# Research Papers — 2026-05 · bundle page 281 of 326 > Full-text markdown of papers #14001–14050 of 16266 published in 2026-05, newest first. AI-readable mirrors of the full 16266-paper 2026-05 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-05/page/281 [← Previous bundle](https://aipod.dev/papers/archive/2026-05/page/280.md) [Next bundle →](https://aipod.dev/papers/archive/2026-05/page/282.md) ### The Cost of Context: Mitigating Textual Bias in Multimodal Retrieval-Augmented Generation Authors: Hoin Jung, Xiaoqian Wang Published: 2026-05-07 Categories: cs.CL, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.05594v1 While Multimodal Large Language Models (MLLMs) are increasingly integrated with Retrieval-Augmented Generation (RAG) to mitigate hallucinations, the introduction of external documents can conceal severe failure modes at the instance level. We identify and formalize the phenomenon of recorruption, where the introduction of even perfectly accurate "oracle" context causes a capable model to abandon an initially correct prediction. Through a mechanistic diagnosis of internal attention matrices, we show that recorruption is driven by a two-fold attentional collapse: (1) visual blindness, characterized by the systemic suppression of visual attention mass ($M_{vis}$) and sharpness ($S_{vis}$), and (2) a structural positional bias that forces the model to prioritize boundary tokens over semantic relevance. Our analysis reveals an Illusion of Success, demonstrating that many seemingly correct RAG outcomes are merely positional coincidences where the model's textual copying bias happens to align with the ground-truth location. To address these vulnerabilities, we propose Bottleneck Attention Intervention for Recovery (BAIR), a parameter-free, inference-time framework that restores visual saliency and applies position-aware penalties to textual distractors. Across medical factuality, social fairness, and geospatial benchmarks, BAIR successfully restores multimodal grounding and improves diagnostic reliability without requiring model retraining or fine-tuning. --- ### Causal Probing for Internal Visual Representations in Multimodal Large Language Models Authors: Zehao Deng, Tianjie Ju, Zheng Wu, Liangbo He, Jun Lan, Huijia Zhu, Weiqiang Wang, Zhuosheng Zhang Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05593v1 Despite the remarkable success of Multimodal Large Language Models (MLLMs) across diverse tasks, the internal mechanisms governing how they encode and ground distinct visual concepts remain poorly understood. To bridge this gap, we propose a causal framework based on activation steering to actively probe and manipulate internal visual representations. Through systematic intervention across four visual concept categories, our results reveal a divergence in concept encoding: entities exhibit distinct localized memorization, whereas abstract concepts are globally distributed across the network. Critically, this divergence uncovers a mechanistic driver of scaling laws: increasing model depth is indispensable for encoding distributed and complex abstract concepts, whereas entity localization remains remarkably invariant to scale. Furthermore, reverse steering uncovers that blocking explicit output triggers a surge in latent activations, exposing a compensatory mechanism between perception and generation. Finally, extending our analysis to visual reasoning, we expose a disconnect between perception and reasoning although MLLMs successfully recognize geometric relations, they treat them merely as static visual features, failing to trigger the procedural execution necessary for abstract problem-solving. --- ### When Can Voting Help, Hurt, or Change Course? Exact Structure of Binary Test-Time Aggregation Authors: Yi Liu Published: 2026-05-07 Categories: cs.LG, cs.IT arXiv: https://arxiv.org/abs/2605.05592v1 Majority voting is one of the few black-box interventions that can improve a fixed stochastic predictor: repeated access can be cheaper than changing a high-capability model. Classical fixed-competence theory makes this intervention look monotone -- more votes help above the majority threshold and hurt below it. We show that this picture is fundamentally incomplete. Under the de Finetti representation for exchangeable repeated correctness, voting is governed by a latent distribution of per-example correctness probabilities. Even simple latent mixtures can generate sharply different voting curves, including nonmonotone behavior and, in an explicit construction, infinitely many trend changes. The full latent law determines the curve, but the curve does not determine the law. The exact object recovered by voting is a signed voting signature: at each binomial variance scale, it records excess latent mass above rather than below the majority threshold. Our main theorem proves that the complete odd-budget curve and this signature are equivalent: the curve increments are signed Hausdorff moments, and the full curve recovers the signature uniquely. This viewpoint explains shape phenomena, branch-symmetric nonidentifiability, realizability, variation, and endpoint rates. It also separates estimation regimes: direct per-example success-probability information targets the full signature, whereas fixed-depth grouped labels reveal only a finite prefix. --- ### In-Context Positive-Unlabeled Learning Authors: Siyan Liu, Yi Chang, Manli Cheng, Qinglong Tian, Pengfei Li Published: 2026-05-07 Categories: stat.ML, cs.LG, stat.CO arXiv: https://arxiv.org/abs/2605.05591v1 Positive-unlabeled (PU) learning addresses binary classification when only a set of labeled positives is available alongside a pool of unlabeled samples drawn from a mixture of positives and negatives. Existing PU methods typically require dataset-specific training or iterative optimization, which limits their applicability when many tasks must be solved quickly or with little tuning. We introduce PUICL, a pretrained transformer that solves PU classification entirely through in-context learning. PUICL is pretrained on synthetic PU datasets generated from randomly instantiated structural causal models, exposing it to a wide range of feature-label relationships and class-prior configurations. At inference time, PUICL receives the labeled positives and the unlabeled samples as a single input and returns class probabilities for the unlabeled rows in one forward pass, with no gradient updates or per-task fitting. On 20 semi-synthetic PU benchmarks derived from the UCI Machine Learning Repository, OpenML, and scikit-learn, PUICL outperforms four standard PU learning baselines in average AUC and accuracy, and is competitive on F1-score. These results show that the in-context learning paradigm extends naturally beyond fully supervised tabular prediction to the semi-supervised PU setting. --- ### Uncertainty-Guided Edge Learning for Deep Image Regression in Remote Sensing Authors: Anh Vu Nguyen, Dino Sejdinovic, Tat-Jun Chin Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.05590v1 Edge learning refers to training machine learning models deployed on edge platforms, typically using new data accumulated onboard. The computational limitations on edge devices affect not only model optimisation, but also calculation of the predictive uncertainty of the current model on the unlabelled data, which is vital for informing model updating. In this paper, we investigate edge learning in the context of performing deep image regression on a remote sensing satellite, where a deep network is executed by an onboard computer to regress a scalar $y$ from an input image, e.g., $y$ is the percentage of pixels indicating cloud coverage or land use. We propose an uncertainty-guided edge learning (UGEL) algorithm that can accurately prioritise the data to speed up training convergence of the on-board regression model. Underpinning UGEL is the calculation of predictive uncertainty based on deep beta regression, where a deep network is used to estimate the parameters of a beta distribution for which the target $y$ for an input image has a high likelihood. Compared to established methods for uncertainty estimation that are either too costly on edge devices (e.g., require many forward passes per sample) or make strict assumptions on the predictive distribution (e.g., Gaussian), deep beta regression is computable in a single forward pass and allows more general predictive distributions. Results show that UGEL delivers faster-converging edge learning than active or semi-supervised learning. Code and models are publicly available at https://github.com/anh-vunguyen/UGEL. --- ### AeroJEPA: Learning Semantic Latent Representations for Scalable 3D Aerodynamic Field Modeling Authors: Francisco Giral, Abhijeet Vishwasrao, Andrea Arroyo Ramo, Mahmoud Golestanian, Federica Tonti, Adrian Lozano-Duran, Steven L. Brunton, Sergio Hoyas, Hector Gomez, Soledad Le Clainche, Ricardo Vinuesa Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.05586v1 Aerodynamic surrogate models are increasingly used to replace repeated high-fidelity CFD evaluations in many-query design settings, but current approaches still face two important limitations: they often scale poorly to the very large fields arising in realistic 3D aerodynamics, and they rarely produce latent representations that are directly useful for analysis and design. We introduce AeroJEPA, a Joint-Embedding Predictive Architecture for aerodynamic field modeling that addresses both issues. Rather than predicting the full flow field directly from geometry, AeroJEPA predicts a target latent representation of the flow from a context latent representation of the geometry and operating conditions, and optionally reconstructs the field through a continuous implicit decoder. This formulation decouples latent prediction from field resolution while encouraging the latent space to organize semantically. We evaluate AeroJEPA on two complementary datasets: HiLiftAeroML, which stresses the method in a high-fidelity regime with extremely large boundary-layer fields, and SuperWing, which tests large-scale generalization and latent-space optimization over a broad family of transonic wings. Across these benchmarks, AeroJEPA is competitive as a continuous surrogate for aerodynamic fields, scales naturally to high-resolution outputs, and learns context and predicted latents that