrogate exactly. For the reported $R=2$ TPU path, the backward pass contains four staircase plan factors. We prove an exact one-reference-tile schedule: the $R=2$ score cotangent is a single reference plan tile times an explicit modifier field built from vector cotangents and dual differences. This yields block-wise cost $O((T+R)LW)$, $O(Ld)$ input storage, and $O(L)$ additional HBM usage for fixed head dimension $d$ and band width $W$ on the balanced fixed-support path. We also formalize the current \texttt{dustbin\_block} path as the same unit-target surrogate on an augmented support, so the adjoint schedule lifts to the single-active-dustbin path used in our TPU runs; this bridge is algebraic and does not claim a general KL-unbalanced or arbitrary-capacity gap model. We provide a local surrogate-bias bound, an a posteriori bias certificate, and a projective contraction certificate for strictly positive active blocks. On synthetic masked problems, the optimized kernel matches exact autodiff of the same centered surrogate to within $10^{-5}$--$10^{-10}$. On TPU v6e-8, a four-configuration Pfam screen completes end-to-end, and a promoted balanced $R=2$ run sustains roughly $8.5$ examples per second through a three-hour budget, reaching step $1437$. Held-out Pfam test shards improve reconstruction from $5.57$ to $2.05$ and sparse CE from $5.53$ to $5.30$ relative to step $0$, with CE logged diagnostically rather than optimized directly; target-barycenter alignment metrics do not materially improve, and a deterministic diagonal reference remains stronger on those metrics. --- ### Performance and Energy Trade-Off Analysis of Hierarchical Federated Learning for Plant Disease Classification Authors: Athanasios Papanikolaou, Athanasios Tziouvaras, Pavlos Stoikos, Apostolos Xenakis, Shameem A Puthiya Parambath, George Floros, Enrica Zereik, Ivan Petrovic, Fabio Bonsignorio Published: 2026-04-28 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2605.08121v1 Early detection of plant diseases is critical for improving crop productivity, while it also facilitates the foundations of precision agriculture. Recent advances in distributed deep learning have enabled plant disease classification models to be trained across geographically distributed agricultural sensing infrastructures. However, deploying such systems in large-scale Internet of Things (IoT) environments, introduces significant challenges related to computational cost, energy consumption, and system efficiency. In this paper, we present a design-space exploration of hierarchical federated learning architectures for plant disease classification, with a particular focus on the trade-offs between predictive performance and energy efficiency. We further introduce a power- and energy-aware optimization framework that enables the systematic evaluation and selection of model-aggregator configurations under varying deployment constraints. The hierarchical federated architecture organizes distributed clients through intermediate aggregation layers, reducing communication and computational overhead. We evaluate multiple convolutional neural network architectures, including EfficientNet-B0, ResNet-50, and MobileNetV3-Large, in combination with different federated aggregation strategies such as FedAvg, FedProx, and FedAvgM. Experimental results demonstrate that different model-aggregator combinations exhibit distinct performance-energy trade-offs. Consequently, we highlight configurations that achieve competitive diagnostic accuracy and significantly reduce system resource requirements. --- ### Feature Repulsion and Spectral Lock-in: An Empirical Study of Two-Layer Network Grokking Authors: Yongzhong Xu Published: 2026-04-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08119v1 Tian (2025) proves a repulsion theorem (Theorem 6) for the matrix $ B = (\widetilde{F}^\top \widetilde{F} + ηI)^{-1} $ during the interactive feature-learning stage of grokking: similar features have negative off-diagonal entries $ B_{j\ell} $, producing an effective repulsive force that drives them apart. However, the theorem does not specify when this mechanism becomes empirically observable, nor whether it leaves a measurable spectral signature in the parameter updates. We test this directly on Tian's modular addition setup ($ M = 71 $, $ K = 2048 $, MSE loss) and observe a clear structure-mechanism dissociation. The predicted sign rule holds robustly on the top-200 most-similar feature pairs across activations (empirical sign-match rising from 0.865 to 0.985 on $ σ= x^2 $ across 5 seeds, and saturating at 1.000 on $ σ= \operatorname{ReLU} $). However, the spectral signature in the parameter updates is strongly activation-dependent. With $ σ= x^2 $, a simple slope detector on the rolling eigengap $ σ_2 / σ_3 $ of $ ΔW $ fires in 15/15 grokking seeds at epoch 174 (IQR (173,174)) and in 0/15 non-grokking controls, with 229$ \times $ late-stage magnitude separation; the spectrum is rank-2. In contrast, with $ σ= \operatorname{ReLU} $, the detector never fires and the spectrum remains effectively rank-1. This dissociation aligns with Tian's Theorem 5 distinction between focused (power-law) and spreading (ReLU) memorization: while the sign structure of $ B $ depends only on $ \widetilde{F}^\top \widetilde{F} $, how feature repulsion translates into weight updates critically depends on the activation derivative $ σ' $. --- ### Modular Retrieval-Augmented Generalization for Human Action Recognition Authors: Peng Liao, Shangsong Liang, Lin Chen, Peijia Zheng Published: 2026-04-28 Categories: eess.SP, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.08117v1 Inertial Measurement Unit (IMU)-based Human Activity Recognition (HAR) aims to interpret and classify user behaviors from temporal motion signals. Recently, deep learning frameworks have advanced this task by learning and extracting discriminative spatiotemporal representations, significantly improving recognition performance. However, IMU-based HAR still faces several critical challenges, particularly limited training samples and static knowledge utilization, both of which severely hinder its large-scale deployment. In this paper, we introduce MoRA, the first Retrieval-Augmented Module specifically designed for motion series. It can be flexibly integrated into any existing HAR model, enhancing recognition performance while maintaining inference efficiency. To address issues such as information redundancy in retrieval results and rigid fusion strategies, we propose an uncertainty-adaptive fusion unit within MoRA. This unit leverages previous physical knowledge from IMU signals to dynamically adjust the fusion strategy between original outputs and retrieved information, enabling more robust recognition. Extensive experiments on ten real-world datasets demonstrate that MoRA significantly improves the performance of existing IMU-based HAR models, consistently delivering stable and effective gains. The source code of MoRA is available at: https://github.com/liavonpenn/mora. --- ### The Safety-Aware Denoiser for Text Diffusion Models Authors: Amman Yusuf, Zhejun Jiang, Mijung Park Published: 2026-04-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08116v3 Recent work on text diffusion models offers a promising alternative to autoregressive generation, but controlling their safety remains underexplored. Existing safety approaches are geared toward autoregressive models and typically rely on post-hoc filtering or inference-time interventions. These are inadequate for effectively addressing safety risks in text diffusion models. We propose the Safety-Aware Denoiser (SAD), a safety-guidance framework in text diffusion models. The SAD modifies the iterative denoising process such that the text sample at the final denoising step is steered toward provably safe regions of the text space. This inference-time method can integrate safety constraints into the denoiser, avoiding computationally expensive retraining of the underlying diffusion model and enabling flexible, lightweight safety guidance. We evaluate the safety of the generated text using the SAD, with respect to hazard taxonomy, memorization, and jailbreak. Experimental results show that SAD substantially reduces unsafe generations while preserving generation quality and fluency, outperforming existing methods. These results demonstrate that our safety guidance during denoising provides an effective and scalable mechanism for enforcing safety in text diffusion models. --- ### The Safety-Aware Denoiser for Text Diffusion Models Authors: Amman Yusuf, Zhejun Jiang, Mijung Park Published: 2026-04-28 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.08116v2 Recent work on text diffusion models offers a promising alternative to autoregressive generation, but controlling their safety remains underexplored. Existing safety approaches are geared toward autoregressive models and typically rely on post-hoc filtering or inference-time interventions. These are inadequate for effectively addressing safety risks in text diffusion models. We propose the Safety-Aware Denoiser (SAD), a safety-guidance framework in text diffusion models. The SAD modifies the iterative denoising process such that the text sample at the final denoising step is steered toward provably safe regions of the text space. This inference-time method can integrate safety constraints into the denoiser, avoiding computationally expensive retraining of the underlying diffusion model and enabling flexible, lightweight safety guidance. We evaluate the safety of the generated text using the SAD, with respect to hazard taxonomy, memorization, and jailbreak. Experimental results show that SAD substantially reduces unsafe generations while preserving generation quality, diversity, and fluency, outperforming existing methods. These results