# Research Papers — 2026-03 · bundle page 207 of 243

> Full-text markdown of papers #10301–10350 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/207
[← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/206.md)
[Next bundle →](https://aipod.dev/papers/archive/2026-03/page/208.md)

### Generic Camera Calibration using Blurry Images

Authors: Zezhun Shi
Published: 2026-03-05
Categories: cs.CV, eess.IV
arXiv: https://arxiv.org/abs/2603.05159v1

Camera calibration is the foundation of 3D vision. Generic camera calibration can yield more accurate results than parametric cam era calibration. However, calibrating a generic camera model using printed calibration boards requires far more images than parametric calibration, making motion blur practically unavoidable for individual users. As a f irst attempt to address this problem, we draw on geometric constraints and a local parametric illumination model to simultaneously estimate feature locations and spatially varying point spread functions, while re solving the translational ambiguity that need not be considered in con ventional image deblurring tasks. Experimental results validate the effectiveness of our approach.

---

### Balancing Privacy-Quality-Efficiency in Federated Learning through Round-Based Interleaving of Protection Techniques

Authors: Yenan Wang, Carla Fabiana Chiasserini, Elad Michael Schiller
Published: 2026-03-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.05158v1

In federated learning (FL), balancing privacy protection, learning quality, and efficiency remains a challenge. Privacy protection mechanisms, such as Differential Privacy (DP), degrade learning quality, or, as in the case of Homomorphic Encryption (HE), incur substantial system overhead. To address this, we propose Alt-FL, a privacy-preserving FL framework that combines DP, HE, and synthetic data via a novel round-based interleaving strategy. Alt-FL introduces three new methods, Privacy Interleaving (PI), Synthetic Interleaving with DP (SI/DP), and Synthetic Interleaving with HE (SI/HE), that enable flexible quality-efficiency trade-offs while providing privacy protection. We systematically evaluate Alt-FL against representative reconstruction attacks, including Deep Leakage from Gradients, Inverting Gradients, When the Curious Abandon Honesty, and Robbing the Fed, using a LeNet-5 model on CIFAR-10 and Fashion-MNIST. To enable fair comparison between DP- and HE-based defenses, we introduce a new attacker-centric framework that compares empirical attack success rates across the three proposed interleaving methods. Our results show that, for the studied attacker model and dataset, PI achieves the most balanced trade-offs at high privacy protection levels, while DP-based methods are preferable at intermediate privacy requirements. We also discuss how such results can be the basis for selecting privacy-preserving FL methods under varying privacy and resource constraints.

---

### The Impact of Preprocessing Methods on Racial Encoding and Model Robustness in CXR Diagnosis

Authors: Dishantkumar Sutariya, Eike Petersen
Published: 2026-03-05
Categories: cs.CV, cs.LG, eess.IV
arXiv: https://arxiv.org/abs/2603.05157v1

Deep learning models can identify racial identity with high accuracy from chest X-ray (CXR) recordings. Thus, there is widespread concern about the potential for racial shortcut learning, where a model inadvertently learns to systematically bias its diagnostic predictions as a function of racial identity. Such racial biases threaten healthcare equity and model reliability, as models may systematically misdiagnose certain demographic groups. Since racial shortcuts are diffuse - non-localized and distributed throughout the whole CXR recording - image preprocessing methods may influence racial shortcut learning, yet the potential of such methods for reducing biases remains underexplored. Here, we investigate the effects of image preprocessing methods including lung masking, lung cropping, and Contrast Limited Adaptive Histogram Equalization (CLAHE). These approaches aim to suppress spurious cues encoding racial information while preserving diagnostic accuracy. Our experiments reveal that simple bounding box-based lung cropping can be an effective strategy for reducing racial shortcut learning while maintaining diagnostic model performance, bypassing frequently postulated fairness-accuracy trade-offs.

---

### SSR-GS: Separating Specular Reflection in Gaussian Splatting for Glossy Surface Reconstruction

Authors: Ningjing Fan, Yiqun Wang
Published: 2026-03-05
Categories: cs.CV, cs.AI, cs.GR
arXiv: https://arxiv.org/abs/2603.05152v1

In recent years, 3D Gaussian splatting (3DGS) has achieved remarkable progress in novel view synthesis. However, accurately reconstructing glossy surfaces under complex illumination remains challenging, particularly in scenes with strong specular reflections and multi-surface interreflections. To address this issue, we propose SSR-GS, a specular reflection modeling framework for glossy surface reconstruction. Specifically, we introduce a prefiltered Mip-Cubemap to model direct specular reflections efficiently, and propose an IndiASG module to capture indirect specular reflections. Furthermore, we design Visual Geometry Priors (VGP) that couple a reflection-aware visual prior via a reflection score (RS) to downweight the photometric loss contribution of reflection-dominated regions, with geometry priors derived from VGGT, including progressively decayed depth supervision and transformed normal constraints. Extensive experiments on both synthetic and real-world datasets demonstrate that SSR-GS achieves state-of-the-art performance in glossy surface reconstruction.

---

### Federated Causal Discovery Across Heterogeneous Datasets under Latent Confounding

Authors: Maximilian Hahn, Alina Zajak, Dominik Heider, Adèle Helena Ribeiro
Published: 2026-03-05
Categories: cs.LG, cs.AI, stat.ML
arXiv: https://arxiv.org/abs/2603.05149v1

Causal discovery across multiple datasets is often constrained by data privacy regulations and cross-site heterogeneity, limiting the use of conventional methods that require a single, centralized dataset. To address these challenges, we introduce fedCI, a federated conditional independence test that rigorously handles heterogeneous datasets with non-identical sets of variables, site-specific effects, and mixed variable types, including continuous, ordinal, binary, and categorical variables. At its core, fedCI uses a federated Iteratively Reweighted Least Squares (IRLS) procedure to estimate the parameters of generalized linear models underlying likelihood-ratio tests for conditional independence. Building on this, we develop fedCI-IOD, a federated extension of the Integration of Overlapping Datasets (IOD) algorithm, that replaces its meta-analysis strategy and enables, for the fist time, federated causal discovery under latent confounding across distributed and heterogeneous datasets. By aggregating evidence federatively, fedCI-IOD not only preserves privacy but also substantially enhances statistical power, achieving performance comparable to fully pooled analyses and mitigating artifacts from low local sample sizes. Our tools are publicly available as the fedCI Python package, a privacy-preserving R implementation of IOD, and a web application for the fedCI-IOD pipeline, providing versatile, user-friendly solutions for federated conditional independence testing and causal discovery.