encode geometry and aerodynamic quantities not used directly as supervision. We further show that the resulting latent space supports controlled interpolation, linear probing, concept-vector arithmetic, and a constrained design latent-optimization experiment. These results suggest that predictive latent learning is a promising direction for scalable and design-meaningful aerodynamic surrogate modeling. --- ### Belief Memory: Agent Memory Under Partial Observability Authors: Junfeng Liao, Qizhou Wang, Jianing Zhu, Bo Du, Rui Yan, Xiuying Chen Published: 2026-05-07 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.05583v2 LLM agents that operate over long context depend on external memory to accumulate knowledge over time. However, existing methods typically store each observation as a single deterministic conclusion (e.g., inferring "API~X failed" from temporary errors), even though such observations are inherently partial and potentially ambiguous. By committing to one conclusion and discarding uncertainty, these methods introduce self-reinforcing error: the agent acts on the stored conclusion, never revisits alternatives, and reinforces the conclusion over time. To address this issue, we propose BeliefMem, which shifts the memory paradigm from committing to a single conclusion per observation to retaining multiple candidate conclusions with their probabilities. Concretely, BeliefMem stores the candidate conclusions as separate memory entries, each carrying a probability that is updated via Noisy-OR rules as new observations arrive. At retrieval, all candidates surface together with their probabilities, keeping alternatives visible to the agent. Since each conclusion in memory retains its probability, BeliefMem preserves the uncertainty that the deterministic paradigm discards, enabling the agent to act with high confidence on well-evidenced knowledge while retaining the capacity to update its confidence when new evidence arrives. Empirical evaluations on LoCoMo and ALFWorld benchmarks show that, even with limited data, BeliefMem achieves the best average performance, remarkably outperforming well-known baselines. More broadly, such probabilistic memory produces substantial gains and explores a new direction for agent memory in partially observable environments. --- ### A Scalable Digital Twin Framework for Energy Optimization in Data Centers Authors: Raphael Hendrigo de Souza Gonçalves, Wendel Marcos dos Santos Published: 2026-05-07 Categories: cs.DC, cs.LG arXiv: https://arxiv.org/abs/2605.05581v1 This study proposes a scalable Digital Twin framework for energy optimization in data centers.The framework integrates IoT-based data acquisition, cloud computing, and machine learning techniques to enable real-time monitoring, forecasting, and intelligent energy management. A controlled small-scale data center environment was developed to monitor variables such as power consumption, temperature, and computational workload. Long Short-Term Memory (LSTM) models were employed to predict energy demand and support operational decision-making. Experimental results demonstrated improvements in energy efficiency, including reductions in power consumption and enhancements in Power Usage Effectiveness (PUE). Despite being evaluated in a constrained environment, the proposed framework demonstrates strong potential as a scalable and cost-effective solution for sustainable data center management. --- ### AlphaCrafter: Harnessing Multi-Agent Workflows for Cross-Sectional Quantitative Trading Authors: Yishuo Yuan, Jiayi Sheng, Sirui Zeng, Jiaqi Wang, Jiaheng Liu Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05580v2 Quantitative trading agents have demonstrated substantial promise in automating factor discovery, signal aggregation, and portfolio execution. However, existing agent-based trading systems predominantly rely on loosely specified natural-language workflows, leading to opaque reasoning processes, inconsistent behaviors across foundation models, and limited controllability and verifiability, all of which introduce significant risks in financial decision-making. To address these limitations, we propose AlphaCrafter, a multi-agent framework built upon a structured agent harness. Instead of treating agent behavior as unconstrained prompt execution, AlphaCrafter encapsulates each agent within programmable policy specifications that integrate procedural workflows, execution constraints, and explicit verification mechanisms. This harness-driven design transforms the entire trading pipeline into a sequence of well-defined, reproducible, and auditable decision processes with explicit execution semantics. Extensive experiments on the CSI 300 and S&P 500 benchmarks demonstrate that AlphaCrafter consistently achieves superior risk-adjusted returns while exhibiting substantially lower cross-model and cross-trial variance. These results suggest that harness-based agent design provides a practical foundation for building more reliable, controllable, and robust multi-agent systems for quantitative trading. --- ### Accelerating LMO-Based Optimization via Implicit Gradient Transport Authors: Won-Jun Jang, Si-Hyeon Lee Published: 2026-05-07 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2605.05577v1 Recent optimizers such as Lion and Muon have demonstrated strong empirical performance by normalizing gradient momentum via linear minimization oracles (LMOs). While variance reduction has been explored to accelerate LMO-based methods, it typically incurs substantial computational overhead due to additional gradient evaluations. At the same time, the theoretical understanding of LMO-based methods remains fragmented across unconstrained and constrained formulations. Motivated by these limitations, we propose \emph{LMO-IGT}, a new class of stochastic LMO-based methods leveraging implicit gradient transport (IGT). We further introduce a unified framework for stochastic LMO-based optimization together with a new stationarity measure, the \emph{regularized support function} (RSF), which bridges gradient-norm and Frank--Wolfe-gap notions within a common framework. By evaluating stochastic gradients at transported points, LMO-IGT accelerates convergence while retaining the single-gradient-per-iteration structure of standard stochastic LMO. Our analysis establishes that stochastic LMO achieves an iteration complexity of $\mathcal{O}(\varepsilon^{-4})$, variance-reduced LMO achieves $\mathcal{O}(\varepsilon^{-3})$ at the cost of additional gradient evaluations, and LMO-IGT achieves $\mathcal{O}(\varepsilon^{-3.5})$ using only a single stochastic gradient per iteration. Empirically, LMO-IGT consistently improves over stochastic LMO counterparts with negligible overhead. Among its instantiations, Muon-IGT achieves the strongest overall performance across evaluated settings, demonstrating that IGT provides an effective and practical acceleration mechanism for modern LMO-based optimization. --- ### AstroAlertBench: Evaluating the Accuracy, Reasoning, and Honesty of Multimodal LLMs in Astronomical Classification Authors: Claire Chen, Jiabao Sean Xiao, Shuze Daniel Liu, Facundo Perez Paolino, Luke Handley, Theophile Jegou du Laz, Ricky Nilsson, Alice Zou, Matthew Graham, Ashish Mahabal Published: 2026-05-07 Categories: astro-ph.IM, cs.AI arXiv: https://arxiv.org/abs/2605.05573v1 Modern astronomical observatories generate a massive volume of multimodal data, creating a critical bottleneck for expert human review. While multimodal large language models (LLMs) have shown promise in interpreting complex visual and textual inputs, their ability to perform specialized scientific classification while providing interpretable reasoning remains understudied. We introduce AstroAlertBench, a comprehensive multimodal benchmark designed to evaluate LLM performance in astronomical event review along a three-stage logical chain: metadata grounding, scientific reasoning, and hierarchical classification over five categories. We use a pilot sample of 1,500 real-world alerts from the Zwicky Transient Facility (ZTF), a wide-field survey that scans the northern sky to detect transient astronomical events. On this dataset, we benchmark 13 frontier closed-source and open-weight LLMs that support visual input. Our results reveal that high accuracy does not always align with model ``honesty,'' defined as the ability to self-evaluate its reasoning, which affects its reliability as a real-world assistant. We further initialize a human-in-the-loop evaluation protocol as a precursor to future community-scale participation. Together, AstroAlertBench provides a framework for developing calibrated and interpretable astronomical assistants. --- ### Text-to-CAD Retrieval: a Strong Baseline Authors: Honghu Pan, Zibo Du, Daxiang Liu, Chengliang Liu, Xiaoling Luo Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.05572v1 Text-based retrieval of Computer-Aided Design (CAD) models is a critical yet underexplored task for the reuse of legacy industrial designs. Existing CAD repositories are typically searched using filenames or directories, which limits the efficiency, scalability, and accuracy of design retrieval. In this paper, we formally introduce text-to-CAD retrieval as a new cross-modal retrieval task, aiming to retrieve semantically relevant CAD models from large-scale databases given natural language queries. Leveraging paired text-CAD annotations from the Text2CAD dataset, we establish a practical benchmark for this task. To achieve text-based retrieval, we propose a unified framework that learns multi-modal CAD embeddings from both procedural sequences and geometric point clouds. Specifically, a sequence encoder captures the construction logic of CAD models, while a point encoder extracts explicit geometric features. A text encoder is used to learn semantic representations of textual queries. During training, we introduce a novel feature decoder that reconstructs masked sequence features via cross-attention with text and point features, encouraging implicit multi-modal alignment. At inference time, we remove this auxiliary decoder to enable efficient retrieval using concatenated sequence-point features. Our framework serves as a strong baseline for text-to-CAD retrieval and lays the foundation for downstream CAD generation paradigms, such as retrieval-augmented generation. The source