demonstrate that our safety guidance during denoising provides an effective and scalable mechanism for enforcing safety in text diffusion models. --- ### Robustness of Transformer-Based Fluence Map Prediction Under Clinically Realistic Perturbations Authors: Ujunwa Mgboh, Rafi Ibn Sultan, Joshua Kim, Kundan Thind, Dongxiao Zhu Published: 2026-04-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00904v1 Learning-based fluence map prediction offers a fast alternative to iterative inverse planning in intensity-modulated radiation therapy (IMRT), but its robustness under realistic distribution shifts remains unclear. We study a two-stage transformer pipeline that maps anatomy (CT and contours) to dose and then to beamlet fluence maps. We compare fluence-stage transformer backbones with hierarchical, global, and hybrid attention, trained with a physics-informed loss enforcing energy consistency. Robustness is evaluated under geometric perturbations, radiometric noise, reduced training data, and domain shifts using a prostate IMRT dataset, with additional evaluation of the dose stage on public datasets. Results show smooth degradation under moderate perturbations but sharp failures under severe rotations and noise. Hierarchical transformers (e.g., SwinUNETR) exhibit slower growth in upper-quartile energy error, indicating improved robustness. We further show that SSIM alone fails to capture clinically relevant errors, highlighting the need for physics-informed evaluation. --- ### A Light Weight Multi-Features-View Convolution Neural Network For Plant Disease Identification Authors: Muhammad Kaleem Ullah Khan Published: 2026-04-28 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.00903v1 Agriculture is a key sector of the economies of developing countries. It serves as a primary source of income and employment for rural populations. However, each year, a large portion of crops is wasted because of pests and diseases. Well-timed prediction of plant diseases is crucial to sustainable, high-quality agricultural production. Detection of plant diseases through conventional methods is both labour-intensive and time-consuming. Researchers have developed image classification based automated techniques for this purpose. Most accurate methods are based on deep convolutional neural networks, which are computationally intensive, with many layers and millions of trainable parameters. In resource-constrained settings, especially in rural areas, it is difficult to deploy deep convolutional neural network models for efficient plant disease identification. To address these issues, an efficient and light-weight Multi-View Convolutional Neural Network is proposed. These additional features aid the proposed model to identify the plant diseases accurately and efficiently with less number of parameters. The proposed model is tested on a benchmark Plantvillage dataset and achieves an improvement of $ 2.9\%$ in classification accuracy compared to the baseline convolutional neural network model, which was trained only on Red, Green, and Blue (RGB) plant images. Compared with state-of-the-art deep convolutional neural network models, the proposed model is less computationally expensive and achieves comparable accuracy for plant disease identification on the PlantVillage dataset. --- ### Validation of Whole-Slide Foundation Models for Image Retrieval in TCGA Data Authors: Tianhao Lei, Parsa Esmaeilkhani, Saghir Alfasly, Wataru Uegami, Judy C. Boughey, Matthew P. Goetz, Krishna R. Kalari, H. R. Tizhoosh Published: 2026-04-28 Categories: cs.CV, cs.IR arXiv: https://arxiv.org/abs/2605.00902v1 Foundation models are reshaping computational histopathology, yet their value for whole-slide image retrieval relative to strong patch-based and supervised aggregation baselines remains unclear. We benchmarked ten pipelines on 9,387 diagnostic slides spanning 17 organs and 60 diagnoses from The Cancer Genome Atlas (TCGA) using patient-level leave-one-patient-out evaluation. Methods included four pre-trained slide foundation models, a supervised attention-based multiple instance learning (ABMIL) aggregator on patch embeddings, and patch-level retrieval across five sampling densities. Performance varied more across organs and diagnoses than across architectures. Although the slide foundation model TITAN achieved the strongest overall results, its advantage was modest; ABMIL and patch-based methods reached comparable Top-1 and Top-3 accuracy, with no model consistently dominant. Morphologically distinctive entities approached ceiling performance, while rare, heterogeneous, and closely related subtypes remained challenging. Misclassifications aligned with organs exhibiting known inter-observer variability, suggesting an intrinsic ceiling for morphology-only retrieval. Performance was driven primarily by patch-level feature representations, with limited benefit from slide-level aggregation, indicating