---

### Act, Think or Abstain: Complexity-Aware Adaptive Inference for Vision-Language-Action Models

Authors: Riccardo Andrea Izzo, Gianluca Bardaro, Matteo Matteucci
Published: 2026-03-05
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2603.05147v2

Current research on Vision-Language-Action (VLA) models predominantly focuses on enhancing generalization through reasoning techniques. While effective, these improvements increase computational complexity and inference latency. Furthermore, these mechanisms are typically applied indiscriminately, wasting resources on trivial tasks while failing to provide the uncertainty estimation necessary to prevent catastrophic failure on out-of-distribution scenarios. Inspired by human cognition, we propose an adaptive framework that dynamically routes VLA execution based on the complexity of the perceived state. Our approach transforms the VLA's vision-language backbone into an active detection tool by projecting latent embeddings into a set of parametric and non-parametric estimators. This allows the system to execute known tasks immediately (Act), reason about ambiguous scenarios (Think), and preemptively halt execution when encountering physical or semantic anomalies (Abstain). We find that a Gaussian Mixture Model fitted to fused vision-language embeddings provides the most reliable task-complexity signal, combining visual novelty with instruction context and cross-modal compatibility. Evaluated on the LIBERO and LIBERO-PRO benchmarks as well as on a real robot, our fused configuration achieves up to 87.5% F1-score across two VLA backbones (SmolVLA and $π_0$), retains 83% with as little as 5% of training data, and surpasses state-of-the-art failure detectors.

---

### Feature Resemblance: Towards a Theoretical Understanding of Analogical Reasoning in Transformers

Authors: Ruichen Xu, Wenjing Yan, Ying-Jun Angela Zhang
Published: 2026-03-05
Categories: cs.CL, cs.LG
arXiv: https://arxiv.org/abs/2603.05143v3

Understanding reasoning in large language models is complicated by evaluations that conflate multiple reasoning types. We isolate analogical reasoning, where a model transfers an attribute between entities that share known properties, and study when such transfer can emerge from training. To make the problem analytically tractable, we study a minimal transformer-style abstraction that isolates how learned representations support analogical reasoning. Within this setting, we prove three key results. First, joint training on similarity and attribution premises enables analogical reasoning through aligned representations. Second, sequential training succeeds only when similarity structure is learned before specific attributes, revealing a curriculum asymmetry. Third, in our stylized setting, two-hop reasoning $(a \to b, b \to c \Rightarrow a \to c)$ can be viewed as analogical reasoning with identity bridges $(b=b)$, which appear explicitly in training data. Together, these results reveal a unified mechanism: entities with shared properties become aligned in representation space, enabling property transfer through feature resemblance. Experiments with architectures up to 8B parameters show qualitative agreement with the theory and suggest that representational geometry plays an important role in analogical reasoning beyond the stylized model.

---

### Recurrent Graph Neural Networks and Arithmetic Circuits

Authors: Timon Barlag, Vivian Holzapfel, Laura Strieker, Jonni Virtema, Heribert Vollmer
Published: 2026-03-05
Categories: cs.CC, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2603.05140v2

We characterise the computational power of recurrent graph neural networks (GNNs) in terms of arithmetic circuits over the real numbers. Our networks are not restricted to aggregate-combine GNNs or other particular types. Generalising similar notions from the literature, we introduce the model of recurrent arithmetic circuits, which can be seen as arithmetic analogues of sequential or logical circuits. These circuits utilise so-called memory gates which are used to store data between iterations of the recurrent circuit. While (recurrent) GNNs work on labelled graphs, we construct arithmetic circuits that obtain encoded labelled graphs as real valued tuples and then compute the same function. For the other direction we construct recurrent GNNs which are able to simulate the computations of recurrent circuits. These GNNs are given the circuit-input as initial feature vectors and then, after the GNN-computation, have the circuit-output among the feature vectors of its nodes. In this way we establish an exact correspondence between the expressivity of recurrent GNNs and recurrent arithmetic circuits operating over real numbers. Our results both deepen our understanding of the capabilities of trained neural networks and open new approaches to study recurrent neural networks using the lens of circuit complexity theory.

---

### Particle-Guided Diffusion for Gas-Phase Reaction Kinetics

Authors: Andrew Millard, Henrik Pedersen
Published: 2026-03-05
Categories: physics.chem-ph, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2603.05139v1

Physics-guided sampling with diffusion model priors has shown promise for solving partial differential equation (PDE) governed problems, but applications to chemically meaningful reaction-transport systems remain limited. We apply diffusion-based guided sampling to gas-phase chemical reactions by training on solutions of the advection-reaction-diffusion (ARD) equation across varying parameters. The method generates physically consistent concentration fields and accurately predicts outlet concentrations, including at unseen parameter values, demonstrating the potential of diffusion models for inference in reactive transport.

---

### Representation Fidelity:Auditing Algorithmic Decisions About Humans Using Self-Descriptions

Authors: Theresa Elstner, Martin Potthast
Published: 2026-03-05
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.05136v1

This paper introduces a new dimension for validating algorithmic decisions about humans by measuring the fidelity of their representations. Representation Fidelity measures if decisions about a person rest on reasonable grounds. We propose to operationalize this notion by measuring the distance between two representations of the same person: (1) an externally prescribed input representation on which the decision is based, and (2) a self-description provided by the human subject of the decision, used solely to validate the input representation. We examine the nature of discrepancies between these representations, how such discrepancies can be quantified, and derive a generic typology of representation mismatches that determine the degree of representation fidelity. We further present the first benchmark for evaluating representation fidelity based on a dataset of loan-granting decisions. Our Loan-Granting Self-Representations Corpus 2025 consists of a large corpus of 30 000 synthetic natural language self-descriptions derived from corresponding representations of applicants in the German Credit Dataset, along with expert annotations of representation mismatches between each pair of representations.

---

### SRasP: Self-Reorientation Adversarial Style Perturbation for Cross-Domain Few-Shot Learning

Authors: Wenqian Li, Pengfei Fang, Hui Xue
Published: 2026-03-05
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.05135v1

Cross-Domain Few-Shot Learning (CD-FSL) aims to transfer knowledge from a seen source domain to unseen target domains, serving as a key benchmark for evaluating the robustness and transferability of models. Existing style-based perturbation methods mitigate domain shift but often suffer from gradient instability and convergence to sharp minima.To address these limitations, we propose a novel crop-global style perturbation network, termed Self-Reorientation Adversarial \underline{S}tyle \underline{P}erturbation (SRasP). Specifically, SRasP leverages global semantic guidance to identify incoherent crops, followed by reorienting and aggregating the style gradients of these crops with the global style gradients within one image. Furthermore, we propose a novel multi-objective optimization function to maximize visual discrepancy while enforcing semantic consistency among global, crop, and adversarial features. Applying the stabilized perturbations during training encourages convergence toward flatter and more transferable solutions, improving generalization to unseen domains. Extensive experiments are conducted on multiple CD-FSL benchmarks, demonstrating consistent improvements over state-of-the-art methods.