code will be released. --- ### Stability of the Monge Map in Semi-Dual Optimal Transport Authors: Anton Selitskiy, David Millard Published: 2026-05-07 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2605.05569v3 This paper shows that the semi-dual formulation of the optimal transport problem has a degenerate saddle-point structure, and that its numerical solution is equivalent to solving a constrained optimization problem. We derive necessary and sufficient conditions for the convergence of Monge maps without requiring optimality of the dual potential. This analysis helps explain why, in practice, numerical algorithms often require more iterations to update the transport map than the potential. --- ### Relaxed Sparsest-Permutation Formulation for Causal Discovery at Scale Authors: Sunmin Oh, Sang-Yun Oh, Gunwoong Park Published: 2026-05-07 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2605.05568v1 Despite the growing availability of large datasets, causal structure learning remains computationally prohibitive at scale. We revisit sparsest-permutation learning for linear structural equation models and show that exact Cholesky factorization is unnecessary for structure recovery. This observation motivates a support-level relaxation that searches for sparse triangular factors over a precision-support screening graph. The relaxed formulation can be efficiently evaluated via masked zero-fill incomplete Cholesky factorization, enabling scalable comparison of candidate orderings. At the population level, we establish soundness for Markov equivalence class (MEC) recovery under no-cancellation and sparsest Markov representation assumptions, as well as robustness to ordering misspecification. Motivated by these guarantees, we introduce SCOPE, a sparse-Cholesky pipeline that provides a scalable implementation of the relaxed formulation. Experiments on synthetic and real datasets demonstrate that SCOPE matches the MEC recovery accuracy of substantially slower baselines, while achieving significantly reduced runtime and scaling to 10k variables. --- ### Locality-aware Private Class Identification for Domain Adaptation with Extreme Label Shift Authors: Chuan-Xian Ren, Cheng-Jun Guo, Hong Yan Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05567v1 Domain adaptation aims to transfer knowledge from a labeled source domain to an unlabeled target domain with different distributions. In real-world scenarios, the label spaces of the two domains often have an inclusion relationship, where some classes exist only in one domain but not the other. These non-overlapping classes are referred to as private classes. Identifying private class samples and mitigating their adverse effects is critical in the literature. Existing methods rely on the assumption that shifts in private classes are large enough to be considered outliers. However, the variance within a single shared class can be significantly larger than the difference between a private class and another shared class, challenging this assumption. Consequently, private classes substantially increase the difficulty of cross-domain classification. To address these issues, based on local transportation and metric properties of optimal transport (OT), a locality-aware private class identification approach is proposed in the form of a score function on transport mass. The effectiveness of the proposed approach is theoretically proven, highlighting the score function's strong ability to distinguish between shared and private class samples. Building on this, we introduce a reliable OT-based method (ReOT) for domain adaptation under severe label shift. ReOT minimizes classification risk while learning the separated cluster structure between the identified shared classes and private classes, effectively avoiding mismatch between shared-private sample pairs, thus ensuring that important knowledge is reliably transported intra-class to mitigate class-conditional discrepancy. Furthermore, a generalization upper bound of the target risk is provided for extreme label shift scenarios, which can be minimized by ReOT. Extensive experiments on benchmarks validate the effectiveness of ReOT. --- ### Nonsense Helps: Prompt Space Perturbation Broadens Reasoning Exploration Authors: Langlin Huang, Chengsong Huang, Jinyuan Li, Donghong Cai, Yuyi Yang, Jiaxin Huang Published: 2026-05-07 Categories: cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2605.05566v1 Reinforcement learning with verifiable rewards, particularly Group Relative Policy Optimization (GRPO), has significantly advanced the reasoning capabilities of Large Language Models (LLMs). However, in complex tasks, GRPO frequently suffers from the ``zero-advantage problem'': when all sampled rollouts for a query fail, the relative advantage collapses to zero. Consequently, the model loses effective training signals for these questions, wasting the training data and computational budget. While simply increasing the sampling budget for these questions is a common remedy, the static sampling policy inherently constrains reasoning exploration, limiting the success rate. In this paper, we propose Lorem Perturbation for Exploration (LoPE), a simple yet effective training framework to break this exploration bottleneck. We posit that task-irrelevant prompt-space perturbations can shift the model's output distribution enough to unlock orthogonal reasoning pathways for hard questions. Specifically, LoPE prepends sequences stochastically assembled from Lorem Ipsum vocabulary (a pseudo-Latin placeholder text) to the prompts before resampling. Experiments across 1.7B, 4B, and 7B models demonstrate that LoPE significantly outperforms resampling with the original prompts. Further analysis reveals that other Latin-based random sequences with low perplexity are also effective perturbations. Our results establish LoPE as a strong baseline for broadening exploration in LLM reinforcement learning. --- ### BitCal-TTS: Bit-Calibrated Test-Time Scaling for Quantized Reasoning Models Authors: Sai Babu Patarlapalli, Surya Teja Avvaru Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05561v1 Post-training quantization makes large reasoning models practical under tight memory and latency budgets, but it can distort the online signals that drive adaptive test-time compute allocation. Under a fixed cap on the number of newly generated tokens, miscalibrated confidence can lead to harmful early halting: the model may surface a plausible final line while the underlying reasoning is still wrong, or the controller may stop before the trace has stabilized. We study this interaction for greedy 4-bit inference and propose BitCal-TTS, a lightweight runtime controller that combines (i) inexpensive online proxies for token-level uncertainty and reasoning-trace stability, (ii) a bit-conditioned confidence rescaling that is conservative at low nominal precision, and (iii) a bit-aware post-marker confirmation horizon designed for GSM8K-style structured outputs. The method requires no fine-tuning of the base model and integrates with standard Hugging Face 4-bit inference using forward hooks for logits and last-layer hidden states. On small evaluation shards of GSM8K with Qwen2.5 Instruct models, BitCal-TTS improves exact-match accuracy over a non-bit-aware adaptive baseline at the 7B and 14B scales while preserving substantial token savings relative to fixed-budget decoding. At a token cap of B=512, on the evaluation shards we report (N=54 for 7B and N=35 for 14B; not the full GSM8K test set), accuracy gains are +3.7 points (7B) and +2.8 points (14B), with the premature-stop rate falling from 14.8% to 11.1% on 7B and from 17.1% to 11.4% on 14B. We report Wilson 95% confidence intervals throughout and explicitly discuss the limited statistical power of the partial-shard comparisons. We release code and figure-generation scripts to support full reproduction. --- ### Who Prices Cognitive Labor in the Age of Agents? Compute-Anchored Wages Authors: Siqi Zhu Published: 2026-05-07 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2605.05558v2 A natural intuition about the economics of AI agents is that, because agents can be replicated at very low marginal cost, agent labor may be supplied highly elastically, placing downward pressure on cognitive-labor wages when it closely substitutes for human labor. We argue this framing is wrong in mechanism but partially correct in conclusion, and that the correction matters for both theory and policy. \textbf{Agents are not labor; they are a production technology that converts compute capital $K_c$ into effective units of cognitive labor $L_A$.} Once this is recognized, the elastic-supply margin that anchors the equilibrium wage migrates from the labor market to the compute capital market. Building on the classic factor-pricing framework \citep{mankiw2020}, we derive a \emph{Compute-Anchored Wage} (CAW) bound stating that, on tasks where human and agent-produced cognitive labor are substitutes, the competitive human wage is bounded above by $λ\cdot k \cdot r_c$, where $r_c$ is the rental rate of compute capital, $k$ is the compute intensity of one effective agent-produced cognitive labor unit, and $λ$ is the relative human-to-agent productivity. We generalize the result through constant elasticity of substitution (CES) aggregation, separate substitutable from complementary tasks, and discuss factor-share consequences. The conclusion is concise: \emph{the price-setter for cognitive labor is no longer the labor market.