aggregation may be unnecessary in many settings. These findings argue against a universally optimal architecture and instead support organ-resolved benchmarking, diagnosis-aware or ensemble strategies, stronger feature representations, and multimodal retrieval frameworks. Notably, even the best model achieved only $\approx 68\% \pm 21\%$ retrieval accuracy on TCGA, and some subtypes showed $0\%$ accuracy across all methods, highlighting fundamental limitations of morphology-based representations and the need for substantial progress before reliable clinical deployment. --- ### RA-CMF: Region-Adaptive Conditional MeanFlow for CT Image Reconstruction Authors: Md Shifatul Ahsan Apurba, Md Selim, Jin Chen Published: 2026-04-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00901v2 The use of CT imaging is important for screening, diagnosis, therapy planning, and prognosis of lung cancers. Unfortunately, due to differences in imaging protocols and scanner models, CT images acquired by different means may show large differences in noise statistics, contrast, and texture. In this study, we develop a novel conditional MeanFlow pipeline for CT image reconstruction. We introduce a conditional MeanFlow network that models the reconstruction trajectory by predicting image-conditioned flow fields given intermediate image states. The image reconstruction network is trained with a MeanFlow consistency loss along with the image reconstruction loss. In order to provide a spatially adaptive refinement process, we integrate a regional reinforcement learning-driven policy network into our approach. The policy network receives information about the MeanFlow rollouts and provides predictions in terms of tile-wise refinement budgets, stopping criteria, and total budget allocation of refinement processes. Our policy network is trained through reinforcement learning in a policy gradient framework, where the goal of the training reward is to maximize reconstruction quality while minimizing unnecessary computations and avoiding instabilities. In this way, our approach combines conditional flow-based reconstruction with reinforcement learning-based spatial reconstruction control. Our results show high accuracy in the tumor ROI, with the average radiomic feature CCC being $0.93 \pm 0.09$, an average PSNR of $31.94 \pm 2.64$, and average SSIM of $0.97 \pm 0.03$. Moreover, there is an improvement in the overall quality of images, with an average PSNR of $34.23 \pm 1.71$ and average SSIM of $0.95 \pm 0.01$. --- ### RA-CMF: Region-Adaptive Conditional MeanFlow for CT Image Reconstruction Authors: Md Shifatul Ahsan Apurba, Md Selim, Jin Chen Published: 2026-04-28 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2605.00901v1 The use of CT imaging is important for screening, diagnosis, therapy planning, and prognosis of lung cancers. Unfortunately, due to differences in imaging protocols and scanner models, CT images acquired by different means may show large differences in noise statistics, contrast, and texture. In this study, we develop a novel conditional MeanFlow pipeline for CT image reconstruction. We introduce a conditional MeanFlow network that models the enhancement trajectory by predicting image-conditioned flow fields given intermediate image states. The image enhancement network is trained with a MeanFlow consistency loss along with the image reconstruction loss. In order to provide an adaptive refinement process in terms of spatial location of enhancements, we integrate a regional reinforcement learning-driven policy network into our approach. The policy network receives information about the MeanFlow rollouts and provides predictions in terms of tile-wise refinement budgets, stopping criteria, and total budget allocation of enhancement processes. Our policy network is trained through reinforcement learning in a policy gradient framework, where the goal of the training reward is to maximize improvement of enhancements while minimizing unnecessary computations and avoiding instabilities. In this way, our approach combines conditional flow-based enhancement with reinforcement learning-based spatial enhancement control. This allows our approach to focus more attention on enhancing difficult areas while stabilizing areas already showing sufficient quality. Our results show high accuracy in the tumor ROI, with the average radiomic feature CCC being 0.96, an average PSNR of 31.30 $\pm$ 4.16, and average SSIM of 0.94 $\pm$ 0.07. Moreover, there is an improvement in the overall quality of images, with an average PSNR of 34.23 $\pm$ 1.71 and average SSIM of 0.95 $\pm$ 0.01. --- ### LatentDiff: Scaling Semantic Dataset Comparison to Millions of Images Authors: James Flora, Kowshik Thopalli, Akshay R. Kulkarni, Weng-Keen Wong, Shusen Liu Published: 2026-04-28 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.00899v1 We present LatentDiff, a scalable framework for semantic