---

### LBM: Hierarchical Large Auto-Bidding Model via Reasoning and Acting

Authors: Yewen Li, Zhiyi Lyu, Peng Jiang, Qingpeng Cai, Fei Pan, Bo An, Peng Jiang
Published: 2026-03-05
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2603.05134v1

The growing scale of ad auctions on online advertising platforms has intensified competition, making manual bidding impractical and necessitating auto-bidding to help advertisers achieve their economic goals. Current auto-bidding methods have evolved to use offline reinforcement learning or generative methods to optimize bidding strategies, but they can sometimes behave counterintuitively due to the black-box training manner and limited mode coverage of datasets, leading to challenges in understanding task status and generalization in dynamic ad environments. Large language models (LLMs) offer a promising solution by leveraging prior human knowledge and reasoning abilities to improve auto-bidding performance. However, directly applying LLMs to auto-bidding faces difficulties due to the need for precise actions in competitive auctions and the lack of specialized auto-bidding knowledge, which can lead to hallucinations and suboptimal decisions. To address these challenges, we propose a hierarchical Large autoBidding Model (LBM) to leverage the reasoning capabilities of LLMs for developing a superior auto-bidding strategy. This includes a high-level LBM-Think model for reasoning and a low-level LBM-Act model for action generation. Specifically, we propose a dual embedding mechanism to efficiently fuse two modalities, including language and numerical inputs, for language-guided training of the LBM-Act; then, we propose an offline reinforcement fine-tuning technique termed GQPO for mitigating the LLM-Think's hallucinations and enhancing decision-making performance without simulation or real-world rollout like previous multi-turn LLM-based methods. Experiments demonstrate the superiority of a generative backbone based on our LBM, especially in an efficient training manner and generalization ability.

---

### MedCoRAG: Interpretable Hepatology Diagnosis via Hybrid Evidence Retrieval and Multispecialty Consensus

Authors: Zheng Li, Jiayi Xu, Zhikai Hu, Hechang Chen, Lele Cong, Yunyun Wang, Shuchao Pang
Published: 2026-03-05
Categories: cs.AI, cs.MA
arXiv: https://arxiv.org/abs/2603.05129v1

Diagnosing hepatic diseases accurately and interpretably is critical, yet it remains challenging in real-world clinical settings. Existing AI approaches for clinical diagnosis often lack transparency, structured reasoning, and deployability. Recent efforts have leveraged large language models (LLMs), retrieval-augmented generation (RAG), and multi-agent collaboration. However, these approaches typically retrieve evidence from a single source and fail to support iterative, role-specialized deliberation grounded in structured clinical data. To address this, we propose MedCoRAG (i.e., Medical Collaborative RAG), an end-to-end framework that generates diagnostic hypotheses from standardized abnormal findings and constructs a patient-specific evidence package by jointly retrieving and pruning UMLS knowledge graph paths and clinical guidelines. It then performs Multi-Agent Collaborative Reasoning: a Router Agent dynamically dispatches Specialist Agents based on case complexity; these agents iteratively reason over the evidence and trigger targeted re-retrievals when needed, while a Generalist Agent synthesizes all deliberations into a traceable consensus diagnosis that emulates multidisciplinary consultation. Experimental results on hepatic disease cases from MIMIC-IV show that MedCoRAG outperforms existing methods and closed-source models in both diagnostic performance and reasoning interpretability.

---

### Measuring the Redundancy of Decoder Layers in SpeechLLMs

Authors: Adel Moumen, Guangzhi Sun, Philip C Woodland
Published: 2026-03-05
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2603.05121v2

Speech Large Language Models route speech encoder representations into an LLM decoder that typically accounts for over 90% of total parameters. We study how much of this decoder capacity is actually needed for speech tasks. Across two LLM families and three scales (1-8B), we show that decoder redundancy is largely inherited from the pretrained LLM: text and speech inputs yield similar redundant blocks. We then measure excess capacity by pruning decoder layers and analysing post-pruning healing to increase robustness. Our findings show that 7-8B models retain good ASR performance with only 60% of decoder layers, and the same trend extends to smaller scales with reduced pruning tolerance. We then generalise to speech translation, and show that the same blocks of layers are redundant across speech encoders, tasks and languages, indicating that a more global redundancy structure exists, enabling a single pruned and multi-tasks SpeechLLM backbone to be deployed.

---

### Bidirectional Curriculum Generation: A Multi-Agent Framework for Data-Efficient Mathematical Reasoning

Authors: Boren Hu, Xiao Liu, Boci Peng, Xinping Zhao, Xiaoran Shang, Yun Zhu, Lijun Wu
Published: 2026-03-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.05120v1

Enhancing mathematical reasoning in Large Language Models typically demands massive datasets, yet data efficiency remains a critical bottleneck. While Curriculum Learning attempts to structure this process, standard unidirectional approaches (simple-to-complex) suffer from inefficient sample utilization: they blindly escalate complexity even when foundational gaps persist, leading to wasted computation on unsolvable problems. To maximize the instructional value of every training sample, we introduce a novel Bidirectional Curriculum Generation framework. Unlike rigid trajectories, our multi-agent ecosystem mimics adaptive pedagogy to establish a closed feedback loop. It dynamically generates data by either complicating problems to challenge the model or, crucially, simplying them to repair specific reasoning failures. This mechanism ensures that the model consumes only the most effective data at any given stage. Grounded in the Optimal Pacing Theorem, our approach optimizes the learning trajectory, significantly outperforming baselines while achieving superior reasoning performance with substantially fewer instruction samples.

---

### FedBCD:Communication-Efficient Accelerated Block Coordinate Gradient Descent for Federated Learning

Authors: Junkang Liu, Fanhua Shang, Yuanyuan Liu, Hongying Liu, Yuangang Li, YunXiang Gong
Published: 2026-03-05
Categories: cs.LG, cs.AI
arXiv: https://arxiv.org/abs/2603.05116v1

Although Federated Learning has been widely studied in recent years, there are still high overhead expenses in each communication round for large-scale models such as Vision Transformer. To lower the communication complexity, we propose a novel Federated Block Coordinate Gradient Descent (FedBCGD) method for communication efficiency. The proposed method splits model parameters into several blocks, including a shared block and enables uploading a specific parameter block by each client, which can significantly reduce communication overhead. Moreover, we also develop an accelerated FedBCGD algorithm (called FedBCGD+) with client drift control and stochastic variance reduction. To the best of our knowledge, this paper is the first work on parameter block communication for training large-scale deep models. We also provide the convergence analysis for the proposed algorithms. Our theoretical results show that the communication complexities of our algorithms are a factor $1/N$ lower than those of existing methods, where $N$ is the number of parameter blocks, and they enjoy much faster convergence than their counterparts. Empirical results indicate the superiority of the proposed algorithms compared to state-of-the-art algorithms. The code is available at https://github.com/junkangLiu0/FedBCGD.