} --- ### An extremely coarse feedback signal is sufficient for learning human-aligned visual representations Authors: Yash Mehta, Michael F. Bonner Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.05556v1 Artificial neural networks trained on visual tasks develop internal representations resembling those of the primate visual system, a discovery that has guided a decade of computational neuroscience. Research on building brain-aligned models has progressively embraced finer-grained supervisory signals, from object classification to contrastive self-supervised objectives that maximize distinctions among individual images, yet the role of supervisory signal granularity on brain alignment remains largely unexamined. Here we systematically investigate how the coarseness of a learning signal shapes representational alignment with human vision. We parametrically vary the level of signal granularity using a data-driven approach that partitions a set of training images into varied numbers of categories (2, 4, 8, 16, ..., 64) via PCA-based splits of pretrained embeddings. We train hundreds of neural networks across convolutional and transformer architectures on these coarse classification tasks and compare their representations to macaque electrophysiology recordings and human fMRI responses. We find that networks trained to distinguish as few as 8 broad categories learn representations that match or exceed the neural alignment of models distinguishing 1,000-classes. Even more strikingly, these coarsely trained networks align more closely with human perceptual similarity judgments than all other models evaluated, including networks trained with fine-grained supervision or self-supervision as well as leading large-scale vision models. These results demonstrate that human-like visual representations emerge from remarkably coarse feedback, reframing what learning signals vision may require and opening a path toward building AI systems that are more aligned with human perception. --- ### FedeKD: Energy-Based Gating for Robust Federated Knowledge Distillation under Heterogeneous Settings Authors: Quang-Huy Nguyen, Jiaqi Wang, Wei-shinn Ku Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.05553v1 Federated learning (FL) operates in heterogeneous environments, where variations in data distributions and asymmetric model design often result in negative transfer. While federated knowledge distillation (FKD) avoids direct model parameter sharing, existing methods typically rely on public datasets or assume that transferred knowledge is uniformly reliable, which limits their robustness in practice. This paper presents FedeKD, a reliability-aware FKD framework that makes sample-wise trust estimation an explicit component of knowledge transfer, without relying on additional public data. Each client maintains a high-capacity private model for local learning and a lightweight shared proxy model for cross-client knowledge exchange. During training, proxy models are aggregated on the server to form a global proxy, which is then used to guide updates of the private models. At the core of FedeKD is an energy-based gating mechanism that converts task-specific private-proxy disagreement into sample-wise trust weights for backward distillation. This mechanism enables sample-wise weighting of knowledge transfer, where the proxy model contributes more to reliable samples while down-weighting unreliable ones. Extensive experiments on six real-world datasets demonstrate that FedeKD significantly reduces negative transfer under heterogeneous settings while maintaining strong predictive performance. --- ### A Novel Graph-Regulated Disentangling Mamba Model with Sparse Tokens for Enhanced Tree Species Classification from MODIS Time Series Authors: Motasem Alkayid, Zhengsen Xu, Saeid Taleghanidoozdoozan, Yimin Zhu, Megan Greenwood, Quinn Ledingham, Zack Dewis, Mabel Heffring, Naser El-Sheimy, Lincoln Linlin Xu Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.05549v1 Although tree species classification from Moderate Resolution Imaging Spectroradiometer (MODIS) time series data is critical for supporting various environmental applications, it is a challenging task due to several key difficulties: the subtle signature differences among tree species, strong spatial-spectral-temporal information coupling, and the difficulty of modeling large-scale topological context information. To better address these challenges, this paper presents a novel Graph-regulated Disentangled Sparse Mamba model (GDS-Mamba) for enhanced tree species classification, with the following contributions. (1) First, to improve large-scale context modeling, we design a mini-batch graph-regulated approach that explicitly explores topological correlation effects among input images. (2) Second, to disentangle the high-dimensional spatial-spectral-temporal information coupling for improved feature extraction, we propose a novel disentangling Mamba architecture tailored for capturing independent spatial patterns, spectral signatures, and temporal phenology behaviors in MODIS time series. (3) Third, to improve efficiency and subtle feature learning, we design novel sparse token approaches that adaptively learn the optimum subset of tokens to better address the correlation decay problem that bottlenecks standard Mamba models. Extensive experiments using large-scale annual MOD13Q1 data across two Canadian provinces (i.e., Alberta and Saskatchewan) achieved an overall accuracy of 93.94\% in Alberta and 80.19\% in cross-provincial evaluations, outperforming twelve state-of-the-art classification models. --- ### Characterizing Brazilian Atlantic Forest Restoration Outcomes with Geospatial AlphaEarth Embeddings Authors: Alice Heiman Published: 2026-05-07 Categories: cs.CV arXiv: https://arxiv.org/abs/2605.05547v2 The Atlantic Forest in Brazil is a critical biodiversity hotspot, yet less than 12-15% of its original cover remains. Although monitoring forest restoration on a large scale is essential, traditional methods are limited by the impracticality of on-the-ground reporting on such a scale and by the saturation of remote-sensing indices such as NDVI. Furthermore, reforestation is a gradual process as opposed to the rapid spectral changes caused by deforestation. In this study, we examine 1,729 restoration sites in São Paulo, using satellite embeddings from the AlphaEarth Foundation's model to evaluate their effectiveness in characterising early restoration success. We introduce the concept of a 'Reference Trajectory Embedding', defining a metric of restoration success based on cosine similarity to reference sites of mature secondary forest. We observe distinct clusters in embedding space according to different land use and land cover (LULC) types, and we can identify sites with clear change vectors. However, the signal can be noisy, and embeddings may require further fine-tuning to capture and predict site metadata beyond LULC. --- ### SPARK: Self-Play with Asymmetric Reward from Knowledge Graphs Authors: Hyobin Park, Taeseop Kim, Dong-Geol Choi Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05546v1 Self-play reinforcement learning has shown strong performance in domains with formally verifiable structure, such as mathematics and coding, where both problem generation and reward computation can be grounded in explicit rules. Extending this paradigm to scientific literature is more challenging: the relationships among multi-modal elements within and across documents are rarely made explicit in text, which makes automatic generation of relational reasoning questions difficult and weakens the reliability of reward signals. We propose SPARK (Self-Play with Asymmetric Reward from Knowledge Graphs), a framework that automatically constructs a unified knowledge graph (KG) from multi-document scientific literature and uses it as the structural basis for self-play. KG paths over multimodal nodes serve as a source for generating relational reasoning questions, and structured facts stored in the KG provide a basis for verifiable reward computation. A single small vision-language model (sVLM) alternates between Proposer and Solver roles under information asymmetry against a fixed KG, a design that we believe can be naturally extended toward online adaptation in future work. We evaluate SPARK on public benchmarks and a self-constructed cross-document multi-hop QA dataset. Results show that SPARK consistently outperforms flat-corpus-based self-play baselines, and the performance gap widens as hop count increases, suggesting that KG-structure grounding contributes to relational multi-hop reasoning beyond what unstructured corpus grounding can provide. --- ### Adaptive Q-Chunking for Offline-to-Online Reinforcement Learning Authors: Nandiraju Gireesh, Yuanliang Ju, He Wang Published: 2026-05-07 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2605.05544v1 Offline-to-online reinforcement learning with action chunking eliminates multi-step off-policy bias and enables temporally coherent exploration, but all existing methods use a fixed chunk size across every state. This is suboptimal: near contact events the agent needs short chunks for reactive control, while during free-space motion long chunks provide better credit assignment. The natural solution is to train critics for several chunk sizes and select the best one at each state, but naive comparison of learned critic values systematically collapses to the shortest chunk due to discount-scale mismatch, and degrades to noise in low-value states. We propose Adaptive Q-Chunking (AQC), which resolves both failures by comparing the advantage of each chunk size relative to a per-horizon baseline, normalized by the discount factor. This criterion converts biased wrong answers into unbiased near-random choices when no genuine signal exists, and becomes discriminative when a particular scale enables better planning. We prove theoretical bounds on the advantage selector's noise immunity and on the value dominance of adaptive chunking over any fixed chunk size. We demonstrate that AQC achieves state-of-the-art offline and online success rates on OGBench and Robomimic, and can be applied to enhance the performance of large-scale VLA models that predict action sequences, significantly boosting performance on RoboCasa-GR1 tasks. --- ### Autoregressive One-Step Generative Modeling for Dynamical System Forecasting Authors: Tianyue Yang, Xiao Xue Published: 2026-05-07 Categories: cs.LG, physics.flu-dyn arXiv: https://arxiv.org/abs/2605.05540v2 Fast surrogate modeling for high-dimensional physical dynamics requires more than low short-term error: useful models must roll out efficiently while preserving the statistical structure of long trajectories. Neural operators provide inexpensive autoregressive forecasts but can drift in turbulent regimes, whereas rolling diffusion and latent generative surrogates can represent stochastic transitions at the cost of multi-step denoising, noise-schedule design, or auxiliary compression models. We propose MeanFlow Long-term Invariant Spatiotemporal Consistency Autoregressive Models (MeLISA), a latent-free autoregressive generative surrogate built on pixel-space