dataset comparison that operates directly in the latent space of pretrained vision encoders. By combining sparse autoencoder-based divergence testing with density ratio estimation, LatentDiff identifies interpretable semantic differences between datasets at a fraction of the computational cost of caption-based alternatives. We also introduce Noisy-Diff, a benchmark capturing realistic sparse distribution shifts that cause existing methods to struggle. Experiments demonstrate that LatentDiff achieves superior accuracy while remaining robust to settings where an extremely small fraction of images (from 5% to <1% ) differ semantically. --- ### A Deep Learning Model for Battery State Prediction towards Intelligent Energy Management Authors: Athanasios Koukosias, Vasileios Tzanidakis, Sotiris Athanasiou, Kostas Kolomvatsos Published: 2026-04-28 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2605.00898v2 Accurate forecasting of battery health indicators, including remaining capacity and lifetime, is of paramount importance for ensuring the reliability, safety, and operational efficiency of applications such as electric vehicles and large scale energy storage infrastructures. The result of the forecasting can be adopted to build an advanced monitoring mechanism for continuous checking batteries' health status to assist in the efficient real-time management of numerous applications. This research investigates the development and implementation of a Deep Learning (DL) model for the prediction of the future state and performance of industrial electrochemical energy storage systems. To address this challenge, we propose a dedicated computational framework that integrates advanced neural network architectures with large-scale training datasets, enabling precise modeling of batteries degradation dynamics and operational trends. The proposed approach provides a decision support mechanism for the optimal management of batteries facilitating both predictive maintenance and the efficient allocation of energy resources. Our findings highlight the potential of DL-based predictive modeling to significantly contribute to the advancement of sustainable and intelligent energy management systems. --- ### SPAT: A Semantic Port-Aware Adaptive-Rate Transmission Protocol for Semantic Communication Authors: Yunhao Wang, Shuai Ma, Bin Shen, Shouhan Shi, Youlong Wu, Guangming Shi, Xiang Cheng Published: 2026-04-28 Categories: eess.SP, cs.CV, eess.IV arXiv: https://arxiv.org/abs/2605.00897v1 With the evolution of 6G, semantic communication has emerged as a promising paradigm by prioritizing the delivery of task-relevant meaning over strict bit-level correctness. However, existing transport mechanisms still rely on explicit port headers and bit-level validation, making them vulnerable to header corruption and the resulting packet loss. To address this issue, this paper proposes a Semantic Port-Aware Adaptive-Rate Transmission Protocol (SPAT) for semantic communication. The proposed framework jointly embeds source and destination port information into semantic representations, thereby reducing dependence on explicit port headers while enabling robust port-aware transmission. Furthermore, a differentiated semantic processing mechanism is developed for uplink and downlink scenarios, where port identification is introduced for uplink service recognition and destination-aware conditional gating is designed for downlink selective decoding. In addition, an adaptive-rate controller is incorporated to dynamically adjust the number of transmitted semantic channels according to channel conditions and feature importance, thereby improving both robustness and transmission efficiency. Experimental results on the AFHQ and ImageNet-10 datasets, together with real-world experimental measurements, demonstrate that SPAT consistently outperforms TCP, UDP, and SITP in reconstruction quality across different SNRs while maintaining low-latency transmission. --- ### When Less Is More: Simplicity Beats Complexity for Physics-Constrained InSAR Phase Unwrapping Authors: Prabhjot Singh, Manmeet Singh Published: 2026-04-28 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.00896v1 Operational phase unwrapping is the primary computational bottleneck in InSAR-based volcanic and seismic monitoring. We challenge the industry trend of adopting high-complexity computer vision architectures, such as attention mechanisms, without validating their suitability for physics-constrained geophysical regression. We present the first large-scale architectural ablation study on a global LiCSAR benchmark (20 frames, 39,724 patches, 651M pixels). Our results reveal a significant "complexity penalty": a vanilla U-Net (7.76M parameters) achieves $R^2=0.834$ and RMSE $= 1.01$ cm, outperforming 11.37M-parameter attention-based models by 34% in $R^2$ and 51% in RMSE. Power Spectral Density (PSD) analysis provides the physical