---

### UniPAR: A Unified Framework for Pedestrian Attribute Recognition

Authors: Minghe Xu, Rouying Wu, Jiarui Xu, Minhao Sun, Zikang Yan, Xiao Wang, ChiaWei Chu, Yu Li
Published: 2026-03-05
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.05114v1

Pedestrian Attribute Recognition is a foundational computer vision task that provides essential support for downstream applications, including person retrieval in video surveillance and intelligent retail analytics. However, existing research is frequently constrained by the ``one-model-per-dataset" paradigm and struggles to handle significant discrepancies across domains in terms of modalities, attribute definitions, and environmental scenarios. To address these challenges, we propose UniPAR, a unified Transformer-based framework for PAR. By incorporating a unified data scheduling strategy and a dynamic classification head, UniPAR enables a single model to simultaneously process diverse datasets from heterogeneous modalities, including RGB images, video sequences, and event streams. We also introduce an innovative phased fusion encoder that explicitly aligns visual features with textual attribute queries through a late deep fusion strategy. Experimental results on the widely used benchmark datasets, including MSP60K, DukeMTMC, and EventPAR, demonstrate that UniPAR achieves performance comparable to specialized SOTA methods. Furthermore, multi-dataset joint training significantly enhances the model's cross-domain generalization and recognition robustness in extreme environments characterized by low light and motion blur. The source code of this paper will be released on https://github.com/Event-AHU/OpenPAR

---

### Decoupling Task and Behavior: A Two-Stage Reward Curriculum in Reinforcement Learning for Robotics

Authors: Kilian Freitag, Knut Åkesson, Morteza Haghir Chehreghani
Published: 2026-03-05
Categories: cs.LG, cs.RO
arXiv: https://arxiv.org/abs/2603.05113v1

Deep Reinforcement Learning is a promising tool for robotic control, yet practical application is often hindered by the difficulty of designing effective reward functions. Real-world tasks typically require optimizing multiple objectives simultaneously, necessitating precise tuning of their weights to learn a policy with the desired characteristics. To address this, we propose a two-stage reward curriculum where we decouple task-specific objectives from behavioral terms. In our method, we first train the agent on a simplified task-only reward function to ensure effective exploration before introducing the full reward that includes auxiliary behavior-related terms such as energy efficiency. Further, we analyze various transition strategies and demonstrate that reusing samples between phases is critical for training stability. We validate our approach on the DeepMind Control Suite, ManiSkill3, and a mobile robot environment, modified to include auxiliary behavioral objectives. Our method proves to be simple yet effective, substantially outperforming baselines trained directly on the full reward while exhibiting higher robustness to specific reward weightings.

---

### SPIRIT: Perceptive Shared Autonomy for Robust Robotic Manipulation under Deep Learning Uncertainty

Authors: Jongseok Lee, Ribin Balachandran, Harsimran Singh, Jianxiang Feng, Hrishik Mishra, Marco De Stefano, Rudolph Triebel, Alin Albu-Schaeffer, Konstantin Kondak
Published: 2026-03-05
Categories: cs.RO, cs.AI
arXiv: https://arxiv.org/abs/2603.05111v1

Deep learning (DL) has enabled impressive advances in robotic perception, yet its limited robustness and lack of interpretability hinder reliable deployment in safety critical applications. We propose a concept termed perceptive shared autonomy, in which uncertainty estimates from DL based perception are used to regulate the level of autonomy. Specifically, when the robot's perception is confident, semi-autonomous manipulation is enabled to improve performance; when uncertainty increases, control transitions to haptic teleoperation for maintaining robustness. In this way, high-performing but uninterpretable DL methods can be integrated safely into robotic systems. A key technical enabler is an uncertainty aware DL based point cloud registration approach based on the so called Neural Tangent Kernels (NTK). We evaluate perceptive shared autonomy on challenging aerial manipulation tasks through a user study of 15 participants and realization of mock-up industrial scenarios, demonstrating reliable robotic manipulation despite failures in DL based perception. The resulting system, named SPIRIT, improves both manipulation performance and system reliability. SPIRIT was selected as a finalist of a major industrial innovation award.

---

### BLINK: Behavioral Latent Modeling of NK Cell Cytotoxicity

Authors: Iman Nematollahi, Jose Francisco Villena-Ossa, Alina Moter, Kiana Farhadyar, Gabriel Kalweit, Abhinav Valada, Toni Cathomen, Evelyn Ullrich, Maria Kalweit
Published: 2026-03-05
Categories: cs.CV, cs.LG
arXiv: https://arxiv.org/abs/2603.05110v2

Machine learning models of cellular interaction dynamics hold promise for understanding cell behavior. Natural killer (NK) cell cytotoxicity is a prominent example of such interaction dynamics and is commonly studied using time-resolved multi-channel fluorescence microscopy. Although tumor cell death events can be annotated at single frames, NK cytotoxic outcome emerges over time from cellular interactions and cannot be reliably inferred from frame-wise classification alone. We introduce BLINK, a trajectory-based recurrent state-space model that serves as a cell world model for NK-tumor interactions. BLINK learns latent interaction dynamics from partially observed NK-tumor interaction sequences and predicts apoptosis increments that accumulate into cytotoxic outcomes. Experiments on long-term time-lapse NK-tumor recordings show improved cytotoxic outcome detection and enable forecasting of future outcomes, together with an interpretable latent representation that organizes NK trajectories into coherent behavioral modes and temporally structured interaction phases. BLINK provides a unified framework for quantitative evaluation and structured modeling of NK cytotoxic behavior at the single-cell level.

---

### Diff-ES: Stage-wise Structural Diffusion Pruning via Evolutionary Search

Authors: Zongfang Liu, Shengkun Tang, Zongliang Wu, Xin Yuan, Zhiqiang Shen
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05105v1

Diffusion models have achieved remarkable success in high-fidelity image generation but remain computationally demanding due to their multi-step denoising process and large model sizes. Although prior work improves efficiency either by reducing sampling steps or by compressing model parameters, existing structured pruning approaches still struggle to balance real acceleration and image quality preservation. In particular, prior methods such as MosaicDiff rely on heuristic, manually tuned stage-wise sparsity schedules and stitch multiple independently pruned models during inference, which increases memory overhead. However, the importance of diffusion steps is highly non-uniform and model-dependent. As a result, schedules derived from simple heuristics or empirical observations often fail to generalize and may lead to suboptimal performance. To this end, we introduce \textbf{Diff-ES}, a stage-wise structural \textbf{Diff}usion pruning framework via \textbf{E}volutionary \textbf{S}earch, which optimizes the stage-wise sparsity schedule and executes it through memory-efficient weight routing without model duplication. Diff-ES divides the diffusion trajectory into multiple stages, automatically discovers an optimal stage-wise sparsity schedule via evolutionary search, and activates stage-conditioned weights dynamically without duplicating model parameters. Our framework naturally integrates with existing structured pruning methods for diffusion models including depth and width pruning. Extensive experiments on DiT and SDXL demonstrate that Diff-ES consistently achieves wall-clock speedups while incurring minimal degradation in generation quality, establishing state-of-the-art performance for structured diffusion model pruning.

