# Research Papers — 2026-03 · bundle page 185 of 243 > Full-text markdown of papers #9201–9250 of 12123 published in 2026-03, newest first. AI-readable mirrors of the full 12123-paper 2026-03 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-03/page/185 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/184.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/186.md) ### Writing literature reviews with AI: principles, hurdles and some lessons learned Authors: Saadi Lahlou, Annabelle Gouttebroze, Atrina Oraee, Julian Madera Published: 2026-03-08 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2603.20235v1 We qualitatively compared literature reviews produced with varying degrees of AI assistance. The same LLM, given the same corpus of 280 papers but different selections, produced dramatically different reviews, from mainstream and politically neutral to critical and post-colonial, though neither orientation was intended. LLM outputs always appear at first glance to be well written, well informed and thought out, but closer reading reveals gaps, biases and lack of depth. Our comparison of six versions shows a series of pitfalls and suggests precautions necessary when using AI assistance to make a literature review. Main issues are: (1) The bias of ignorance (you do not know what you do not get) in the selection of relevant papers. (2) Alignment and digital sycophancy: commercial AI models slavishly take you further in the direction they understand you give them, reinforcing biases. (3) Mainstreaming: because of their statistical nature, LLM productions tend to favor mainstream perspectives and content; in our case there was only 20% overlap between paper selections by humans and the LLM. (4) Limited capacity for creative restructuring, with vague and ambiguous statements. (5) Lack of critical perspective, coming from distant reading and political correctness. Most pitfalls can be addressed by prompting, but only if the user knows the domain well enough to detect them. There is a paradox: producing a good AI-assisted review requires expertise that comes from reading the literature, which is precisely what AI was meant to reduce. Overall, AI can improve the span and quality of the review, but the gain of time is not as massive as one would expect, and a press-button strategy leaving AI to do the work is a recipe for disaster. We conclude with recommendations for those who write, or assess, such LLM-augmented reviews. --- ### Emergency Lane-Change Simulation: A Behavioral Guidance Approach for Risky Scenario Generation Authors: Chen Xiong, Cheng Wang, Yuhang Liu, Zirui Wu, Ye Tian Published: 2026-03-08 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2603.20234v1 In contemporary autonomous driving testing, virtual simulation has become an important approach due to its efficiency and cost effectiveness. However, existing methods usually rely on reinforcement learning to generate risky scenarios, making it difficult to efficiently learn realistic emergency behaviors. To address this issue, we propose a behavior guided method for generating high risk lane change scenarios. First, a behavior learning module based on an optimized sequence generative adversarial network is developed to learn emergency lane change behaviors from an extracted dataset. This design alleviates the limitations of existing datasets and improves learning from relatively few samples. Then, the opposing vehicle is modeled as an agent, and the road environment together with surrounding vehicles is incorporated into the operating environment. Based on the Recursive Proximal Policy Optimization strategy, the generated trajectories are used to guide the vehicle toward dangerous behaviors for more effective risk scenario exploration. Finally, the reference trajectory is combined with model predictive control as physical constraints to continuously optimize the strategy and ensure physical authenticity. Experimental results show that the proposed method can effectively learn high risk trajectory behaviors from limited data and generate high risk collision scenarios with better efficiency than traditional methods such as grid search and manual design. --- ### Generalized Stock Price Prediction for Multiple Stocks Combined with News Fusion Authors: Pei-Jun Liao, Hung-Shin Lee, Yao-Fei Cheng, Li-Wei Chen, Hung-yi Lee, Hsin-Min Wang Published: 2026-03-08 Categories: q-fin.ST, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2603.19286v1 Predicting stock prices presents challenges in financial forecasting. While traditional approaches such as ARIMA and RNNs are prevalent, recent developments in Large Language Models (LLMs) offer alternative methodologies. This paper introduces an approach that integrates LLMs with daily financial news for stock price prediction. To address the challenge of processing news data and identifying relevant content, we utilize stock name embeddings within attention mechanisms. Specifically, we encode news articles using a pre-trained LLM and implement three attention-based pooling techniques -- self-attentive, cross-attentive, and position-aware self-attentive pooling -- to filter news based on stock relevance. The filtered news embeddings, combined with historical stock prices, serve as inputs to the prediction model. Unlike prior studies that focus on individual stocks, our method trains a single generalized model applicable across multiple stocks. Experimental results demonstrate a 7.11% reduction in Mean Absolute Error (MAE) compared to the baseline, indicating the utility of stock name embeddings for news filtering and price forecasting within a generalized framework. --- ### Beam-aware Kernelized Contextual Bandits for User Association and Beamforming in mmWave Vehicular Networks Authors: Xiaoyang He, Manabu Tsukada Published: 2026-03-08 Categories: cs.IT, cs.LG arXiv: https://arxiv.org/abs/2603.19285v1 Timely channel information is necessary for vehicles to determine both the serving base station (BS) and the beamforming vector, but frequent estimation of fast-fading mmWave channels incurs significant overhead. To address this challenge, we propose a Beam-aware Kernelized Contextual Upper Confidence Bound (BKC-UCB) algorithm that estimates instantaneous transmission rates without additional channel measurements by exploiting historical contexts such as vehicle location and velocity, together with past observed transmission rates. Specifically, BKC-UCB leverages kernel methods to capture the nonlinear relationship between context and transmission rate by mapping contexts into a reproducing kernel Hilbert space (RKHS), where linear learning becomes feasible. Rather than treating each beam as an independent arm, the beam index is embedded into the context, enabling BKC-UCB to exploit correlations among beams to accelerate convergence. Furthermore, an event-triggered information sharing mechanism is incorporated into BKC-UCB, enabling information exchange only when significant explorations are conducted to improve learning efficiency with limited communication overhead. --- ### CDEoH: Category-Driven Automatic Algorithm Design With Large Language Models Authors: Yu-Nian Wang, Shen-Huan Lyu, Ning Chen, Jia-Le Xu, Baoliu Ye, Qingfu Zhang Published: 2026-03-08 Categories: cs.NE, cs.AI arXiv: https://arxiv.org/abs/2603.19284v1 With the rapid advancement of large language models (LLMs), LLM-based heuristic search methods have demonstrated strong capabilities in automated algorithm generation. However, their evolutionary processes often suffer from instability and premature convergence. Existing approaches mainly address this issue through prompt engineering or by jointly evolving thought and code, while largely overlooking the critical role of algorithmic category diversity in maintaining evolutionary stability. To this end, we propose Category Driven Automatic Algorithm Design with Large Language Models (CDEoH), which explicitly models algorithm categories and jointly balances performance and category diversity in population management, enabling parallel exploration across multiple algorithmic paradigms. Extensive experiments on representative combinatorial optimization problems across multiple scales demonstrate that CDEoH effectively mitigates convergence toward a single evolutionary direction, significantly enhancing evolutionary stability and achieving consistently superior average performance across tasks and scales. --- ### InfoMamba: An Attention-Free Hybrid Mamba-Transformer Model Authors: Youjin Wang, Jiaqiao Zhao, Rong Fu, Run Zhou, Ruizhe Zhang, Jiani Liang, Suisuai Cao, Feng Zhou Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.18031v2 Balancing fine-grained local modeling with long-range dependency capture under computational constraints remains a central challenge in sequence modeling. While Transformers provide strong token mixing, they suffer from quadratic complexity, whereas Mamba-style selective state-space models (SSMs) scale linearly but often struggle to capture high-rank and synchronous global interactions. We present a consistency boundary analysis that characterizes when diagonal short-memory SSMs can approximate causal attention and identifies structural gaps that remain. Motivated by this analysis, we propose InfoMamba, an attention-free hybrid architecture. InfoMamba replaces token-level self-attention with a concept bottleneck linear filtering layer that serves as a minimal-bandwidth global interface and integrates it with a selective recurrent stream through information-maximizing fusion (IMF). IMF dynamically injects global context into the SSM dynamics and encourages complementary information usage through a mutual-information-inspired objective. Extensive experiments on classification, dense prediction, and non-vision tasks show that InfoMamba consistently outperforms strong Transformer and SSM baselines, achieving competitive accuracy-efficiency trade-offs while maintaining near-linear scaling. --- ### InfoMamba: An Attention-Free Hybrid Mamba-Transformer Model Authors: Youjin Wang, Jiaqiao Zhao, Rong Fu, Run Zhou, Ruizhe Zhang, Jiani Liang, Suisuai Cao, Feng Zhou Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.18031v1 Balancing fine-grained local modeling with long-range dependency capture under computational constraints remains a central challenge in sequence