justification: while attention excels at capturing sharp semantic edges in natural images, it injects unphysical high-frequency artifacts ($>0.3$ cycles/pixel) into geophysical fields, violating the fundamental smoothness constraints of elastic surface deformation. With a 2.92ms inference latency (a $2.5\times$ speedup), the vanilla U-Net is the only candidate to comfortably meet the sub-100ms requirement for operational early-warning systems. This work bridges the "publication-to-practice" gap by proving that convolutional locality outperforms modern complexity for smooth-field regression, advocating for physics-informed simplicity in ML4RS. Code available at https://github.com/prabhjotschugh/When-Less-is-More-InSAR-Phase-Unwrapping --- ### Transfer Learning for Tonal Noise Prediction in VRF Units Using Thermodynamic and Vibration Signals Authors: ZhiWei Su, Ding Wang, Yuan Guo, Yang Qiao, HongJun Cao Published: 2026-04-28 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2605.00895v1 The second-order harmonic (2f) component generated by twin-rotary compressor is a dominant low-frequency noise source of variable refrigerant flow (VRF) outdoor units, yet its amplitude fluctuates strongly with environmental thermal load and valve opening, making it difficult to assess accurately using conventional mechanism-based models. This paper proposes an unsupervised transfer learning method based on Domain-invariant Partial Least Squares (Di-PLS) to accurately predict 2f noise levels under new conditions using different signals. Prediction models utilizing thermodynamic signals and acceleration signals are constructed respectively, and the generalization performance of the proposed Di-PLS is systematically compared with traditional Partial Least Squares (PLS). Results demonstrate that Di-PLS significantly outperforms PLS by extracting cross-condition common features and minimizing the distribution discrepancy between the source and target domains. Specifically, the acceleration-based Di-PLS model achieves the best performance, maintaining prediction errors within 3 dB for all test cases. This superiority over thermodynamic-based models highlights a physical insight: while thermodynamic states drive dynamic changes, structural vibration possesses a stronger and more direct causal link to acoustic radiation. --- ### State-Dependent Lyapunov Analysis of Rank-1 Matrix Factorization Authors: Jaehong Moon Published: 2026-04-28 Categories: math.NA, cs.LG, math.DS, math.OC arXiv: https://arxiv.org/abs/2604.26993v3 We develop a state-dependent Lyapunov framework for gradient descent on rank-1 matrix factorization. A parameterized quadratic certificate $I(δ;\,\cdot)$ generates strictly nested sublevel sets. Their ordering assigns each point a state $δ$, while a boundary-inward property makes this state monotone along gradient-descent trajectories. Together with internal chain transitivity, this geometry identifies the limiting dynamics even when all relevant stationary points are unstable. For scalar and rank-1 factorization, it yields convergence to a global minimizer below the stability threshold and to a balanced period-$2$ orbit in a post-critical interval, for almost every initialization in an explicit region. We formalize the mechanism through structural and dynamical axioms. Within the origin-centered, exchange-symmetric quadratic class, the axioms determine the ordered level-set geometry uniquely up to a monotone relabeling of the state, recovering the scalar geometry identified by Liang--Mont{ú}far. Additional analytic and numerical examples suggest broader applicability. --- ### State-Dependent Lyapunov Analysis of Rank-1 Matrix Factorization Authors: Jaehong Moon Published: 2026-04-28 Categories: math.NA, cs.LG, math.OC arXiv: https://arxiv.org/abs/2604.26993v2 We study gradient descent for rank-1 matrix factorization through a state-dependent Lyapunov perspective. The central object is a parameterized quadratic certificate $I(δ;\,\cdot)$ whose boundary-inward property induces a monotone state parameter $δ_t$, thereby certifying that the trajectory is confined to a shrinking family of level sets. For certified initializations below the critical step size, this mechanism proves convergence to global minimizers. Above the critical step size, the same monotone-state mechanism instead leads to a balanced terminal regime; for a range of post-critical step sizes, the reduced dynamics exhibit period-2 behavior consistent with edge-of-stability phenomena. We further show that the scalar certificate is not an ad hoc algebraic construction: under structural axioms and a natural state-parameter normalization, it is uniquely determined by the monotonicity mechanism. Numerical experiments suggest that this state-dependent Lyapunov mechanism persists beyond the proved cases, including two-dimensional rank-1 approximation and quartic augmentations of scalar factorization. ---