---

### ARC-TGI: Human-Validated Task Generators with Reasoning Chain Templates for ARC-AGI

Authors: Jens Lehmann, Syeda Khushbakht, Nikoo Salehfard, Nur A Zarin Nishat, Dhananjay Bhandiwad, Andrei Aioanei, Sahar Vahdati
Published: 2026-03-05
Categories: cs.CL, cs.AI, cs.LG
arXiv: https://arxiv.org/abs/2603.05099v1

The Abstraction and Reasoning Corpus (ARC-AGI) probes few-shot abstraction and rule induction on small visual grids, but progress is difficult to measure on static collections of hand-authored puzzles due to overfitting, dataset leakage, and memorisation. We introduce ARC-TGI (ARC Task Generators Inventory), an open-source framework for task-family generators: compact Python programs that sample diverse ARC-AGI tasks while preserving a latent rule. ARC-TGI is built around a solver-facing representation: each generated task is paired with natural-language input and transformation reasoning chains and partially evaluated Python code implementing sampling, transformation, and episode construction. Crucially, ARC-TGI supports task-level constraints so that training examples collectively expose the variations needed to infer the underlying rule, a requirement for human-solvable ARC tasks that independent per-example sampling often fails to guarantee. All generators undergo human refinement and local verification to keep both grids and reasoning traces natural and consistent under variation. We release 461 generators covering 180 ARC-Mini tasks, 215 ARC-AGI-1 tasks (200 train, 15 test), and 66 ARC-AGI-2 tasks (55 train, 11 test), enabling scalable dataset sampling and controlled benchmarking.

---

### GEM-TFL: Bridging Weak and Full Supervision for Forgery Localization through EM-Guided Decomposition and Temporal Refinement

Authors: Xiaodong Zhu, Yuanming Zheng, Suting Wang, Junqi Yang, Yuhong Yang, Weiping Tu, Zhongyuan Wang
Published: 2026-03-05
Categories: cs.CV, cs.AI
arXiv: https://arxiv.org/abs/2603.05095v1

Temporal Forgery Localization (TFL) aims to precisely identify manipulated segments within videos or audio streams, providing interpretable evidence for multimedia forensics and security. While most existing TFL methods rely on dense frame-level labels in a fully supervised manner, Weakly Supervised TFL (WS-TFL) reduces labeling cost by learning only from binary video-level labels. However, current WS-TFL approaches suffer from mismatched training and inference objectives, limited supervision from binary labels, gradient blockage caused by non-differentiable top-k aggregation, and the absence of explicit modeling of inter-proposal relationships. To address these issues, we propose GEM-TFL (Graph-based EM-powered Temporal Forgery Localization), a two-phase classification-regression framework that effectively bridges the supervision gap between training and inference. Built upon this foundation, (1) we enhance weak supervision by reformulating binary labels into multi-dimensional latent attributes through an EM-based optimization process; (2) we introduce a training-free temporal consistency refinement that realigns frame-level predictions for smoother temporal dynamics; and (3) we design a graph-based proposal refinement module that models temporal-semantic relationships among proposals for globally consistent confidence estimation. Extensive experiments on benchmark datasets demonstrate that GEM-TFL achieves more accurate and robust temporal forgery localization, substantially narrowing the gap with fully supervised methods.

---

### From Baselines to Transport Geodesics: Axiomatic Attribution via Optimal Generative Flows

Authors: Cenwei Zhang, Lin Zhu, Manxi Lin, Lei You
Published: 2026-03-05
Categories: cs.LG, cs.AI, cs.CV
arXiv: https://arxiv.org/abs/2603.05093v2

Feature attributions often hide a critical modeling choice: they explain a prediction along a counterfactual path from a reference state to an input. Different baselines, interpolations, and generative trajectories define different paths and can therefor produce different explanations. We study this path ambiguity as a modeling problem. Our central question is whether the path can be chosen by the data-generating transport process, rather than by a hand-designed interpolation or by the sensitivity geometry of the model being explained. We separate attribution into fixed-path credit allocation and path selection. For a fixed path, we prove that the Aumann-Shapley line integral is the unique attribution rule under standard fixed-path axioms and explicit coordinate-trace regularity. For path selection, we minimize kinetic action over flows that transport a reference distribution to the data distribution, yielding a transport-geodesic attribution principle. We approximate this ideal with Rectified Flow and Reflow and derive stability bounds linking vector-field error to attribution error. Experiments show that lower-action, transport-consistent paths produce more stable and structured explanations, preserving competitive deletion faithfulness, without claiming data-manifold membership. Our code is available at https://github.com/cenweizhang/OTFlowSHAP.

---

### Aura: Universal Multi-dimensional Exogenous Integration for Aviation Time Series

Authors: Jiafeng Lin, Mengren Zheng, Simeng Ye, Yuxuan Wang, Huan Zhang, Yuhui Liu, Zhongyi Pei, Jianmin Wang
Published: 2026-03-05
Categories: cs.LG, cs.AI, cs.CL
arXiv: https://arxiv.org/abs/2603.05092v2

Time series forecasting has witnessed an increasing demand across diverse industrial applications, where accurate predictions are pivotal for informed decision-making. Beyond numerical time series data, reliable forecasting in practical scenarios requires integrating diverse exogenous factors. Such exogenous information is often multi-dimensional or even multimodal, introducing heterogeneous interactions that unimodal time series models struggle to capture. In this paper, we delve into an aviation maintenance scenario and identify three distinct types of exogenous factors that influence temporal dynamics through distinct interaction modes. Based on this empirical insight, we propose Aura, a universal framework that explicitly organizes and encodes heterogeneous external information according to its interaction mode with the target time series. Specifically, Aura utilizes a tailored tripartite encoding mechanism to embed heterogeneous features into well-established time series models, ensuring seamless integration of non-sequential context. Extensive experiments on a large-scale, three-year industrial dataset from China Southern Airlines, covering the Boeing 777 and Airbus A320 fleets, demonstrate that Aura consistently achieves state-of-the-art performance across all baselines and exhibits superior adaptability. Our findings highlight Aura's potential as a general-purpose enhancement for aviation safety and reliability.

---

### Orthogonal Spatial-temporal Distributional Transfer for 4D Generation

Authors: Wei Liu, Shengqiong Wu, Bobo Li, Haoyu Zhao, Hao Fei, Mong-Li Lee, Wynne Hsu
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05081v1

In the AIGC era, generating high-quality 4D content has garnered increasing research attention. Unfortunately, current 4D synthesis research is severely constrained by the lack of large-scale 4D datasets, preventing models from adequately learning the critical spatial-temporal features necessary for high-quality 4D generation, thus hindering progress in this domain. To combat this, we propose a novel framework that transfers rich spatial priors from existing 3D diffusion models and temporal priors from video diffusion models to enhance 4D synthesis. We develop a spatial-temporal-disentangled 4D (STD-4D) Diffusion model, which synthesizes 4D-aware videos through disentangled spatial and temporal latents. To facilitate the best feature transfer, we design a novel Orthogonal Spatial-temporal Distributional Transfer (Orster) mechanism, where the spatiotemporal feature distributions are carefully modeled and injected into the STD-4D Diffusion. Furthermore, during the 4D construction, we devise a spatial-temporal-aware HexPlane (ST-HexPlane) to integrate the transferred spatiotemporal features, thereby improving 4D deformation and 4D Gaussian feature modeling. Experiments demonstrate that our method significantly outperforms existing approaches, achieving superior spatial-temporal consistency and higher-quality 4D synthesis.