modeling. While Transformers provide strong token mixing, they suffer from quadratic complexity, whereas Mamba-style selective state-space models (SSMs) scale linearly but often struggle to capture high-rank and synchronous global interactions. We present a consistency boundary analysis that characterizes when diagonal short-memory SSMs can approximate causal attention and identifies structural gaps that remain. Motivated by this analysis, we propose InfoMamba, an attention-free hybrid architecture. InfoMamba replaces token-level self-attention with a concept bottleneck linear filtering layer that serves as a minimal-bandwidth global interface and integrates it with a selective recurrent stream through information-maximizing fusion (IMF). IMF dynamically injects global context into the SSM dynamics and encourages complementary information usage through a mutual-information-inspired objective. Extensive experiments on classification, dense prediction, and non-vision tasks show that InfoMamba consistently outperforms strong Transformer and SSM baselines, achieving competitive accuracy-efficiency trade-offs while maintaining near-linear scaling. --- ### Quine: Realizing LLM Agents as Native POSIX Processes Authors: Hao Ke Published: 2026-03-08 Categories: cs.OS, cs.AI, cs.PL, cs.SE arXiv: https://arxiv.org/abs/2603.18030v2 Current LLM agent frameworks often implement isolation, scheduling, and communication at the application layer, even though these mechanisms are already provided by mature operating systems. Instead of introducing another application-layer orchestrator, this paper presents Quine, a runtime architecture and reference implementation that realizes LLM agents as native POSIX processes. The mapping is explicit: identity is PID, interface is standard streams and exit status, state is memory, environment variables, and filesystem, and lifecycle is fork/exec/exit. A single executable implements this model by recursively spawning fresh instances of itself. By grounding the agent abstraction in the OS process model, Quine inherits isolation, composition, and resource control directly from the kernel, while naturally supporting recursive delegation, context renewal via exec, and shell-native composition. The design also exposes where the POSIX process model stops: processes provide a robust substrate for execution, but not a complete runtime model for cognition. In particular, the analysis points toward two immediate extensions beyond process semantics: task-relative worlds and revisable time. A reference implementation of Quine is publicly available on GitHub. --- ### Engineering Verifiable Modularity in Transformers via Per-Layer Supervision Authors: J. Clayton Kerce Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.18029v1 Transformers resist surgical control. Ablating an attention head identified as critical for capitalization produces minimal behavioral change because distributed redundancy compensates for damage. This Hydra effect renders interpretability illusory: we may identify components through correlation, but cannot predict or control their causal role. We demonstrate that architectural interventions can expose hidden modularity. Our approach combines dual-stream processing separating token and contextual representations, per-layer supervision providing independent gradient signal at each depth, and gated attention regularizing toward discrete activation patterns. When trained with per-layer supervision, models produce ablation effects 5 to 23 times larger than architecturally identical controls trained with standard objectives. This enables 4 times greater control leverage on targeted behaviors: scaling identified attention heads produces smooth, predictable changes in model output. The key finding is architectural. Without per-layer supervision, ablation damage concentrates near zero with low variance (Winograd standard deviation 0.63%). With per-layer supervision, effects spread widely (standard deviation 6.32%), revealing which predictions depend on which circuits. The larger variance is not measurement noise but the signature of unmasked modularity. We validate our approach through three components: engineered features that capture computational dynamics rather than vocabulary structure (validated by near-zero correlation with raw activation clustering), an architecture providing positive control for modularity, and causal experiments demonstrating functional reorganization where different tasks route through different attention heads. This es tablishes a methodology for transforming interpretability from passive observation to active control. --- ### What on Earth is AlphaEarth? Hierarchical structure and functional interpretability for global land cover Authors: Ivan Felipe Benavides-Martinez, Justin Guthrie, Jhon Edwin Arias, Yeison Alberto Garces-Gomez, Angela Ines Guzman-Alvis, Cristiam Victoriano Portilla-Cabrera, Somnath Mondal, Andrew J. Allyn, Auroop R. Ganguly Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.16911v1 Geospatial foundation models generate high-dimensional embeddings that achieve strong predictive performance, yet their internal organization remains obscure, limiting their scientific use. Recent interpretability studies relate Google AlphaEarth Foundations (GAEF) embeddings to continuous environmental variables, but it is still unclear whether the embedding space exhibits a functional or hierarchical organization, in which some dimensions act as specialized representations while others encode shared or broader geospatial structure. In this work, we propose a functional interpretability framework that reverse-engineers the role of embedding dimensions by characterizing their contribution to land cover structure from observed classification behavior. The approach combines large-scale experimentation with a structural analysis of embedding-class relationships based on feature importance patterns and progressive ablation. Our results show that embedding dimensions exhibit consistent and non-uniform functional behavior, allowing them to be categorized along a hierarchical functional spectrum: specialist dimensions associated with specific land cover classes, low- and mid-generalist dimensions capturing shared characteristics between classes, and highgeneralist dimensions reflecting broader environmental gradients. Critically, we find that accurate land cover classification (98% of baseline performance) can be achieved using as few as 2 to 12 of the 64 available dimensions, depending on the class. This demonstrates substantial redundancy in the embedding space and offers a pathway toward significant reductions in computational cost. Together, these findings reveal that AlphaEarth embeddings are not only physically informative, but also functionally organized into a hierarchical structure, providing practical guidance for dimension selection in operational classification tasks. --- ### Machine Learning Based Identification of Solvents from Post-Desiccation Patterns Authors: Jesús Israel Morán-Cortés, Felipe Pacheco-Vázquez Published: 2026-03-08 Categories: cond-mat.soft, cs.LG, physics.app-ph, physics.comp-ph, physics.data-an arXiv: https://arxiv.org/abs/2603.15660v1 We introduce an optimized protocol of fracture pattern classification using an artificial neural network to identify the solvent involved in the desiccation cracking process of starch-liquid slurries, even after it has been completely evaporated. For this purpose, image analysis techniques were used to characterize patterns obtained from drying suspensions using single solvents (water, ethanol, acetone) and two-component solvents (water-ethanol mixtures at different concentrations). Frequency histograms were generated based on nine morphological features, taking into account their size, shape, geometry and orientational ordering. Subsequently, we used these histograms as input data into artificial neural network variants to determine the set of features that lead to the higher accuracy in solvent identification. We obtained an average accuracy of $96(\pm 1)\%$ considering all solvents in the analysis. The highest accuracy was obtained with sets of features that include the crack area distribution. The proposed protocol can help to determine the combination of features that optimize pattern recognition in other fields of science and engineering. --- ### Did You Check the Right Pocket? Cost-Sensitive Store Routing for Memory-Augmented Agents Authors: Madhava Gaikwad Published: 2026-03-08 Categories: cs.AI, cs.CL, cs.IR arXiv: https://arxiv.org/abs/2603.15658v1 Memory-augmented agents maintain multiple specialized stores, yet most systems retrieve from all stores for every query, increasing cost and introducing irrelevant context. We formulate memory retrieval as a store-routing problem and evaluate it using coverage, exact match, and token efficiency metrics. On downstream question answering, an oracle router achieves higher accuracy while using substantially fewer context tokens compared to uniform retrieval, demonstrating that selective retrieval improves both efficiency and performance. Our results show that routing decisions are a first-class component of memory-augmented agent design and motivate learned routing mechanisms for scalable multi-store systems. We additionally formalize store selection as a cost-sensitive decision problem that trades answer accuracy against retrieval cost, providing a principled interpretation of routing policies. --- ### Attribution-Guided Model Rectification of Unreliable Neural Network Behaviors Authors: Peiyu Yang, Naveed Akhtar, Jiantong Jiang, Ajmal Mian Published: 2026-03-08 Categories: cs.LG, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2603.15656v1 The performance of neural network models deteriorates due to their unreliable behavior on non-robust features of corrupted samples. Owing to their opaque nature, rectifying models to address this problem often necessitates arduous data cleaning and model retraining, resulting in huge computational and manual overhead. In this work, we leverage rank-one model editing to establish an attribution-guided model rectification framework that effectively locates and corrects model unreliable behaviors. We first distinguish our rectification setting from existing model editing, yielding a formulation that corrects