MeanFlow. MeLISA defines a blockwise stochastic transition kernel that generates each forecast block with a single model evaluation, avoiding latent encoders and iterative diffusion solvers at inference time. To stabilize long-horizon rollouts, MeLISA combines a Window-Consistency MeanFlow objective that learns conditional spatiotemporal generation from partially observed temporal windows with a Time Increment Consistency loss that constrains multi-lag finite increments and targets temporal-correlation structure. We evaluate MeLISA with compact UNet and scalable DiT backbones on two high-resolution benchmarks, extended 2D Kolmogorov flow at $256 \times 256$ and turbulent channel-flow slice at $192 \times 192$. MeLISA outperforms neural-operator baselines on short-term forecasting accuracy and long-horizon statistical metrics, including energy spectra, turbulent kinetic energy, and mixing-rate-related dynamics, while achieving inference speeds comparable to, and in some cases faster than, neural operators. To our knowledge, this is the first method for high-resolution one-step generation for physical dynamical systems with performance comparable to state-of-the-art deterministic surrogates. --- ### AgenticRAG: Agentic Retrieval for Enterprise Knowledge Bases Authors: Susheel Suresh, Hazel Mak, Shangpo Chou, Fred Kroon, Sahil Bhatnagar Published: 2026-05-07 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2605.05538v1 We present AgenticRAG, a practical agentic harness for retrieval and analysis over enterprise knowledge bases. Standard RAG pipelines place significant burden of grounding on the search stack, constraining the language model to a fixed candidate set chosen deep in the retrieval process. Our approach reduces this overdependence by layering a lightweight harness on top of existing enterprise search infrastructure, equipping a reasoning LLM with search, find, open, and summarize tools enabling the model to iteratively retrieve information, navigate within documents, and analyze evidence autonomously. On three open benchmarks we observe substantial gains: $49.6\%$ recall@1 on BRIGHT (+21.8 pp over the best embedding baseline), 0.96 factuality on WixQA ($+13\%$ relative improvement), and $92\%$ answer correctness on FinanceBench--within 2 pp of oracle access to true evidence. Ablation studies show that the most significant factor is the shift from single-shot retrieval to agentic tool use ($5.9\times$ improvement), while multi-query search and in-document navigation contribute to both quality and efficiency. We present various design choices in our agentic harness that were informed by pre-production deployments. Our results demonstrate its suitability for real-world enterprise production environments. --- ### Housing Potential Common Data Model and City Digital Twin Authors: Megan Katsumi, Mark Fox, Anderson Wong, Divnoor Chatha Published: 2026-05-07 Categories: cs.AI arXiv: https://arxiv.org/abs/2605.05535v1 The evaluation of housing potential requires consideration of a location from multiple perspectives, ranging from zoning and land use to population characteristics and access to services. This research introduces the Housing Potential Common Data Model (HPCDM) to overcome existing data silos, serving as a standard to support integration and interoperability across the diverse range of datasets that are required for housing potential analysis. This report details the evaluation of the model along with the creation of a City Digital Twin for housing and a pilot dashboard application to demonstrate a practical implementation. Beyond the technical framework, this work identifies critical barriers to adoption and provides actionable mitigation strategies for urban planners and stakeholders. --- ### Adversarial Graph Neural Network Benchmarks: Towards Practical and Fair Evaluation Authors: Tran Gia Bao Ngo, Zulfikar Alom, Federico Errica, Murat Kantarcioglu, Cuneyt Gurcan Akcora Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.05534v1 Adversarial learning and the robustness of Graph Neural Networks (GNNs) are topics of widespread interest in the machine learning community, as documented by the number of adversarial attacks and defenses designed for these purposes. While a rigorous evaluation of these adversarial methods is necessary to understand the robustness of GNNs in real-world applications, we posit that many works in the literature do not share the same experimental settings, leading to ambiguous and potentially contradictory scientific conclusions. In this benchmark, we demonstrate the importance of adopting fair, robust, and standardized evaluation protocols in adversarial GNN research. We perform a comprehensive re-evaluation of seven widely used attacks and eight recent defenses under both poisoning and evasion scenarios, across six popular graph datasets. Our study spans over 453,000 experiments conducted within a unified framework. We observe substantial differences in adversarial attack performance when evaluated under a fair and robust procedure. Our findings reveal that previously overlooked factors, such as target node selection and the training process of the attacked model, have a profound impact on attack effectiveness, to the extent of completely distorting performance insights. These results underscore the urgent need for standardized evaluations in adversarial graph machine learning. --- ### A Few Good Clauses: Comparing LLMs vs Domain-Trained Small Language Models on Structured Contract Extraction Authors: Nicole Lincoln, Nick Whitehouse, Jaron Mar, Rivindu Perera Published: 2026-05-07 Categories: cs.CL, cs.CY arXiv: https://arxiv.org/abs/2605.05532v1 This paper evaluates whether a domain trained Small Language Model (SLM) can outperform frontier Large Language Models on structured contract extraction at radically lower cost. We test Olava Extract, a self hosted legal domain Mixture of Experts model, against five frontier models. Olava Extract achieved the strongest aggregate performance in the study, with a macro F1 of 0.812 and a micro F1 of 0.842, while reducing inference cost by 78% to 97% compared with the frontier models tested. It also achieved the highest precision scores, producing fewer hallucinated and unsupported extractions, an important distinction in legal workflows where hallucinations create operational risk and downstream review burden. The findings shows that high performing, human comparable legal AI no longer requires the largest externally hosted models. More broadly, they challenge the assumption that commercially valuable enterprise AI capability must remain tied to ever larger models, massive infrastructure expenditure, and centrally hosted providers. --- ### Energy Generative Modeling: A Lyapunov-based Energy Matching Perspective Authors: Yixuan Wang, Wenqian Xue, Warren E. Dixon Published: 2026-05-07 Categories: cs.LG arXiv: https://arxiv.org/abs/2605.05530v1 Generative models based on static scalar energy functions represent an emerging paradigm in which a single time independent potential drives sample generation through its gradient field, eliminating the need for time conditioning entirely. We unify the training and sampling phases of this paradigm, conventionally treated as separate procedures, within a single framework: density transport on the Wasserstein space, cast as a nonlinear control problem in which the Kullback Leibler (KL) divergence serves as a Lyapunov function. Training and sampling are then two instances of this same master dynamics, differing only in initial condition. Within this autonomous framework we develop two analytic results. First, since the Lyapunov certificate is asymptotic, we derive a finite step stopping criterion for Langevin sampling and prove that no Lyapunov certificate exists for the deterministic gradient flow on the same energy landscape. Second, the reformulation brings the toolkit of nonlinear control theory to bear on static scalar energy generative modeling, that is, we show that additive composition of trained scalar energies retains an explicit Gibbs invariant measure and inherits the closed-loop Lyapunov certificate. Beyond these immediate results, this reformulation bridges static scalar energy generative models with the full toolkit of nonlinear control theory, opening the door to barrier functions for constrained generation and contraction metrics for accelerated sampling. Experiments on synthetic distributions validate the theoretical predictions. --- ### Discrete Elastic Ribbons: A Unified Discrete Differential Geometry Framework for One-Dimensional Energy Models Authors: Shivam Kumar Panda, M Khalid Jawed Published: 2026-05-07 Categories: cs.CE, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2605.05529v1 Elastic ribbons, slender structures whose length ($L$), width ($W$), and thickness ($b$) satisfy $L \gg W \gg b$, exhibit mechanical behaviors intermediate between one-dimensional rods ($L \gg W, b$) and two-dimensional plates ($L, W \gg b$). In quadratic Kirchhoff-type rod-based frameworks, such as Discrete Elastic Rods (DER), the governing equilibrium equations are independent of width, and therefore these models cannot capture width-dependent mechanical effects. Reduced centerline-based ribbon models attempt to capture width dependence via coupled bending-twisting energies. However, their relative accuracy remain unclear due to the absence of a unified simulation framework. In this work, we formulate a framework grounded in discrete differential geometry where the energy is expressed as functions of coupled bending-twisting strain measures along the centerline, rather than a linear sum of quadratic bending and twisting energies in DER. We derive analytical gradients and Hessians of the energy that enable implicit time integration. Within this unified setting, we compare five ribbon models: Kirchhoff, Sadowsky, Wunderlich, Sano, and Audoly. As a benchmark, a straight ribbon is longitudinally constrained into a pre-buckled arch and subjected to transverse displacement, inducing a supercritical pitchfork bifurcation. Predicted bifurcation thresholds are compared against shell-based finite element simulations, with the Sano model providing the closest agreement in capturing width-dependent shifts. Our high-performance JAX-based implementation achieves $\mathcal{O}(N)$ per-iteration cost and also confirms that