---

### MoRe: Motion-aware Feed-forward 4D Reconstruction Transformer

Authors: Juntong Fang, Zequn Chen, Weiqi Zhang, Donglin Di, Xuancheng Zhang, Chengmin Yang, Yu-Shen Liu
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05078v2

Reconstructing dynamic 4D scenes remains challenging due to the presence of moving objects that corrupt camera pose estimation. Existing optimization methods alleviate this issue with additional supervision, but they are mostly computationally expensive and impractical in real-time applications. To address these limitations, we propose MoRe, a feedforward 4D reconstruction network that efficiently recovers dynamic 3D scenes from monocular videos. Built upon a strong static reconstruction backbone, MoRe employs an attention-forcing strategy to disentangle dynamic motion from static structure. To further enhance robustness, we fine-tune the model on large-scale, diverse datasets encompassing both dynamic and static scenes. Moreover, our grouped causal attention captures temporal dependencies and adapts to varying token lengths across frames, ensuring temporally coherent geometry reconstruction. Extensive experiments on multiple benchmarks demonstrate that MoRe achieves high-quality dynamic reconstructions with exceptional efficiency.

---

### UniM: A Unified Any-to-Any Interleaved Multimodal Benchmark

Authors: Yanlin Li, Minghui Guo, Kaiwen Zhang, Shize Zhang, Yiran Zhao, Haodong Li, Congyue Zhou, Weijie Zheng, Yushen Yan, Shengqiong Wu, Wei Ji, Lei Cui et al.
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05075v1

In real-world multimodal applications, systems usually need to comprehend arbitrarily combined and interleaved multimodal inputs from users, while also generating outputs in any interleaved multimedia form. This capability defines the goal of any-to-any interleaved multimodal learning under a unified paradigm of understanding and generation, posing new challenges and opportunities for advancing Multimodal Large Language Models (MLLMs). To foster and benchmark this capability, this paper introduces the UniM benchmark, the first Unified Any-to-Any Interleaved Multimodal dataset. UniM contains 31K high-quality instances across 30 domains and 7 representative modalities: text, image, audio, video, document, code, and 3D, each requiring multiple intertwined reasoning and generation capabilities. We further introduce the UniM Evaluation Suite, which assesses models along three dimensions: Semantic Correctness & Generation Quality, Response Structure Integrity, and Interleaved Coherence. In addition, we propose UniMA, an agentic baseline model equipped with traceable reasoning for structured interleaved generation. Comprehensive experiments demonstrate the difficulty of UniM and highlight key challenges and directions for advancing unified any-to-any multimodal intelligence. The project page is https://any2any-mllm.github.io/unim.

---

### MI-DETR: A Strong Baseline for Moving Infrared Small Target Detection with Motion Integration

Authors: Nian Liu, Jin Gao, Zhen Liang, Shubo Lin, Sikui Zhang, Fudong Ge, Liang Li, Weiming Hu
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05071v2

Detecting moving infrared small targets is challenging because tiny, low-contrast targets occupy few pixels and are easily obscured by dynamic backgrounds. Existing multi-frame methods aggregate temporal information across frames to capture motion. However, dynamic background changes can generate similar motion cues, making it difficult to distinguish between target motion and background interference. Furthermore, even when motion cues are extracted, combining them with current-frame appearance features remains difficult. To address these issues, we propose Motion Integration DETR (MI-DETR), a three-stage framework that explicitly models motion and fuses it with appearance features. First, to suppress background clutter while preserving target-related motion cues, Recurrent Interpretable Motion Cue Aggregation (RIMCA) maintains a recurrent temporal state that accumulates motion across consecutive frames, producing a causal and spatially aligned motion representation. Second, to integrate spatial and temporal information, Pathway Mutual Interaction (PMI) preserves separate appearance and motion pathways while enabling bidirectional feature exchange between them. Finally, an RT-DETR-based detector uses these refined features for end-to-end target localization. Experiments on DAUB-R, ITSDT-15K, and IRDST-H show that explicit motion modeling and pathway interaction effectively improve moving infrared small target detection.

---

### MI-DETR: A Strong Baseline for Moving Infrared Small Target Detection with Bio-Inspired Motion Integration

Authors: Nian Liu, Jin Gao, Shubo Lin, Yutong Kou, Sikui Zhang, Fudong Ge, Zhiqiang Pu, Liang Li, Gang Wang, Yizheng Wang, Weiming Hu
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05071v1

Infrared small target detection (ISTD) is challenging because tiny, low-contrast targets are easily obscured by complex and dynamic backgrounds. Conventional multi-frame approaches typically learn motion implicitly through deep neural networks, often requiring additional motion supervision or explicit alignment modules. We propose Motion Integration DETR (MI-DETR), a bio-inspired dual-pathway detector that processes one infrared frame per time step while explicitly modeling motion. First, a retina-inspired cellular automaton (RCA) converts raw frame sequences into a motion map defined on the same pixel grid as the appearance image, enabling parvocellular-like appearance and magnocellular-like motion pathways to be supervised by a single set of bounding boxes without extra motion labels or alignment operations. Second, a Parvocellular-Magnocellular Interconnection (PMI) Block facilitates bidirectional feature interaction between the two pathways, providing a biologically motivated intermediate interconnection mechanism. Finally, a RT-DETR decoder operates on features from the two pathways to produce detection results. Surprisingly, our proposed simple yet effective approach yields strong performance on three commonly used ISTD benchmarks. MI-DETR achieves 70.3% mAP@50 and 72.7% F1 on IRDST-H (+26.35 mAP@50 over the best multi-frame baseline), 98.0% mAP@50 on DAUB-R, and 88.3% mAP@50 on ITSDT-15K, demonstrating the effectiveness of biologically inspired motion-appearance integration. Code is available at https://github.com/nliu-25/MI-DETR.