unreliable behavior while preserving model performance and reducing reliance on large budgets of cleansed samples. We further reveal a bottleneck of model rectifying arising from heterogeneous editability across layers. To target the primary source of misbehavior, we introduce an attribution-guided layer localization method that quantifies layer-wise editability and identifies the layer most responsible for unreliabilities. Extensive experiments demonstrate the effectiveness of our method in correcting unreliabilities observed for neural Trojans, spurious correlations and feature leakage. Our method shows remarkable performance by achieving its editing objective with as few as a single cleansed sample, which makes it appealing for practice. --- ### AgriPath: A Systematic Exploration of Architectural Trade-offs for Crop Disease Classification Authors: Hamza Mooraj, George Pantazopoulos, Alessandro Suglia Published: 2026-03-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.13354v3 Reliable crop disease detection requires models that perform consistently across diverse acquisition conditions, yet existing evaluations often focus on single architectural families or lab-generated datasets. This work presents a systematic empirical comparison of three model paradigms for fine-grained crop disease classification: Convolutional Neural Networks (CNNs), contrastive Vision-Language Models (VLMs), and generative VLMs. To enable controlled analysis of domain effects, we introduce AgriPath-LF16, a benchmark of 111k images spanning 16 crops and 41 diseases with explicit separation between laboratory and field imagery, alongside a balanced 30k subset for standardised training and evaluation. We train and evaluate all models under unified protocols across full, lab-only, and field-only training regimes using macro-F1 and Parse Success Rate (PSR) to account for generative reliability (i.e., output parsability measured via PSR). The results reveal distinct performance profiles: CNNs achieve the highest accuracy on in-domain imagery but exhibit pronounced degradation under domain shift; contrastive VLMs provide a robust and parameter-efficient alternative with competitive cross-domain performance; generative VLMs demonstrate the strongest resilience to distributional variation, albeit with additional failure modes stemming from free-text generation. These findings highlight that architectural choice should be guided by deployment context rather than aggregate performance alone. --- ### Optimizing LLM Annotation of Classroom Discourse through Multi-Agent Orchestration Authors: Bakhtawar Ahtisham, Kirk Vanacore, Rene F. Kizilcec Published: 2026-03-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.13353v1 Large language models (LLMs) are increasingly positioned as scalable tools for annotating educational data, including classroom discourse, interaction logs, and qualitative learning artifacts. Their ability to rapidly summarize instructional interactions and assign rubric-aligned labels has fueled optimism about reducing the cost and time associated with expert human annotation. However, growing evidence suggests that single-pass LLM outputs remain unreliable for high-stakes educational constructs that require contextual, pedagogical, or normative judgment, such as instructional intent or discourse moves. This tension between scale and validity sits at the core of contemporary education data science. In this work, we present and empirically evaluate a hierarchical, cost-aware orchestration framework for LLM-based annotation that improves reliability while explicitly modeling computational tradeoffs. Rather than treating annotation as a one-shot prediction problem, we conceptualize it as a multi-stage epistemic process comprising (1) an unverified single-pass annotation stage, in which models independently assign labels based on the rubric; (2) a self-verification stage, in which each model audits its own output against rubric definitions and revises its label if inconsistencies are detected; and (3) a disagreement-centric adjudication stage, in which an independent adjudicator model examines the verified labels and justifications and determines a final label in accordance with the rubric. This structure mirrors established human annotation workflows in educational research, where initial coding is followed by self-checking and expert resolution of disagreements. --- ### Local Precise Refinement: A Dual-Gated Mixture-of-Experts for Enhancing Foundation Model Generalization against Spectral Shifts Authors: Xi Chen, Maojun Zhang, Yu Liu, Shen Yan Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.13352v3 Domain Generalization Semantic Segmentation (DGSS) in spectral remote sensing is severely challenged by spectral shifts across diverse acquisition conditions, which cause significant performance degradation for models deployed in unseen domains. While fine-tuning foundation models is a promising direction, existing methods employ global, homogeneous adjustments. This "one-size-fits-all" tuning struggles with the spatial heterogeneity of land cover, causing semantic confusion. We argue that the key to robust DGSS lies not in a single global adaptation, but in performing fine-grained, spatially-adaptive refinement of a foundation model's features. To achieve this, we propose SpectralMoE, a novel fine-tuning framework for DGSS. It operationalizes this principle by utilizing a Mixture-of-Experts (MoE) architecture to perform \textbf{local precise refinement} on the foundation model's features, incorporating depth features estimated from selected RGB bands of the spectral remote sensing imagery to guide the fine-tuning process. Specifically, SpectralMoE employs a dual-gated MoE architecture that independently routes visual and depth features to top-k selected experts for specialized refinement, enabling modality-specific adjustments. A subsequent cross-attention mechanism then judiciously fuses the refined structural cues into the visual stream, mitigating semantic ambiguities caused by spectral variations. Extensive experiments show that SpectralMoE sets a new state-of-the-art on multiple DGSS benchmarks across hyperspectral, multispectral, and RGB remote sensing imagery. --- ### From Garbage to Gold: A Data-Architectural Theory of Predictive Robustness Authors: Terrence J. Lee-St. John, Jordan L. Lawson, Bartlomiej Piechowski-Jozwiak Published: 2026-03-08 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2603.12288v1 Tabular machine learning presents a paradox: modern models achieve state-of-the-art performance using high-dimensional (high-D), collinear, error-prone data, defying the "Garbage In, Garbage Out" mantra. To help resolve this, we synthesize principles from Information Theory, Latent Factor Models, and Psychometrics, clarifying that predictive robustness arises not solely from data cleanliness, but from the synergy between data architecture and model capacity. Partitioning predictor-space "noise" into "Predictor Error" and "Structural Uncertainty" (informational deficits from stochastic generative mappings), we prove that leveraging high-D sets of error-prone predictors asymptotically overcomes both types of noise, whereas cleaning a low-D set is fundamentally bounded by Structural Uncertainty. We demonstrate why "Informative Collinearity" (dependencies from shared latent causes) enhances reliability and convergence efficiency, and explain why increased dimensionality reduces the latent inference burden, enabling feasibility with finite samples. To address practical constraints, we propose "Proactive Data-Centric AI" to identify predictors that enable robustness efficiently. We also derive boundaries for Systematic Error Regimes and show why models that absorb "rogue" dependencies can mitigate assumption violations. Linking latent architecture to Benign Overfitting, we offer a first step towards a unified view of robustness to Outcome Error and predictor-space noise, while also delineating when traditional DCAI's focus on label cleaning remains powerful. By redefining data quality from item-level perfection to portfolio-level architecture, we provide a theoretical rationale for "Local Factories" -- learning from live, uncurated enterprise "data swamps" -- supporting a deployment paradigm shift from "Model Transfer" to "Methodology Transfer'' to overcome static generalizability limitations. --- ### Context-Enriched Natural Language Descriptions of Vessel Trajectories Authors: Kostas Patroumpas, Alexandros Troupiotis-Kapeliaris, Giannis Spiliopoulos, Panagiotis Betchavas, Dimitrios Skoutas, Dimitris Zissis, Nikos Bikakis Published: 2026-03-08 Categories: cs.AI, cs.CL, cs.DB arXiv: https://arxiv.org/abs/2603.12287v1 We address the problem of transforming raw vessel trajectory data collected from AIS into structured and semantically enriched representations interpretable by humans and directly usable by machine reasoning systems. We propose a context-aware trajectory abstraction framework that segments noisy AIS sequences into distinct trips each consisting of clean, mobility-annotated episodes. Each episode is further enriched with multi-source contextual information, such as nearby geographic entities, offshore navigation features, and weather conditions. Crucially, such representations can support generation of controlled natural language descriptions using LLMs. We empirically examine the quality of such descriptions generated using several LLMs over AIS data along with open contextual features. By increasing semantic density and reducing spatiotemporal complexity, this abstraction can facilitate downstream analytics and enable integration with LLMs for higher-level maritime reasoning tasks. --- ### Toward Epistemic Stability: Engineering Consistent Procedures for Industrial LLM Hallucination Reduction Authors: Brian Freeman, Adam Kicklighter, Matt Erdman, Zach Gordon Published: 2026-03-08 Categories: cs.SE, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2603.10047v2 Hallucinations in large language models (LLMs) are outputs that are syntactically coherent but factually incorrect or contextually inconsistent. They are persistent obstacles in high-stakes industrial settings such as engineering design, enterprise resource planning, and IoT telemetry platforms. We present and compare five prompt engineering