Sano model introduces negligible per-iteration overhead relative to standard DER. --- ### Anatomy of a Query: W5H Dimensions and FAR Patterns for Text-to-SQL Evaluation Authors: Vicki Stover Hertzberg, Eduardo Valverde, Joyce C. Ho Published: 2026-05-07 Categories: cs.DB, cs.CL arXiv: https://arxiv.org/abs/2605.05525v1 Natural language interfaces to databases have gained popularity, yet the theoretical foundations for evaluating and designing these systems remain underdeveloped. We present QUEST (Query Understanding Evaluation through Semantic Translation), a framework resting on two independently motivated components: the FAR structural invariant, which holds that every well-formed query reduces to Filter, Aggregate, and Return operations; and the W5H dimensional framework, which holds that all filtering criteria map to six semantic dimensions (Who, What, Where, When, Why, and How). Validated across five text-to-SQL datasets (n = 120,464), FAR conformance is universal across all domains and schema types, while W5H dimensional profiles vary substantially. Healthcare queries are strongly concentrated in temporal (WHEN: 80.4%) and person-centric (WHO: 73.0%) dimensions far exceeding general-domain benchmarks, and causal (WHY) and mechanistic (HOW) reasoning are near-zero everywhere, with apparent HOW exceptions reflecting quantitative aggregation rather than genuine procedural reasoning. These results identify a frontier that must be crossed for genuine machine reasoning over structured data. --- ### Prompt Chaining in Practice: A Case Study in Automated Scholarly Report Generation Authors: Andrei Lazarev Published: 2026-05-06 Categories: cs.CL, cs.AI, cs.CE, cs.DL, cs.SE arXiv: https://arxiv.org/abs/2607.27210v1 The exponential growth of scholarly publications requires automated tools for effective information synthesis. However, simple, single-shot prompting methods often lack the reliability and quality required for complex synthesis tasks. This paper introduces and empirically evaluates a multi-stage prompt chaining methodology as a more reliable architectural pattern for such tasks. This approach is implemented in our system, AI SciBrief, which automatically generates scholarly digests. We conducted a comparative experiment, measuring the performance of our prompt chaining method against a carefully optimized single-shot baseline. Both systems were evaluated against a human-authored "gold standard" report for the "Education" domain. The results demonstrate a significant difference in reliability: our prompt chaining method achieved a 100% success rate, whereas the optimized baseline failed in 50% of its runs. In terms of quality, the proposed method also demonstrated a clear advantage, achieving a superior ROUGE-L F1-score (0.507 vs. 0.486), driven primarily by higher precision. We conclude that prompt chaining is a more dependable and effective engineering approach for complex, multi-step generative tasks, significantly mitigating the risks of failure and inconsistency inherent in monolithic prompts. --- ### Explaining GAND: A Resource on Gender-Ambiguous Natural Data & Contrastive Attribution Authors: Janiça Hackenbuchner, Jasper Degraeuwe, Arda Tezcan, Joke Daems Published: 2026-05-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.22546v1 Machine translation (MT) systems continue to produce gender-biased translations. In a time where self-expression is paramount, mistranslations based on default behaviour and stereotyping can lead to harm for users of these systems. To better understand how these systems translate gender in the absence of clear gender cues, we need benchmarking resources that reflect gender-ambiguous scenarios in a natural way. To this end, we present GAND, a gender-ambiguous natural data benchmarking resource for MT consisting of English source sentences, specifically designed to analyse the influence of contextual cues on gender in translation. We leverage GAND to conduct an interpretability analysis: we translate a subset of GAND into two grammatical gender languages and extend these with manually crafted contrastive translations. A following feature attribution analysis reveals source words in context that inform the gender translation of an ambiguous referent entity in the target translation. --- ### Semalith v1.4: A Calibrated 184M Safety Classifier Achieving State-of-the-Art Prompt-Injection Detection at 44x Fewer Parameters than Llama-Guard-3-8B Authors: Tejasvi C. Addagada Published: 2026-05-06 Categories: cs.LG, cs.AI, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2607.22545v1 Deploying large language models in financial-services and agentic settings requires safety classifiers that simultaneously handle prompt injection, regulatory compliance, and general harm, a combination no existing open guardrail addresses in a single inference pass. Semalith v1.4 is a 184M-parameter DeBERTa-v3-base classifier performing simultaneous three-axis safety classification including prompt injection, general harm, and financial-services regulatory compliance, in a single forward pass. Its 22-class head (BENIGN, nine prompt-injection sub-types, general-harm, eleven BFSI labels) is trained with a 4-class auxiliary super-category head under jointly weighted loss, on a 76,204-row corpus mined from 49 public sources with SHA-1 deduplication against every held-out evaluation set, with 21 of 22 benchmarks at zero contamination (max 0.22%). Against Llama-Guard-3-8B on 22 held-out benchmarks, Semalith v1.4 wins every prompt-injection evaluation (7/7) and 11 of 18 benchmarks overall at 44x fewer parameters, with FPR = 0.000 on 208 benign agentic prompts vs 0.063 for Llama-Guard-3-8B. On general-harm benchmarks (WildGuardMix, HEx-PHI, HarmBench), Llama-Guard-3 leads; this complementary split is documented in Section 4. Six measured weak spots are disclosed in Section 6. Deployment guidance: v1.3 is recommended for conversational moderation deployments (ToxicChat F1 0.624); v1.4 is recommended when BFSI label coverage or zero-FPR on benign agentic prompts is the priority. --- ### UzWordnet and Generative AI for Learning Uzbek by Game Playing Authors: Alessandro Agostini, Saydobid Khusanov, Mirkamol Mirkamilov Published: 2026-05-06 Categories: cs.CL, cs.HC, cs.SE arXiv: https://arxiv.org/abs/2607.14104v1 This paper presents an educational system architecture that enables learners to practice the Uzbek language through game-playing. The architecture integrates UzWordnet and the largest currently available orthographic dictionary for Uzbek as core lexical resources, together with generative AI as a fundamental component for learning support. We design four educational games to facilitate Uzbek language learning and propose a game-based methodology for improving UzWordnet as a direct by-product of game dynamics. Our approach combines game design and lexical resources to address objectives that are at the same time educational (language learning) and lexical (improvement and enrichment of a lexical resource). --- ### Latent Communication Between Language Model Agents: Channels, Alignment, and the Limits of Text Authors: Markus Wenzel Published: 2026-05-06 Categories: cs.CL, cs.MA arXiv: https://arxiv.org/abs/2607.14103v1 Multi-agent systems (MAS) are utilized in many contexts and many professions. Those MAS rely on inter-agent communication, usually implemented by clear-text message passing. We hypothesize that Large Language Models may have a world model at their disposal that exceeds expressibility in text when complex concepts need to be communicated. Our aim is to approach a proof of this hypothesis with structured experiments. In this work, we show that LLM agents communicating via text lose information, which we quantify via Sparse Autoencoder (SAE) feature analysis. We construct three communication channels and measure concept-discriminating information in each. We first show that the SAE-sparse channel retains a 99.4% probe accuracy at 28-fold compression over the dense-latent channel vs 80.4% for the text channel. We then proceed to examine the same for cross-architecture communication by using sparse latent space alignment. We find for Procrustes alignment a 92% top-1 retrieval between Llama and Mistral. Using a text round-trip, we perform feature survival analysis to find that text serialization destroys 88% of SAE features, replacing them with a different feature set. We attribute the loss to identity replacement, not attenuation. By our analysis, we were able to attribute a 3-10pp performance penalty to the linear Procrustes alignment, improving with nonlinear alignment methods. In a task-level evaluation we find that the latent channel matches the text channel on cross-lingual concept tasks but never exceeds it. Text augmentation with latent features provides no benefit, leading us to negative conclusions for the initial hypothesis: lost features mostly or completely encode surface form, not task-relevant semantics. To pinpoint the practical advantage of latent communication over a text channel, deeper tasks eliciting complex concepts and an corresponding analysis framework are needed. --- ### UniSAGE: Unifying Static and Dynamic Attributes with Hyper-Structure Authors: Taoran Fang, Yan Deng, Chunping Wang, Yang Wang, Lei Chen, Yang Yang Published: 2026-05-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2607.14102v1 With the rapid growth of digital data, real-world applications increasingly involve hierarchical information that combines static attributes with dynamic records. Modeling such heterogeneous data in a unified and generalizable manner remains challenging. Existing approaches often rely on extensive manual design, are tightly coupled to specific data schemas, and typically process static and dynamic attributes in isolation, thereby overlooking their implicit interactions. We propose UniSAGE, a unified framework for modeling data with both static and dynamic attributes. UniSAGE constructs a global attribute graph that represents hierarchical and temporal relationships in a unified structure. To ensure representational consistency, it introduces two orthogonal parameter subspaces that jointly support static aggregation and dynamic reasoning within a shared semantic space. Building on these unified representations, UniSAGE further enables task-specific interaction between static and dynamic