---

### Jagarin: A Three-Layer Architecture for Hibernating Personal Duty Agents on Mobile

Authors: Ravi Kiran Kadaboina
Published: 2026-03-05
Categories: cs.AI, cs.HC, cs.MA
arXiv: https://arxiv.org/abs/2603.05069v3

Personal AI agents face a deployment paradox on mobile: persistent background execution drains the battery and conflicts with platform background limits, yet purely reactive agents miss time-sensitive obligations until the user remembers to ask. We present Jagarin, a three-layer architecture that resolves this through structured hibernation and demand-driven wake. DAWN (Duty-Aware Wake Network) is an on-device scoring engine that runs on the platform's periodic wake and combines four signals (duty-typed optimal action windows, predicted user engagement, the cost of delay, and cross-duty batching) with per-duty adaptive thresholds to decide whether a sleeping agent should stay silent, nudge the user, or offer escalation. ARIA (Agent Relay Identity Architecture) is a commercial email identity proxy that turns institutional email into structured duty records and routes messages by category, removing manual data entry. ACE (Agent-Centric Exchange) is a protocol for machine-readable communication from institutions to personal agents, intended to make email parsing unnecessary in the long run. DAWN and ACE are specified and evaluated in companion papers; this paper describes how the three layers fit together and a working Flutter prototype on Android that combines them with an ephemeral cloud agent invoked only when the user asks. Behavioural signals, thresholds and scoring never leave the device, and every record ARIA extracts is sealed to the device's public key before it is stored, so the relay holds only ciphertext it cannot read. Cloud model exposure is limited to parsing commercial email and to user-initiated escalation, which receives only the structured duty record.

---

### Jagarin: A Three-Layer Architecture for Hibernating Personal Duty Agents on Mobile

Authors: Ravi Kiran Kadaboina
Published: 2026-03-05
Categories: cs.AI, cs.HC, cs.MA
arXiv: https://arxiv.org/abs/2603.05069v2

Personal AI agents face a fundamental deployment paradox on mobile: persistent background execution drains battery and violates platform sandboxing policies, yet purely reactive agents miss time-sensitive obligations until the user remembers to ask. We present Jagarin, a three-layer architecture that resolves this paradox through structured hibernation and demand-driven wake. The first layer, DAWN (Duty-Aware Wake Network), is an on-device heuristic engine that computes a composite urgency score from four signals: duty-typed optimal action windows, user behavioral engagement prediction, opportunity cost of inaction, and cross-duty batch resonance. It uses adaptive per-user thresholds to decide when a sleeping agent should nudge or escalate. The second layer, ARIA (Agent Relay Identity Architecture), is a commercial email identity proxy that routes the full commercial inbox -- obligations, promotional offers, loyalty rewards, and platform updates -- to appropriate DAWN handlers by message category, eliminating cold-start and removing manual data entry. The third layer, ACE (Agent-Centric Exchange), is a protocol framework for direct machine-readable communication from institutions to personal agents, replacing human-targeted email as the canonical channel. Together, these three layers form a complete stack from institutional signal to on-device action, without persistent cloud state, continuous background execution, or privacy compromise. A working Flutter prototype is demonstrated on Android, combining all three layers with an ephemeral cloud agent invoked only on user-initiated escalation.

---

### Cyber Threat Intelligence for Artificial Intelligence Systems

Authors: Natalia Krawczyk, Mateusz Szczepkowski, Adrian Brodzik, Krzysztof Bocianiak
Published: 2026-03-05
Categories: cs.CR, cs.AI
arXiv: https://arxiv.org/abs/2603.05068v1

As artificial intelligence (AI) becomes deeply embedded in critical services and everyday products, it is increasingly exposed to security threats which traditional cyber defenses were not designed to handle. In this paper, we investigate how cyber threat intelligence (CTI) may evolve to address attacks that target AI systems. We first analyze the assumptions and workflows of conventional threat intelligence with the needs of AI-focused defense, highlighting AI-specific assets and vulnerabilities. We then review and organize the current landscape of AI security knowledge. Based on this, we outline what an AI-oriented threat intelligence knowledge base should contain, describing concrete indicators of compromise (IoC) for different AI supply-chain phases and artifacts, and showing how such a knowledge base could support security tools. Finally, we discuss techniques for measuring similarity between collected indicators and newly observed AI artifacts. The review reveals gaps and quality issues in existing resources and identifies potential future research directions toward a practical threat intelligence framework tailored to AI.

---

### Synchronization-based clustering on the unit hypersphere

Authors: Zinaid Kapić, Aladin Crnkić, Goran Mauša
Published: 2026-03-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.05067v1

Clustering on the unit hypersphere is a fundamental problem in various fields, with applications ranging from gene expression analysis to text and image classification. Traditional clustering methods are not always suitable for unit sphere data, as they do not account for the geometric structure of the sphere. We introduce a novel algorithm for clustering data represented as points on the unit sphere $\mathbf{S}^{d-1}$. Our method is based on the $d$-dimensional generalized Kuramoto model. The effectiveness of the introduced method is demonstrated on synthetic and real-world datasets. Results are compared with some of the traditional clustering methods, showing that our method achieves similar or better results in terms of clustering accuracy.

---

### Reward-Conditioned Reinforcement Learning

Authors: Michal Nauman, Marek Cygan, Pieter Abbeel
Published: 2026-03-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.05066v3

Single-task RL agents are typically trained under a fixed reward function, which limits their robustness to reward misspecification and their ability to adapt to changing preferences. We introduce Reward-Conditioned Reinforcement Learning (RCRL), an off-policy method that conditions agents on reward parameterizations while collecting experience under a single nominal objective. By recomputing counterfactual rewards from shared replay data, RCRL exposes the agent to multiple reward objectives without additional environment interaction, connecting single-task RL with ideas from multi-objective and multi-task learning. Across single-task, multi-task, and vision-based benchmarks, RCRL improves sample efficiency under the nominal reward parameterization, enables efficient adaptation to new parameterizations, and supports zero-shot behavioral adjustment at deployment. Our results show that RCRL provides a scalable mechanism for learning robust, steerable policies without sacrificing the simplicity of single-task training.

---

### Deep Learning-Driven Friendly Jamming for Secure Multicarrier ISAC Under Channel Uncertainty

Authors: Bui Minh Tuan, Van-Dinh Nguyen, Diep N. Nguyen, Nguyen Linh Trung, Nguyen Van Huynh, Dinh Thai Hoang, Marwan Krunz, Eryk Dutkiewicz
Published: 2026-03-05
Categories: cs.LG
arXiv: https://arxiv.org/abs/2603.05062v1

Integrated sensing and communication (ISAC) systems promise efficient spectrum utilization by jointly supporting radar sensing and wireless communication. This paper presents a deep learning-driven framework for enhancing physical-layer security in multicarrier ISAC systems under imperfect channel state information (CSI) and in the presence of unknown eavesdropper (Eve) locations. Unlike conventional ISAC-based friendly jamming (FJ) approaches that require Eve's CSI or precise angle-of-arrival (AoA) estimates, our method exploits radar echo feedback to guide directional jamming without explicit Eve's information. To enhance robustness to radar sensing uncertainty, we propose a radar-aware neural network that jointly optimizes beamforming and jamming by integrating a novel nonparametric Fisher Information Matrix (FIM) estimator based on f-divergence. The jamming design satisfies the Cramer-Rao lower bound (CRLB) constraints even in the presence of noisy AoA. For efficient implementation, we introduce a quantized tensor train-based encoder that reduces the model size by more than 100 times with negligible performance loss. We also integrate a non-overlapping secure scheme into the proposed framework, in which specific sub-bands can be dedicated solely to communication. Extensive simulations demonstrate that the proposed solution achieves significant improvements in secrecy rate, reduced block error rate (BLER), and strong robustness against CSI uncertainty and angular estimation errors, underscoring the effectiveness of the proposed deep learning-driven friendly jamming framework under practical ISAC impairments.