strategies intended to reduce the variance of model outputs and move toward repeatable, grounded results without modifying model weights or creating complex validation models. These methods include: (M1) Iterative Similarity Convergence, (M2) Decomposed Model-Agnostic Prompting, (M3) Single-Task Agent Specialization, (M4) Enhanced Data Registry, and (M5) Domain Glossary Injection. Each method is evaluated against an internal baseline using an LLM-as-Judge framework over 100 repeated runs per method (same fixed task prompt, stochastic decoding at tau = 0.7. Under this evaluation setup, M4 (Enhanced Data Registry) received ``Better'' verdicts in all 100 trials; M3 and M5 reached 80% and 77% respectively; M1 reached 75%; and M2 was net negative at 34% when compared to single shot prompting with a modern foundation model. We then developed enhanced version 2 (v2) implementations and assessed them on a 10-trial verification batch; M2 recovered from 34% to 80%, the largest gain among the four revised methods. We discuss how these strategies help overcome the non-deterministic nature of LLM results for industrial procedures, even when absolute correctness cannot be guaranteed. We provide pseudocode, verbatim prompts, and batch logs to support independent assessment. --- ### Gated Adaptation for Continual Learning in Human Activity Recognition Authors: Reza Rahimi Azghan, Gautham Krishna Gudur, Mohit Malu, Edison Thomaz, Giulia Pedrielli, Pavan Turaga, Hassan Ghasemzadeh Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.10046v2 Wearable sensors in Internet of Things (IoT) ecosystems increasingly support applications such as remote health monitoring, elderly care, and smart home automation, all of which rely on robust human activity recognition (HAR). Continual learning systems must balance plasticity (learning new tasks) with stability (retaining prior knowledge), yet AI models often exhibit catastrophic forgetting, where learning new tasks degrades performance on earlier ones. This challenge is especially acute in domain-incremental HAR, where on-device models must adapt to new subjects with distinct movement patterns while maintaining accuracy on prior subjects without transmitting sensitive data to the cloud. We propose a parameter-efficient continual learning framework based on channel-wise gated modulation of frozen pretrained representations. Our key insight is that adaptation should operate through feature selection rather than feature generation: by restricting learned transformations to diagonal scaling of existing features, we preserve the geometry of pretrained representations while enabling subject-specific modulation. We provide a theoretical analysis showing that gating implements a bounded diagonal operator that limits representational drift compared to unconstrained linear transformations. Empirically, freezing the backbone substantially reduces forgetting, and lightweight gates restore lost adaptation capacity, achieving stability and plasticity simultaneously. On PAMAP2 with 8 sequential subjects, our approach reduces forgetting from 39.7% to 16.2% and improves final accuracy from 56.7% to 77.7%, while training less than 2% of parameters. Our method matches or exceeds standard continual learning baselines without replay buffers or task-specific regularization, confirming that structured diagonal operators are effective and efficient under distribution shift. --- ### Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety Authors: David Gringras Published: 2026-03-08 Categories: cs.AI, cs.CL, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2603.10044v3 Safety benchmarks usually test "bare" models that receive prompts and output responses, but real-world deployments "wrap" those models in complex scaffolds. How much do these scaffolds affect model safety as measured by benchmarks? We test six leading models on four pre-registered safety benchmarks with a direct API and three scaffolds: ReAct, multi-agent, and map-reduce. We conducted 62,808 scored evaluations. How safety is measured matters more than scaffolding does: we find that using a multiple choice vs. open-ended format for otherwise-identical benchmark items changes measured safety by 5-20 percentage points (pp). The two formats are scored with different methods (answer extraction and an LLM judge), so the gap is due to measurement rather than differences in latent safety. Using a heuristic to classify model refusals would have led to different findings in five cases. Benchmark choice explains 19.3% of the variation in outcomes; scaffold architecture explains 0.4%, about 45x less. We find that map-reduce scaffolds, a form of structure-destroying delegation that strips answer options by decomposing prompts, reduce pooled measured safety by 7.3 pp (95% CI: 6.4 to 8.1). The pooled effects for ReAct and multi-agent scaffolds are within our pre-registered +/-2 pp margin of equivalence. However, there are large differences across models for specific benchmarks and scaffolds that are hidden by pooled estimates: for example, on the same sycophancy benchmark items, Opus 4.6 has 16.8 pp lower measured safety with a map-reduce scaffold, while Llama 4 has 18.8 pp higher measured safety. Composite reliability is G = 0.000 (95% CI: (0.000, 0.752)). This wide confidence interval, which spans "of little use" to "very good", does not support using a single composite measure of model safety as the basis for go/no-go decisions about model deployment. --- ### Safety Under Scaffolding: How Evaluation Conditions Shape Measured Safety Authors: David Gringras Published: 2026-03-08 Categories: cs.AI, cs.CL, cs.CY, cs.LG arXiv: https://arxiv.org/abs/2603.10044v2 A safety score earned on a benchmark need not predict how the same model behaves once it is wrapped in an agentic scaffold the benchmark never tested. We ran six frontier models through four deployment configurations (direct API, ReAct, multi-agent critic, map-reduce delegation): N = 62,808 blinded, pre-registered, equivalence-tested evaluations across four safety benchmarks (BBQ, TruthfulQA, XSTest/OR-Bench, sycophancy), plus three supporting analyses. ReAct and multi-agent scaffolds stay within a pre-registered +/-2 pp equivalence margin; map-reduce delegation degrades measured safety (NNH = 14), though that loss is largely a measurement artifact: on identical items, multiple-choice versus open-ended phrasing shifts the measured safety rate by 5-20 pp, and decomposition silently strips the multiple-choice options. Roughly 40-89% of the per-model map-reduce loss is this format conversion rather than reasoning disruption, and an option-preserving variant recovers most of it. Pooled effects also mask sharp model-by-scaffold heterogeneity: under map-reduce, on identical items, Opus loses 16.8 pp while Llama 4 gains 18.8 pp. Structurally, scaffold architecture explains only 0.4% of outcome variance (benchmark choice explains 45x more), and the generalizability coefficient is G = 0.000 (bootstrap 95% CI (0.000, 0.752)). An interval that wide is enough on its own to undermine the utility of any single composite safety number as a deployment criterion. These are the "easy cases"; consequential properties like scheming and CBRN uplift have no obvious reason to be less format- or scaffold-sensitive. Code, data, and prompts are released as ScaffoldSafety. --- ### No Certificate for Alignment: Two Independent Impossibilities and the Pareto Frontier of Achievable Safety Guarantees Authors: Ayushi Agarwal Published: 2026-03-08 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2603.08761v2 We argue that formal certification of AI alignment over open-ended or unbounded input domains is impossible under standard assumptions in computational complexity and learning theory, and characterise what remains achievable. Two structurally independent impossibility theorems support this position. The semantic barrier (Theorem 1): deciding whether a system satisfies any non-trivial alignment property over the full input domain is NP-hard for feedforward networks and undecidable for Turing-complete architectures -- a direct consequence of neural-network verification complexity and Rice's Theorem. The statistical barrier (Theorem 2): any verification procedure that is both sound and tractable cannot satisfy Completeness over the full input domain -- a direct consequence of the impossibility of certifying infinite-domain properties from finite observations. These two theorems jointly entail a trilemma: no procedure can simultaneously satisfy soundness (no misaligned system is certified), completeness (no aligned system is rejected), and tractability (polynomial runtime). Each pair is simultaneously achievable; all three are not. We combine these results as a joint framework of two structurally independent barriers, prove their independence, and characterise the achievable Pareto frontier quantitatively via a constructive coverage-gap lower bound. --- ### Clear, Compelling Arguments: Rethinking the Foundations of Frontier AI Safety Cases Authors: Shaun Feakins, Ibrahim Habli, Phillip Morgan Published: 2026-03-08 Categories: cs.CY, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2603.08760v1 This paper contributes to the nascent debate around safety cases for frontier AI systems. Safety cases are structured, defensible arguments that a system is acceptably safe to deploy in a given context. Historically, they have been used in safety-critical industries, such as aerospace, nuclear or automotive. As a result, safety cases for frontier AI have risen in prominence, both in the safety policies of leading frontier developers and in international research agendas proposed by leaders in generative AI, such as the Singapore Consensus on Global AI Safety Research Priorities and the International AI Safety Report. This paper appraises this work. We note that research conducted within the alignment community which draws explicitly on lessons from the assurance community has significant limitations. We therefore aim to rethink existing approaches to alignment safety cases. We offer lessons from existing methodologies within safety assurance and outline the limitations involved in the alignment community's current approach. Building on this foundation, we present a case study for a safety case focused on Deceptive Alignment and CBRN