attributes via a lightweight hyper-structure mechanism. UniSAGE is fully automated, robust to evolving data schemas, and capable of capturing complex cross-attribute dependencies. Extensive experiments on multiple public benchmarks and a real-world financial behavior dataset demonstrate that UniSAGE consistently outperforms existing methods, achieving performance improvements of over 10% on several tasks. --- ### Ask Before You Diagnose: Safe-Psych, a Sequential Evaluation Benchmark for LLMs in Psychiatry Authors: Oriana Presacan, Andreea Grama, Larisa Irimină, Alireza Nik, Jaya Ojha, Vajira Thambawita, Ciprian I. Băcilă, Bogdan Ionescu, Michael A. Riegler Published: 2026-05-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.13036v1 Large language models (LLMs) are increasingly used for decision support in healthcare, but clinical evidence is often incomplete or evolving. When the available information is insufficient to support a reliable answer, models should request clarification or abstain rather than provide unsupported responses. Existing medical benchmarks, however, typically assume that complete information is available upfront. We introduce Safe-Psych, a sequential benchmark for evaluating how LLMs handle evolving diagnostic uncertainty in clinical psychiatry. Safe-Psych contains over 1,000 real-world psychiatric clinical notes segmented to simulate incremental evidence disclosure, with psychiatrist-derived action labels at each stage: DIAGNOSE, CLARIFY, or ABSTAIN. We evaluate multiple state-of-the-art LLMs in full-information and sequential settings. Our findings show that capability does not ensure calibration: even strong models struggle under incomplete clinical information, with under-abstention exceeding 60% for most models and safety-aware prompting reducing premature commitment only by shifting errors toward excessive abstention. In sequential evaluation, models frequently diagnose before sufficient evidence is available and rarely seek clarification unless explicitly prompted; these premature diagnoses are less accurate than on-time diagnoses. Overall, Safe-Psych reveals a limitation across the evaluated models: recognizing when clinical evidence is incomplete and additional information is needed. We release Safe-Psych to support research on improving LLM safety in healthcare. --- ### CANDI: Contextual Alignment for Niche Domains Question Answering Authors: Megha Chakraborty, Darssan L. Eswaramoorthi, Het Riteshkumar Shah, Madhur Thareja, Michelle A Ihetu, Harshul Raj Surana, Kaushik Roy, Amit Sheth Published: 2026-05-06 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2607.11891v1 The deployment of large language models (LLMs) in specialized domains like medical diagnostics and financial advisory necessitates evaluating capabilities beyond general knowledge. Traditional question-answering benchmarks often fail to capture the nuanced contextual grounding, user awareness, and domain understanding these fields require. To address this, we introduce CANDI-QA (Contextual Alignment for Niche Domains Question Answering), a novel dataset evaluating LLMs on delivering accurate, context-sensitive, and user-aligned answers in specialized settings. CANDI-QA features expert-curated question-answer pairs structured into two categories: (1) Information Assistance Questions, which are direct, factual queries requiring precise extraction, and (2) Applied Inference Questions, which are multi-hop reasoning tasks needing situational inference to generate actionable insights. We evaluate over ten diverse language models, from compact open-source to state-of-the-art proprietary systems. As a robust baseline, we present MTSS-Net, a lightweight neuro-symbolic framework combining neural retrieval with rule-based reasoning. Our findings highlight the profound challenges of achieving contextual alignment in niche domains, revealing the limitations of current LLMs without enhanced contextual or symbolic integration. Ultimately, CANDI-QA serves as a critical benchmark for advancing research in context-aware language models, stimulating the development of robust, trustworthy AI for high-stakes domains. --- ### Role-Based Agentic AI for Intent-Driven Network and Service Orchestration Authors: Juan Parra-Ullauri, Talha Ahmed Khan, Daniel McHugh, Shipra Kapoor, Alistair Duke, Alicia Hey, Andy Corston-Petrie Published: 2026-05-06 Categories: cs.NI, cs.AI arXiv: https://arxiv.org/abs/2606.20580v1 Telecommunication networks are increasingly complex due to heterogeneous technologies, diverse service requirements, and growing demands for resource efficiency and business agility. Intent-Based Networking (IBN) and, more recently, agentic AI have emerged as promising paradigms to address this complexity through autonomous network management. However, existing approaches primarily focus on operational orchestration within Operations Support Systems (OSS) and lack an integrated framework that spans Business Support Systems (BSS) and OSS, limiting the realisation of true intent-to-business-to-network coordination. This paper presents a role-based multi-agent architecture (MAS) for end-to-end intent orchestration that mirrors Communication Service Provider (CSP) organisational structures. The proposed framework applies principles of functional decomposition, explicit task ownership, privacy-preserving domain separation, and domain-specific expertise within a hierarchical four-layer agent system spanning customer engagement, strategic planning, service delivery, and infrastructure provisioning. Leadership agents coordinate planning activities, whilst specialised service and resource agents are dynamically instantiated according to intent requirements. A proof-of-concept implementation demonstrates the feasibility of bridging the BSS-OSS divide through structured agent coordination, illustrating how agentic MAS can support accountable and scalable intent-driven service orchestration. --- ### From Memorization to Creation: Evaluating the Cognitive Depth of LLM-Generated Educational Questions Authors: Xiaolong Wang, Zhe Zhao, Song Lai, Chaoli Zhang, Zijie Geng, Yu Tong, Ye Wei, Qingsong Wen Published: 2026-05-06 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2606.18257v1 While LLMs show promise in automating educational content creation, their ability to generate questions that stimulate higher-order thinking remains understudied. This work evaluates six widely-used LLMs through a Bloom's Taxonomy lens, focusing on their capacity to transcend rote memorization and achieve cognitive leaps. Using a hybrid human--AI evaluation protocol, we generate and analyze 20{,}700 questions across computer science, K--12 math, and social-science domains. Key contributions include: (1) a fine-grained prompting strategy that reduces question repetitiveness by 24.45\% for Qwen2.5-7B-Instruct, and increases the proportion of higher-order cognitive level outputs by 11.53\% for InternLM3-8B-Instruct; (2) quantitative metrics for cognitive shift intensity (CogShift) and category drift, revealing InternLM3's superior performance in multi-level transitions; (3) an interpretability analysis revealing metric-level correlations that enhance the transparency of Chain-of-Thought prompting. Our findings highlight the importance of cognitive-aware prompt design and provide benchmarks for deploying LLMs in personalized learning systems. --- ### AI SciBrief as a Gateway to Research: A Framework for Onboarding Students into New Research Areas Authors: Andrei Lazarev, Dmitrii Sedov Published: 2026-05-06 Categories: cs.CY, cs.AI, cs.CE, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2606.12413v1 Students at all levels of higher education face a significant barrier in the form of information overload, which often paralyzes the initial stages of the research process and suppresses motivation. In response, this article introduces a pedagogical framework that leverages AI SciBrief, a platform powered by a Large Language Model (LLM) designed to automatically generate digests of scientific trends. We describe how this multidisciplinary tool - with initial coverage in finance, medicine, and education - can be integrated into the curriculum to overcome this "entry barrier." The framework provides concrete methodologies for utilizing these digests to facilitate topic selection for term papers, accelerate literature reviews for dissertations, and enable postgraduate students to continuously monitor emerging trends. We conclude that AI SciBrief functions as a "gateway to research" effectively reducing students' cognitive load and empowering them to transition more rapidly from information searching to knowledge creation. --- ### SODE: Analyzing Social Dynamics in LLM Agents Authors: Inseo Jung, Yoonseok Oh, Kyungryul Back, Jinkyu Kim, Jungbeom Lee Published: 2026-05-06 Categories: cs.MA, cs.AI arXiv: https://arxiv.org/abs/2605.23949v1 As Large Language Models (LLMs) evolve into interactive agents, understanding their behavioral alignment within human social dynamics becomes essential. While behavioral game theory offers a framework to study these interactions, previous work has predominantly relied on outcome-based metrics such as average scores. This focus overlooks the mechanisms that facilitate sustainable cooperation, as identical scores can be derived from vastly different strategies. To bridge this gap, we introduce SODE (Social Dynamics Evaluation), a framework that evaluates LLM agents across three evolutionary dimensions: Direct Reciprocity for strategy adaptation, Indirect Reciprocity for reputation sensitivity, and Group Dynamics for cooperative resilience. Applying SODE reveals systematic divergences: instruction-tuned models often exhibit "passive compliance" that renders them vulnerable to exploitation, while reasoning models prioritize short-horizon optimization, destabilizing long-term cooperation. Notably, we demonstrate that a "long-horizon framing" can unlock reciprocal capabilities in reasoning models. Thus, SODE offers a systematic, mechanism-grounded benchmark for aligning AI agents with complex human social dynamics. --- ### The Insurability Frontier of AI Risk: Mapping Threats to Affirmative Coverage, Silent Exposures, and Exclusions Authors: Alex Leung, Rex Zhang, Ervin Ling, Kentaroh Toyoda, SiewMei Loh Published: 2026-05-06 Categories: q-fin.RM, cs.AI, cs.CR, cs.CY, econ.GN