capabilities, drawing on existing, theoretical safety case "sketches" created by the alignment safety case community. Overall, we contribute holistic insights from the field of safety assurance via rigorous theory and methodologies that have been applied in safety-critical contexts. We do so in order to create a better foundational framework for robust, defensible and useful safety case methodologies which can help to assure the safety of frontier AI systems. --- ### EDMFormer: Genre-Specific Self-Supervised Learning for Music Structure Segmentation Authors: Sahal Sajeer, Krish Patel, Oscar Chung, Joel Song Bae Published: 2026-03-08 Categories: cs.SD, cs.AI arXiv: https://arxiv.org/abs/2603.08759v1 Music structure segmentation is a key task in audio analysis, but existing models perform poorly on Electronic Dance Music (EDM). This problem exists because most approaches rely on lyrical or harmonic similarity, which works well for pop music but not for EDM. EDM structure is instead defined by changes in energy, rhythm, and timbre, with different sections such as buildup, drop, and breakdown. We introduce EDMFormer, a transformer model that combines self-supervised audio embeddings using an EDM-specific dataset and taxonomy. We release this dataset as EDM-98: a group of 98 professionally annotated EDM tracks. EDMFormer improves boundary detection and section labelling compared to existing models, particularly for drops and buildups. The results suggest that combining learned representations with genre-specific data and structural priors is effective for EDM and could be applied to other specialized music genres or broader audio domains. --- ### Generalized Reduction to the Isotropy for Flexible Equivariant Neural Fields Authors: Alejandro García-Castellanos, Gijs Bellaard, Remco Duits, Daniel Pelt, Erik J Bekkers Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.08758v1 Many geometric learning problems require invariants on heterogeneous product spaces, i.e., products of distinct spaces carrying different group actions, where standard techniques do not directly apply. We show that, when a group $G$ acts transitively on a space $M$, any $G$-invariant function on a product space $X \times M$ can be reduced to an invariant of the isotropy subgroup $H$ of $M$ acting on $X$ alone. Our approach establishes an explicit orbit equivalence $(X \times M)/G \cong X/H$, yielding a principled reduction that preserves expressivity. We apply this characterization to Equivariant Neural Fields, extending them to arbitrary group actions and homogeneous conditioning spaces, and thereby removing the major structural constraints imposed by existing methods. --- ### Intentional Deception as Controllable Capability in LLM Agents Authors: Jason Starace, Terence Soule Published: 2026-03-08 Categories: cs.AI arXiv: https://arxiv.org/abs/2603.07848v1 As LLM-based agents increasingly operate in multi-agent systems, understanding adversarial manipulation becomes critical for defensive design. We present a systematic study of intentional deception as an engineered capability, using LLM-to-LLM interactions within a text-based RPG where parameterized behavioral profiles (9 alignments x 4 motivations, yielding 36 profiles with explicit ethical ground truth) serve as our experimental testbed. Unlike accidental deception from misalignment, we investigate a two-stage system that infers target agent characteristics and generates deceptive responses steering targets toward actions counter to their beliefs and motivations. We find that deceptive intervention produces differential effects concentrated in specific behavioral profiles rather than distributed uniformly, and that 88.5% of successful deceptions employ misdirection (true statements with strategic framing) rather than fabrication, indicating fact-checking defenses would miss the large majority of adversarial responses. Motivation, inferable at 98%+ accuracy, serves as the primary attack vector, while belief systems remain harder to identify (49% inference ceiling) or exploit. These findings identify which agent profiles require additional safeguards and suggest that current fact-verification approaches are insufficient against strategically framed deception. --- ### An Efficient and Effective Evaluator for Text2SQL Models on Unseen and Unlabeled Data Authors: Trinh Pham, Thanh Tam Nguyen, Viet Huynh, Hongzhi Yin, Quoc Viet Hung Nguyen Published: 2026-03-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.07841v2 Recent advances in large language models have strengthened Text2SQL systems that translate natural language questions into database queries. A persistent deployment challenge is to assess a newly trained Text2SQL system on an unseen and unlabeled dataset when no verified answers are available. This situation arises frequently because database content and structure evolve, privacy policies slow manual review, and carefully written SQL labels are costly and time-consuming. Without timely evaluation, organizations cannot approve releases or detect failures early. FusionSQL addresses this gap by working with any Text2SQL models and estimating accuracy without reference labels, allowing teams to measure quality on unseen and unlabeled datasets. It analyzes patterns in the system's own outputs to characterize how the target dataset differs from the material used during training. FusionSQL supports pre-release checks, continuous monitoring of new databases, and detection of quality decline. Experiments across diverse application settings and question types show that FusionSQL closely follows actual accuracy and reliably signals emerging issues. Our code is available at https://github.com/phkhanhtrinh23/FusionSQL. --- ### Training-free Temporal Object Tracking in Surgical Videos Authors: Subhadeep Koley, Abdolrahim Kadkhodamohammadi, Santiago Barbarisi, Danail Stoyanov, Imanol Luengo Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07839v1 Purpose: In this paper, we present a novel approach for online object tracking in laparoscopic cholecystectomy (LC) surgical videos, targeting localisation and tracking of critical anatomical structures and instruments. Our method addresses the challenges of costly pixel-level annotations and label inconsistencies inherent in existing datasets. Methods: Leveraging the inherent object localisation capabilities of pre-trained text-to-image diffusion models, we extract representative features from surgical frames without any training or fine-tuning. Our tracking framework uses these features, along with cross-frame interactions via an affinity matrix inspired by query-key-value attention, to ensure temporal continuity in the tracking process. Results: Through a pilot study, we first demonstrate that diffusion features exhibit superior object localisation and consistent semantics across different decoder levels and temporal frames. Later, we perform extensive experiments to validate the effectiveness of our approach, showcasing its superiority over competitors for the task of temporal object tracking. Specifically, we achieve a per-pixel classification accuracy of 79.19%, mean Jaccard Score of 56.20%, and mean F-Score of 79.48% on the publicly available CholeSeg8K dataset. Conclusion: Our work not only introduces a novel application of text-to-image diffusion models but also contributes to advancing the field of surgical video analysis, offering a promising avenue for accurate and cost-effective temporal object tracking in minimally invasive surgery videos. --- ### AI Steerability 360: A Toolkit for Steering Large Language Models Authors: Erik Miehling, Karthikeyan Natesan Ramamurthy, Praveen Venkateswaran, Irene Ko, Pierre Dognin, Moninder Singh, Tejaswini Pedapati, Avinash Balakrishnan, Matthew Riemer, Dennis Wei, Inge Vejsbjerg, Elizabeth M. Daly et al. Published: 2026-03-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2603.07837v1 The AI Steerability 360 toolkit is an extensible, open-source Python library for steering LLMs. Steering abstractions are designed around four model control surfaces: input (modification of the prompt), structural (modification of the model's weights or architecture), state (modification of the model's activations and attentions), and output (modification of the decoding or generation process). Steering methods exert control on the model through a common interface, termed a steering pipeline, which additionally allows for the composition of multiple steering methods. Comprehensive evaluation and comparison of steering methods/pipelines is facilitated by use case classes (for defining tasks) and a benchmark class (for performance comparison on a given task). The functionality provided by the toolkit significantly lowers the barrier to developing and comprehensively evaluating steering methods. The toolkit is Hugging Face native and is released under an Apache 2.0 license at https://github.com/IBM/AISteer360. --- ### DistillGuard: Evaluating Defenses Against LLM Knowledge Distillation Authors: Bo Jiang Published: 2026-03-08 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.07835v1 Knowledge distillation from proprietary LLM APIs poses a growing threat to model providers, yet defenses against this attack remain fragmented and unevaluated. We present DistillGuard, a framework for systematically evaluating output-level defenses against LLM knowledge distillation. We introduce a taxonomy of three defense categories -- output perturbation, data poisoning, and information throttling -- and evaluate nine defense configurations using a standardized pipeline with Qwen3-14B as teacher and Qwen2.5-7B-Instruct as student across three benchmarks (MATH-500, HumanEval+, MT-Bench). Our results reveal that, in a same-family distillation setting against a naive attacker, most output-level defenses are surprisingly ineffective: paraphrasing-based perturbation barely degrades distilled student quality, and data poisoning primarily impairs conversational fluency while leaving task-specific capabilities intact. Only chain-of-thought removal substantially impairs mathematical reasoning (31.4\% vs.\ 67.8\% baseline), though code generation remains unaffected. These findings demonstrate that the effectiveness of distillation defenses is highly task-dependent and that current output-level approaches are insufficient to broadly prevent knowledge theft. --- ### AI Misuse in Education Is a Measurement Problem: Toward a Learning Visibility Framework Authors: Eduardo Davalos, Yike Zhang Published: 2026-03-08 Categories: cs.CY, cs.AI, cs.HC arXiv: https://arxiv.org/abs/2603.07834v1 The rapid integration of conversational AI systems into educational settings has intensified ethical concerns about academic integrity, fairness, and students' cognitive development. Institutional responses have largely centered on AI detection tools and restrictive policies, yet such approaches have proven unreliable and ethically contentious. This paper reframes AI misuse in education not primarily as a detection problem, but as a measurement problem rooted in the loss of visibility into the learning process. When AI enters the assessment loop, educators often retain access to final outputs but lose valuable insight into how those outputs were produced. Drawing on research in cognitive offloading, learning analytics, and multimodal timeline reconstruction, we propose the Learning Visibility Framework, grounded in three principles: clear specification and modeling of acceptable AI use, recognition of learning processes as assessable evidence alongside outcomes, and the establishment of transparent timelines of student activity. Rather than promoting surveillance, the framework emphasizes transparency and shared evidence as foundations for ethical AI integration in classroom settings. By shifting focus from adversarial detection toward process visibility, this work offers a principled pathway for aligning AI use with educational values while preserving trust and transparency between students and educators --- ### Gradient Iterated Temporal-Difference Learning Authors: Théo Vincent, Kevin Gerhardt, Yogesh Tripathi, Habib Maraqten, Adam White, Martha White, Jan Peters, Carlo D'Eramo Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.07833v2 Temporal-difference (TD) learning is highly effective at controlling and evaluating an agent's long-term outcomes. Most approaches in this paradigm implement a semi-gradient update to boost the learning speed, which consists of ignoring the gradient of the bootstrapped estimate. While popular, this type of update is prone to divergence, as Baird's counterexample illustrates. Gradient TD methods were introduced to overcome this issue, but have not been widely used, potentially due to issues with learning speed compared to semi-gradient methods. Recently, iterated TD learning was developed to increase the learning speed of TD methods. For that, it learns a sequence of action-value functions in parallel, where each function is optimized to represent the application of the Bellman operator over the previous function in the sequence. While promising, this algorithm can be unstable due to its semi-gradient nature, as each function tracks a moving target. In this work, we modify iterated TD learning by computing the gradients over those moving targets, aiming to build a powerful gradient TD method that competes with semi-gradient methods. Our evaluation reveals that this algorithm, called Gradient Iterated Temporal-Difference learning, has a competitive learning speed against semi-gradient methods across various benchmarks, including Atari games, a result that no prior work on gradient TD methods has demonstrated. --- ### GazeShift: Unsupervised Gaze Estimation and Dataset for VR Authors: Gil Shapira, Ishay Goldin, Evgeny Artyomov, Donghoon Kim, Yosi Keller, Niv Zehngut Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07832v3 Gaze estimation is instrumental in modern virtual reality (VR) systems. Despite significant progress in remote-camera gaze estimation, VR gaze research remains constrained by data scarcity, particularly the lack of large-scale, accurately labeled datasets captured with the off-axis camera configurations typical of modern headsets. Gaze annotation is difficult since fixation on intended targets cannot be guaranteed. To address these challenges, we introduce VRGaze, the first large-scale off-axis gaze estimation dataset for VR, comprising 2.1 million near-eye infrared images collected from 68 participants. We further propose GazeShift, an attention-guided unsupervised framework for learning gaze representations without labeled data. Unlike prior redirection-based methods that rely on multi-view or 3D geometry, GazeShift is tailored to near-eye imagery, achieving effective gaze-appearance disentanglement in a compact, real-time model. GazeShift embeddings can be optionally adapted to individual users via lightweight few-shot calibration, achieving a 1.84° mean error on VRGaze. On the remote-camera MPIIGaze dataset, the model achieves a 7.15° person-agnostic error, doing so with 10x fewer parameters and 35x fewer FLOPs than baseline methods. Deployed natively on a VR headset GPU, inference takes only 5 ms. Combined with demonstrated robustness to illumination changes, these results highlight GazeShift as a label-efficient, real-time solution for VR gaze tracking. Project code and the VRGaze dataset are released at https://github.com/gazeshift3/gazeshift --- ### Transferable Optimization Network for Cross-Domain Image Reconstruction Authors: Yunmei Chen, Chi Ding, Xiaojing Ye Published: 2026-03-08 Categories: cs.CV, cs.LG, math.OC arXiv: https://arxiv.org/abs/2603.07831v1 We develop a novel transfer learning framework to tackle the challenge of limited training data in image reconstruction problems. The proposed framework consists of two training steps, both of which are formed as bi-level optimizations. In the first step, we train a powerful universal feature-extractor that is capable of learning important knowledge from large, heterogeneous data sets in various domains. In the second step, we train a task-specific domain-adapter for a new target domain or task with only a limited amount of data available for training. Then the composition of the adapter and the universal feature-extractor effectively explores feature which serve as an important component of image regularization for the new domains, and this leads to high-quality reconstruction despite the data limitation issue. We apply this framework to reconstruct under-sampled MR images with limited data by using a collection of diverse data samples from different domains, such as images of other anatomies, measurements of various sampling ratios, and even different image modalities, including natural images. Experimental results demonstrate a promising transfer learning capability of the proposed method. --- ### Benchmarking Large Language Models for Quebec Insurance: From Closed-Book to Retrieval-Augmented Generation Authors: David Beauchemin, Richard Khoury Published: 2026-03-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.07825v1 The digitization of insurance distribution in the Canadian province of Quebec, accelerated by legislative changes such as Bill 141, has created a significant "advice gap", leaving consumers to interpret complex financial contracts without professional guidance. While Large Language Models (LLMs) offer a scalable solution for automated advisory services, their deployment in high-stakes domains hinges on strict legal accuracy and trustworthiness. In this paper, we address this challenge by introducing AEPC-QA, a private gold-standard benchmark of 807 multiple-choice questions derived from official regulatory certification (paper) handbooks. We conduct a comprehensive evaluation of 51 LLMs across two paradigms: closed-book generation and retrieval-augmented generation (RAG) using a specialized corpus of Quebec insurance documents. Our results reveal three critical insights: 1) the supremacy of inference-time reasoning, where models leveraging chain-of-thought processing (e.g. o3-2025-04-16, o1-2024-12-17) significantly outperform standard instruction-tuned models; 2) RAG acts as a knowledge equalizer, boosting the accuracy of models with weak parametric knowledge by over 35 percentage points, yet paradoxically causing "context distraction" in others, leading to catastrophic performance regressions; and 3) a "specialization paradox", where massive generalist models consistently outperform smaller, domain-specific French fine-tuned ones. These findings suggest that while current architectures approach expert-level proficiency (~79%), the instability introduced by external context retrieval necessitates rigorous robustness calibration before autonomous deployment is viable. --- ### Column Generation for the Micro-Transit Zoning Problem Authors: Hins Hu, Rishav Sen, Jose Paolo Talusan, Abhishek Dubey, Aron Laszka, Samitha Samaranayake Published: 2026-03-08 Categories: math.OC, cs.AI arXiv: https://arxiv.org/abs/2603.07821v1 Along with the rapid development of new urban mobility options like ride-sharing over the past decade, on-demand micro-transit services stand out as a middle ground, bridging the gap between fixed-line mass transit and single-request ride-hailing, balancing ridership maximization and travel time minimization. Micro-transit adoption can have significant social impact. It improves urban sustainability, through lower energy consumption and reduced emissions, while enhancing equitable mobility access for disadvantaged communities, thanks to its lower vehicle miles per passenger, flexible schedules, and affordable pricing. However, effective operation of micro-transit services requires planning geo-fenced zones in advance, which involves solving a challenging combinatorial optimization problem. Existing approaches enumerate candidate zones first and selects a fixed number of optimal zones in the second step. In this paper, we generalize the Micro-Transit Zoning Problem (MZP) to allow a global budget rather than imposing a size limit for candidate zones. We also design a Column Generation (CG) framework to solve the problem and several pricing heuristics to accelerate computation. Extensive numerical experiments across major U.S. cities demonstrate that our approach produces higher-quality solutions more efficiently and scales better in the generalized setting. --- ### Fusion Complexity Inversion: Why Simpler Cross View Modules Outperform SSMs and Cross View Attention Transformers for Pasture Biomass Regression Authors: Mridankan Mandal Published: 2026-03-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2603.07819v5 Accurate