arXiv: https://arxiv.org/abs/2605.18784v2 The rapid diffusion of agentic AI has created a new coverage problem for commercial insurance: some AI-mediated losses are now affirmatively insured, some create silent-AI exposure under legacy cyber, technology errors-and-omissions (E&O), directors-and-officers (D&O), employment practices liability (EPLI), crime, and media policies, and others are being actively excluded. This paper maps that emerging boundary by coding 55 AI threat classes against 26 insurance products, endorsements, and exclusion regimes using public carrier materials and OWASP/MITRE threat catalogs. We identify a four-tier insurability frontier: affirmatively insured perils, silent-AI exposures, actively excluded perils, and perils outside conventional private insurance structures. Our coding measures publicly claimed positioning rather than executed contract wording; the headline statistics describe what carriers publicly state about coverage, not what would be paid in any specific claim. Three patterns emerge. First, affirmative AI coverage is beginning to differentiate by primary risk emphasis: public materials often position Munich Re around model performance and drift, Armilla and parts of the Lloyd's market around hallucination and broader AI liability, Tokio Marine Kiln and CFC around IP and technology E&O concerns, Apollo ibott around emerging autonomous system liability, and Coalition around deepfake and AI-enabled cyber response. Second, legacy lines retain silent-AI exposure where AI is an instrumentality rather than the legal cause of loss. Third, foundation model concentration is the clearest genuinely novel insurability frontier because upstream model failure can correlate losses across many cedents at once; the relevant market design question is which insurability constraint each candidate structure relaxes, not merely which systemic risk template exists. --- ### Large Language Models for Web Accessibility: A Systematic Literature Review Authors: Wajdi Aljedaani, Rubel Hassan Mollik Published: 2026-05-06 Categories: cs.DL, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2605.13873v1 Web accessibility aims to ensure that web content and services are usable by people with diverse abilities. In recent years, Large Language Models (LLMs) have been increasingly explored to support accessibility-related tasks on the web, such as content generation, issue detection, and remediation. However, little is known about the characteristics of these approaches, the accessibility issues they target, the standards they follow, and how they are evaluated. In this paper, we present a systematic literature review of 38 peer-reviewed studies that investigate the use of LLMs in web accessibility contexts. We begin by performing a comprehensive search of scientific publications to identify relevant studies. We then conduct a comparative analysis to examine the accessibility tasks addressed, the LLM models and prompting strategies employed, the system architectures adopted, the accessibility issues and guidelines considered, and the evaluation methods used across studies. Our findings show that most studies apply LLMs to text-centric and structurally explicit accessibility tasks, with WCAG serving as the primary reference framework and limited consideration of cognitive accessibility guidelines (COGA). The reviewed approaches predominantly rely on general-purpose LLMs and prompt-based interactions, while evaluation practices vary widely and often lack direct involvement of users with disabilities. We envision this review as a consolidated reference for researchers and practitioners seeking to understand the current landscape of LLM-supported web accessibility, and as a foundation to guide future research and tool development in this area. --- ### Two Hebrew folk meteorological proverbs tested: rainfall on Rosh Chodesh and Shabbat Mevarechim as predictors of monthly precipitation (Israel, 1950-2024) Authors: Abraham Itzhak Weinberg Published: 2026-05-06 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2605.10960v1 Folk meteorological proverbs encode centuries of empirical observation by agricultural communities. Two Hebrew proverbs link lunar calendar anchor days to monthly winter rainfall: (i) "If Rosh Chodesh is rainy, the whole month is rainy" and (ii) "If it rains on Shabbat Mevarechim, the whole month is rainy." Shabbat Mevarechim is the last Saturday before each new Hebrew month, preceding Rosh Chodesh by one to seven days. The first proverb is widely known; the second circulates in Hasidic oral tradition with no identified written source. Both have never been formally tested. We analyse 75 years (1950-2024) of daily precipitation data from seven Israeli cities across three climatic regions, comprising 191,758 station-days and 2,422 Hebrew-month observations during the winter rainy season (Marcheshvan-Adar). A rainy Rosh Chodesh increases the probability of a rainy month from 22.2% to 38.6% (lift +16.4 percentage points; chi-square = 57.8, p = 2.9e-14; Bayes factor 1.81). A rainy Shabbat Mevarechim produces a similar effect (lift +16.5 percentage points, p = 8.0e-13), despite preceding Rosh Chodesh by up to seven days. The effect decays with lag and mirrors daily rainfall autocorrelation (r = 0.35-0.44 at lag 1; ~0 at lag 7), consistent with Mediterranean cyclone persistence. A bootstrap permutation test (p < 1e-4) and a 15-year rolling analysis show declining predictive power (-0.20 percentage points per year, p < 0.001), consistent with shortening precipitation events under warming climate conditions. Both proverbs encode real but probabilistic meteorological signals whose reliability is decreasing over time. --- ### Inverse Design of Multi-Layer Sub-Pixel-Resolution RF Passives Through Grayscale Diffusion with Flexible S-Parameter Conditioning Authors: Tommaso Dreossi, Christopher M. Bryant, Hao Liu, Nathan Mirman, Noah Kessler, Michael Frei, Harish Krishnaswamy Published: 2026-05-06 Categories: eess.SP, cs.LG arXiv: https://arxiv.org/abs/2605.08233v1 Inverse design of RF passive components from S-parameters is a high-dimensional, ill-posed problem, and prior generative approaches are limited to single-layer binary-metallization structures. This paper presents an inverse design approach that generates passive components from partial S-parameter inputs on an $8\times8$ mm board discretized at $64\times64$ pixels with sub-pixel grayscale metallization across 1-20 GHz. The framework generates two-layer copper layouts with vias, with hard physical constraints on feed locations enforced through annealed Langevin projection, flexible multi-modal conditioning on partial S-parameter specifications, port locations, dielectric properties, reference topology, and variable port placement. Candidate designs are generated in seconds, with surrogate-predicted S-parameters matching targets to within $0.77 \pm 1.28$ dB weighted mean absolute error. We validate the approach with two fabricated designs on RO4003C: a manufacturable alternative to a hairpin filter whose coupling gaps violate fabrication rules, and a combline bandpass filter designed from scratch given only target S-parameters. --- ### Hierarchical Multi-Fidelity Learning for Predicting Three-Dimensional Flame Wrinkling and Turbulent Burning Velocity Authors: Saghar Zolfaghari, Yu Xie, Junfeng Yang, Safa Jamali Published: 2026-05-06 Categories: cs.LG, physics.comp-ph, physics.flu-dyn arXiv: https://arxiv.org/abs/2605.08232v1 High-fidelity experimental characterization of turbulent premixed flames remains limited by the cost and complexity of advanced diagnostics, particularly under elevated pressures and intense turbulence where measurements of coupled flame morphology and burning dynamics are sparse. Here, we develop a hierarchical multi-fidelity neural network framework (MuFiNNs) to address this challenge by integrating sparse high-fidelity experimental data with structured low-fidelity representations encoding dominant physical trends. The framework combines hierarchical low-fidelity construction with nonlinear multi-fidelity correction to learn coupled geometric and reactive flame behavior while recovering discrepancies that simplified models alone cannot capture. The methodology is applied to expanding turbulent premixed flames to predict three-dimensional flame wrinkling dynamics and turbulent mass burning velocity across varying fuels, pressures, and turbulence intensities. Using experimentally informed low-fidelity trend models with sparse high-fidelity measurements, MuFiNNs accurately reconstruct observed flame behavior, enable interpolation across unseen operating conditions, and demonstrate robust extrapolation beyond the training domain. Importantly, the framework remains effective in noisy, weakly structured, or experimentally inaccessible regimes where conventional data-driven approaches often fail. These results show that hierarchical multi-fidelity learning provides a scalable and physically grounded strategy for predictive combustion modeling in data-limited regimes. More broadly, this work establishes multi-fidelity scientific machine learning as a practical framework for extracting physically meaningful predictive models from sparse experiments, particularly for instability-dominated and turbulence-sensitive reactive flows where high-fidelity data acquisition is demanding. --- ### TRAM: Training Approximate Multiplier Structures for Low-Power AI Accelerators Authors: Chang Meng, Hanyu Wang, Yuyang Ye, Mingfei Yu, Wayne Burleson, Giovanni De Micheli Published: 2026-05-06 Categories: cs.LG, cs.AI, cs.AR arXiv: https://arxiv.org/abs/2605.08231v1 Reducing power consumption in AI accelerators is increasingly important. Approximate computing can reduce power consumption while keeping the accuracy loss small. Since multipliers are power-hungry components in AI models, this paper focuses on synthesizing low-power approximate multipliers (AxMs). Unlike prior works that design AxMs separately from AI model training, we present TRAM, which jointly optimizes the AxM structure and AI model parameters to lower power with small accuracy loss. Experiments show that compared to state-of-the-art AxMs, TRAM achieves up to 25.05% AxM power reduction on CNNs with CIFAR-10, and reduces power by up to 27.09% on vision transformers with ImageNet. ---