estimation of pasture biomass from agricultural imagery is critical for sustainable livestock management, yet existing methods are limited by the small, imbalanced, and sparsely annotated datasets typical of real world monitoring. In this study, adaptation of vision foundation models to agricultural regression is systematically evaluated on the CSIRO Pasture Biomass benchmark, a 357 image dual view dataset with laboratory validated, component wise ground truth for five biomass targets, through 17 configurations spanning four backbones (EfficientNet-B3 to DINOv3-ViT-L), five cross view fusion mechanisms, and a 4x2 metadata factorial. A counterintuitive principle, termed "fusion complexity inversion", is uncovered: on scarce agricultural data, a two layer gated depthwise convolution (R^2 = 0.903) outperforms cross view attention transformers (0.833), bidirectional SSMs (0.819), and full Mamba (0.793, below the no fusion baseline). Backbone pretraining scale is found to monotonically dominate all architectural choices, with the DINOv2 -> DINOv3 upgrade alone yielding +5.0 R^2 points. Training only metadata (species, state, and NDVI) is shown to create a universal ceiling at R^2 ~ 0.829, collapsing an 8.4 point fusion spread to 0.1 points. Actionable guidelines for sparse agricultural benchmarks are established: backbone quality should be prioritized over fusion complexity, local modules preferred over global alternatives, and features unavailable at inference excluded. --- ### Tracking Phenological Status and Ecological Interactions in a Hawaiian Cloud Forest Understory using Low-Cost Camera Traps and Visual Foundation Models Authors: Luke Meyers, Anirudh Potlapally, Yuyan Chen, Mike Long, Tanya Berger-Wolf, Hari Subramoni, Remi Megret, Daniel Rubenstein Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07817v1 Plant phenology, the study of cyclical events such as leafing out, flowering, or fruiting, has wide ecological impacts but is broadly understudied, especially in the tropics. Image analysis has greatly enhanced remote phenological monitoring, yet capturing phenology at the individual level remains challenging. In this project, we deployed low-cost, animal-triggered camera traps at the Pu'u Maka'ala Natural Area Reserve in Hawaii to simultaneously document shifts in plant phenology and flora-faunal interactions. Using a combination of foundation vision models and traditional computer vision methods, we measure phenological trends from images comparable to on-the-ground observations without relying on supervised learning techniques. These temporally fine-grained phenology measurements from camera-trap images uncover trends that coarser traditional sampling fails to detect. When combined with detailed visitation data detected from images, these trends can begin to elucidate drivers of both plant phenology and animal ecology. --- ### HybridStitch: Pixel and Timestep Level Model Stitching for Diffusion Acceleration Authors: Desen Sun, Jason Hon, Jintao Zhang, Sihang Liu Published: 2026-03-08 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2603.07815v1 Diffusion models have demonstrated a remarkable ability in Text-to-Image (T2I) generation applications. Despite the advanced generation output, they suffer from heavy computation overhead, especially for large models that contain tens of billions of parameters. Prior work has illustrated that replacing part of the denoising steps with a smaller model still maintains the generation quality. However, these methods only focus on saving computation for some timesteps, ignoring the difference in compute demand within one timestep. In this work, we propose HybridStitch, a new T2I generation paradigm that treats generation like editing. Specifically, we introduce a hybrid stage that jointly incorporates both the large model and the small model. HybridStitch separates the entire image into two regions: one that is relatively easy to render, enabling an early transition to the smaller model, and another that is more complex and therefore requires refinement by the large model. HybridStitch employs the small model to construct a coarse sketch while exploiting the large model to edit and refine the complex regions. According to our evaluation, HybridStitch achieves 1.83$\times$ speedup on Stable Diffusion 3, which is faster than all existing mixture of model methods. --- ### Learning embeddings of non-linear PDEs: the Burgers' equation Authors: Pedro Tarancón-Álvarez, Leonid Sarieddine, Pavlos Protopapas, Raul Jimenez Published: 2026-03-08 Categories: math.AP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2603.07812v1 Embeddings provide low-dimensional representations that organize complex function spaces and support generalization. They provide a geometric representation that supports efficient retrieval, comparison, and generalization. In this work we generalize the concept to Physics Informed Neural Networks. We present a method to construct solution embedding spaces of nonlinear partial differential equations using a multi-head setup, and extract non-degenerate information from them using principal component analysis (PCA). We test this method by applying it to viscous Burgers' equation, which is solved simultaneously for a family of initial conditions and values of the viscosity. A shared network body learns a latent embedding of the solution space, while linear heads map this embedding to individual realizations. By enforcing orthogonality constraints on the heads, we obtain a principal-component decomposition of the latent space that is robust to training degeneracies and admits a direct physical interpretation. The obtained components for Burgers' equation exhibit rapid saturation, indicating that a small number of latent modes captures the dominant features of the dynamics. --- ### Neural Precoding in Complex Projective Spaces Authors: Zaid Abdullah, Merouane Debbah, Symeon Chatzinotas, Bjorn Ottersten Published: 2026-03-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07811v1 Deep-learning (DL)-based precoding in multi-user multiple-input single-output (MU-MISO) systems involves training DL models to map features derived from channel coefficients to labels derived from precoding weights. Traditionally, complex-valued channel and precoder coefficients are parameterized using either their real and imaginary components or their amplitude and phase. However, precoding performance depends on magnitudes of inner products between channel and precoding vectors, which are invariant to global phase rotations. Conventional representations fail to exploit this symmetry, leading to inefficient learning and degraded generalization. To address this, we propose a DL framework based on complex projective space (CPS) parameterizations of both the wireless channel and the weighted minimum mean squared error (WMMSE) precoder vectors. By removing the global phase redundancies inherent in conventional representations, the proposed framework enables the DL model to learn geometry-aligned and physically distinct channel-precoder mappings. Two CPS parameterizations based on real-valued embeddings and complex hyperspherical coordinates are investigated and benchmarked against two baseline methods. Simulation results demonstrate substantial improvements in sum-rate performance and generalization, with negligible increase in model complexity. --- ### MWM: Mobile World Models for Action-Conditioned Consistent Prediction Authors: Han Yan, Zishang Xiang, Zeyu Zhang, Hao Tang Published: 2026-03-08 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2603.07799v1 World models enable planning in imagined future predicted space, offering a promising framework for embodied navigation. However, existing navigation world models often lack action-conditioned consistency, so visually plausible predictions can still drift under multi-step rollout and degrade planning. Moreover, efficient deployment requires few-step diffusion inference, but existing distillation methods do not explicitly preserve rollout consistency, creating a training-inference mismatch. To address these challenges, we propose MWM, a mobile world model for planning-based image-goal navigation. Specifically, we introduce a two-stage training framework that combines structure pretraining with Action-Conditioned Consistency (ACC) post-training to improve action-conditioned rollout consistency. We further introduce Inference-Consistent State Distillation (ICSD) for few-step diffusion distillation with improved rollout consistency. Our experiments on benchmark and real-world tasks demonstrate consistent gains in visual fidelity, trajectory accuracy, planning success, and inference efficiency. Code: https://github.com/AIGeeksGroup/MWM. Website: https://aigeeksgroup.github.io/MWM. --- ### Toward Global Intent Inference for Human Motion by Inverse Reinforcement Learning Authors: Sarmad Mehrdad, Maxime Sabbah, Vincent Bonnet, Ludovic Righetti Published: 2026-03-08 Categories: cs.RO, cs.LG arXiv: https://arxiv.org/abs/2603.07797v1 This paper investigates whether a single, unified cost function can explain and predict human reaching movements, in contrast with existing approaches that rely on subject- or posture-specific optimization criteria. Using the Minimal Observation Inverse Reinforcement Learning (MO-IRL) algorithm, together with a seven-dimensional set of candidate cost terms, we efficiently estimate time-varying cost weights for a standard planar reaching task. MO-IRL provides orders-of-magnitude faster convergence than bilevel formulations, while using only a fraction of the available data, enabling the practical exploration of time-varying cost structures. Three levels of generality are evaluated: Subject-Dependent Posture-Dependent, Subject-Dependent Posture-Independent, and Subject-Independent Posture-Independent. Across all cases, time-varying weights substantially improve trajectory reconstruction, yielding an average 27% reduction in RMSE compared to the baseline. The inferred costs consistently highlight a dominant role for joint-acceleration regulation, complemented by smaller contributions from torque-change smoothness. Overall, a single subject- and posture-agnostic time-varying cost function is shown to predict human reaching trajectories with high accuracy, supporting the existence of a unified optimality principle governing this class of movements. --- ### 4DRC-OCC: Robust Semantic Occupancy Prediction Through Fusion of 4D Radar and Camera Authors: David Ninfa, Andras Palffy, Holger Caesar Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07794v1 Autonomous driving requires robust perception across diverse environmental conditions, yet 3D semantic occupancy prediction remains challenging under adverse weather and lighting. In this work, we present the first study combining 4D radar and camera data for 3D semantic occupancy prediction. Our fusion leverages the complementary strengths of both modalities: 4D radar provides reliable range, velocity, and angle measurements in challenging conditions, while cameras contribute rich semantic and texture information. We further show that integrating depth cues from camera pixels enables lifting 2D images to 3D, improving scene reconstruction accuracy. Additionally, we introduce a fully automatically labeled dataset for training semantic occupancy models, substantially reducing reliance on costly manual annotation. Experiments demonstrate the robustness of 4D radar across diverse scenarios, highlighting its potential to advance autonomous vehicle perception. --- ### Dual-Metric Evaluation of Social Bias in Large Language Models: Evidence from an Underrepresented Nepali Cultural Context Authors: Ashish Pandey, Tek Raj Chhetri Published: 2026-03-08 Categories: cs.CL, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2603.07792v1 Large language models (LLMs) increasingly influence global digital ecosystems, yet their potential to perpetuate social and cultural biases remains poorly understood in underrepresented contexts. This study presents a systematic analysis of representational biases in seven state-of-the-art LLMs: GPT-4o-mini, Claude-3-Sonnet, Claude-4-Sonnet, Gemini-2.0-Flash, Gemini-2.0-Lite, Llama-3-70B, and Mistral-Nemo in the Nepali cultural context. Using Croissant-compliant dataset of 2400+ stereotypical and anti-stereotypical sentence pairs on gender roles across social domains, we implement an evaluation framework, Dual-Metric Bias Assessment (DMBA), combining two metrics: (1) agreement with biased statements and (2) stereotypical completion tendencies. Results show models exhibit measurable explicit agreement bias, with mean bias agreement ranging from 0.36 to 0.43 across decoding configurations, and an implicit completion bias rate of 0.740-0.755. Importantly, implicit completion bias follows a non-linear, U-shaped relationship with temperature, peaking at moderate stochasticity (T=0.3) and declining slightly at higher temperatures. Correlation analysis under different decoding settings revealed that explicit agreement strongly aligns with stereotypical sentence agreement but is a weak and often negative predictor of implicit completion bias, indicating generative bias is poorly captured by agreement metrics. Sensitivity analysis shows increasing top-p amplifies explicit bias, while implicit generative bias remains largely stable. Domain-level analysis shows implicit bias is strongest for race and sociocultural stereotypes, while explicit agreement bias is similar across gender and sociocultural categories, with race showing the lowest explicit agreement. These findings highlight the need for culturally grounded datasets and debiasing strategies for LLMs in underrepresented societies. --- ### SGI: Structured 2D Gaussians for Efficient and Compact Large Image Representation Authors: Zixuan Pan, Kaiyuan Tang, Jun Xia, Yifan Qin, Lin Gu, Chaoli Wang, Jianxu Chen, Yiyu Shi Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07789v2 2D Gaussian Splatting has emerged as a novel image representation technique that can support efficient rendering on low-end devices. However, scaling to high-resolution images requires optimizing and storing millions of unstructured Gaussian primitives independently, leading to slow convergence and redundant parameters. To address this, we propose Structured Gaussian Image (SGI), a compact and efficient framework for representing high-resolution images. SGI decomposes a complex image into multi-scale local spaces defined by a set of seeds. Each seed corresponds to a spatially coherent region and, together with lightweight multi-layer perceptrons (MLPs), generates structured implicit 2D neural Gaussians. This seed-based formulation imposes structural regularity on otherwise unstructured Gaussian primitives, which facilitates entropy-based compression at the seed level to reduce the total storage. However, optimizing seed parameters directly on high-resolution images is a challenging and non-trivial task. Therefore, we designed a multi-scale fitting strategy that refines the seed representation in a coarse-to-fine manner, substantially accelerating convergence. Quantitative and qualitative evaluations demonstrate that SGI achieves up to 7.5x compression over prior non-quantized 2D Gaussian methods and 1.6x over quantized ones, while also delivering 1.6x and 6.5x faster optimization, respectively, without degrading, and often improving, image fidelity. Code is available at https://github.com/zx-pan/SGI. --- ### Vision Transformers that Never Stop Learning Authors: Caihao Sun, Mingqi Yuan, Shiyuan Wang, Jiayu Chen Published: 2026-03-08 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.07787v1 Loss of plasticity refers to the progressive inability of a model to adapt to new tasks and poses a fundamental challenge for continual learning. While this phenomenon has been extensively studied in homogeneous neural architectures, such as multilayer perceptrons, its mechanisms in structurally heterogeneous, attention-based models such as Vision Transformers (ViTs) remain underexplored. In this work, we present a systematic investigation of loss of plasticity in ViTs, including a fine-grained diagnosis using local metrics that capture parameter diversity and utilization. Our analysis reveals that stacked attention modules exhibit increasing instability that exacerbates plasticity loss, while feed-forward network modules suffer even more pronounced degradation. Furthermore, we evaluate several approaches for mitigating plasticity loss. The results indicate that methods based on parameter re-initialization fail to recover plasticity in ViTs, whereas approaches that explicitly regulate the update process are more effective. Motivated by this insight, we propose ARROW, a geometry-aware optimizer that preserves plasticity by adaptively reshaping gradient directions using an online curvature estimate for the attention module. Extensive experiments show that ARROW effectively improves plasticity and maintains better performance on newly encountered tasks. --- ### OrdinalBench: A Benchmark Dataset for Diagnosing Generalization Limits in Ordinal Number Understanding of Vision-Language Models Authors: Yusuke Tozaki, Hisashi Miyamori Published: 2026-03-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.07786v1 Vision-Language Models (VLMs) have advanced across multimodal benchmarks but still show clear gaps in ordinal number understanding, i.e., the ability to track relative positions and generalize to large indices. We present OrdinalBench, a diagnostic benchmark that standardizes ordinal number understanding as an evaluation task for VLMs. The core task is N-th object identification, defined by a starting reference and traversal rule. Task difficulty is controlled along three axes: (i) ordinal magnitude, from small numbers to extreme cases up to 300; (ii) arrangement complexity, from single loops to maze-like paths; and (iii) object count. The benchmark provides 39,000 question-answer pairs, each annotated with a ground-truth reasoning trajectory and balanced across difficulty levels for controlled large-scale testing. Beyond answer-only evaluation, our framework requires models to generate structured stepwise traces of the counting process and provides an open evaluation toolkit that measures both final accuracy and step-level path consistency. Zero-shot evaluations of GPT-5, Gemini 2.5 Flash Lite, Qwen2.5-VL, InternVL3.5, and Molmo reveal sharp degradation under large-ordinal and complex-path conditions, highlighting weak generalization despite strong scores on standard multimodal tasks. By framing ordinal number understanding as a core target, OrdinalBench provides a reproducible benchmark and diagnostic framework for developing VLMs with stronger sequential reasoning. All data and code are available at https://ordinalbench.github.io/ --- ### ProgAgent:A Continual RL Agent with Progress-Aware Rewards Authors: Jinzhou Tan, Gabriel Adineera, Jinoh Kim Published: 2026-03-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2603.07784v1 We present ProgAgent, a continual reinforcement learning (CRL) agent that unifies progress-aware reward learning with a high-throughput, JAX-native system architecture. Lifelong robotic learning grapples with catastrophic forgetting and the high cost of reward specification. ProgAgent tackles these by deriving dense, shaped rewards from unlabeled expert videos through a perceptual model that estimates task progress across initial, current, and goal observations. We theoretically interpret this as a learned state-potential function, delivering robust guidance in line with expert behaviors. To maintain stability amid online exploration - where novel, out-of-distribution states arise - we incorporate an adversarial push-back refinement that regularizes the reward model, curbing overconfident predictions on non-expert trajectories and countering distribution shift. By embedding this reward mechanism into a JIT-compiled loop, ProgAgent supports massively parallel rollouts and fully differentiable updates, rendering a sophisticated unified objective feasible: it merges PPO with coreset replay and synaptic intelligence for an enhanced stability-plasticity balance. Evaluations on ContinualBench and Meta-World benchmarks highlight ProgAgent's advantages: it markedly reduces forgetting, boosts learning speed, and outperforms key baselines in visual reward learning (e.g., Rank2Reward, TCN) and continual learning (e.g., Coreset, SI) - surpassing even an idealized perfect memory agent. Real-robot trials further validate its ability to acquire complex manipulation skills from noisy, few-shot human demonstrations. ---