---

### Asymptotic Behavior of Multi--Task Learning: Implicit Regularization and Double Descent Effects

Authors: Ayed M. Alrashdi, Oussama Dhifallah, Houssem Sifaou
Published: 2026-03-05
Categories: cs.LG, cs.IT
arXiv: https://arxiv.org/abs/2603.05060v2

Multi--task learning seeks to improve the generalization error by leveraging the common information shared by multiple related tasks. One challenge in multi--task learning is identifying formulations capable of uncovering the common information shared between different but related tasks. This paper provides a precise asymptotic analysis of a popular multi--task formulation associated with misspecified perceptron learning models. The main contribution of this paper is to precisely determine the reasons behind the benefits gained from combining multiple related tasks. Specifically, we show that combining multiple tasks is asymptotically equivalent to a traditional formulation with additional regularization terms that help improve the generalization performance. Another contribution is to empirically study the impact of combining tasks on the generalization error. In particular, we empirically show that the combination of multiple tasks postpones the double descent phenomenon and can mitigate it asymptotically.

---

### A 360-degree Multi-camera System for Blue Emergency Light Detection Using Color Attention RT-DETR and the ABLDataset

Authors: Francisco Vacalebri-Lloret, Lucas Banchero, Jose J. Lopez, Jose M. Mossi
Published: 2026-03-05
Categories: cs.CV, cs.AI, eess.IV
arXiv: https://arxiv.org/abs/2603.05058v1

This study presents an advanced system for detecting blue lights on emergency vehicles, developed using ABLDataset, a curated dataset that includes images of European emergency vehicles under various climatic and geographic conditions. The system employs a configuration of four fisheye cameras, each with a 180-degree horizontal field of view, mounted on the sides of the vehicle. A calibration process enables the azimuthal localization of the detections. Additionally, a comparative analysis of major deep neural network algorithms was conducted, including YOLO (v5, v8, and v10), RetinaNet, Faster R-CNN, and RT-DETR. RT-DETR was selected as the base model and enhanced through the incorporation of a color attention block, achieving an accuracy of 94.7 percent and a recall of 94.1 percent on the test set, with field test detections reaching up to 70 meters. Furthermore, the system estimates the approach angle of the emergency vehicle relative to the center of the car using geometric transformations. Designed for integration into a multimodal system that combines visual and acoustic data, this system has demonstrated high efficiency, offering a promising approach to enhancing Advanced Driver Assistance Systems (ADAS) and road safety.

---

### MUTEX: Leveraging Multilingual Transformers and Conditional Random Fields for Enhanced Urdu Toxic Span Detection

Authors: Inayat Arshad, Fajar Saleem, Ijaz Hussain
Published: 2026-03-05
Categories: cs.CL, cs.AI
arXiv: https://arxiv.org/abs/2603.05057v1

Urdu toxic span detection remains limited because most existing systems rely on sentence-level classification and fail to identify the specific toxic spans within those text. It is further exacerbated by the multiple factors i.e. lack of token-level annotated resources, linguistic complexity of Urdu, frequent code-switching, informal expressions, and rich morphological variations. In this research, we propose MUTEX: a multilingual transformer combined with conditional random fields (CRF) for Urdu toxic span detection framework that uses manually annotated token-level toxic span dataset to improve performance and interpretability. MUTEX uses XLM RoBERTa with CRF layer to perform sequence labeling and is tested on multi-domain data extracted from social media, online news, and YouTube reviews using token-level F1 to evaluate fine-grained span detection. The results indicate that MUTEX achieves 60% token-level F1 score that is the first supervised baseline for Urdu toxic span detection. Further examination reveals that transformer-based models are more effective at implicitly capturing the contextual toxicity and are able to address the issues of code-switching and morphological variation than other models.

---

### CLIP-driven Zero-shot Learning with Ambiguous Labels

Authors: Jinfu Fan, Jiangnan Li, Xiaowen Yan, Xiaohui Zhong, Wenpeng Lu, Linqing Huang
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05053v1

Zero-shot learning (ZSL) aims to recognize unseen classes by leveraging semantic information from seen classes, but most existing methods assume accurate class labels for training instances. However, in real-world scenarios, noise and ambiguous labels can significantly reduce the performance of ZSL. To address this, we propose a new CLIP-driven partial label zero-shot learning (CLIP-PZSL) framework to handle label ambiguity. First, we use CLIP to extract instance and label features. Then, a semantic mining block fuses these features to extract discriminative label embeddings. We also introduce a partial zero-shot loss, which assigns weights to candidate labels based on their relevance to the instance and aligns instance and label embeddings to minimize semantic mismatch. As the training goes on, the ground-truth labels are progressively identified, and the refined labels and label embeddings in turn help improve the semantic alignment of instance and label features. Comprehensive experiments on several datasets demonstrate the advantage of CLIP-PZSL.

---

### MCEL: Margin-Based Cross-Entropy Loss for Error-Tolerant Quantized Neural Networks

Authors: Mikail Yayla, Akash Kumar
Published: 2026-03-05
Categories: cs.LG, cs.AR
arXiv: https://arxiv.org/abs/2603.05048v1

Robustness to bit errors is a key requirement for the reliable use of neural networks (NNs) on emerging approximate computing platforms and error-prone memory technologies. A common approach to achieve bit error tolerance in NNs is injecting bit flips during training according to a predefined error model. While effective in certain scenarios, training-time bit flip injection introduces substantial computational overhead, often degrades inference accuracy at high error rates, and scales poorly for larger NN architectures. These limitations make error injection an increasingly impractical solution for ensuring robustness on future approximate computing platforms and error-prone memory technologies. In this work, we investigate the mechanisms that enable NNs to tolerate bit errors without relying on error-aware training. We establish a direct connection between bit error tolerance and classification margins at the output layer. Building on this insight, we propose a novel loss function, the Margin Cross-Entropy Loss (MCEL), which explicitly promotes logit-level margin separation while preserving the favorable optimization properties of the standard cross-entropy loss. Furthermore, MCEL introduces an interpretable margin parameter that allows robustness to be tuned in a principled manner. Extensive experimental evaluations across multiple datasets of varying complexity, diverse NN architectures, and a range of quantization schemes demonstrate that MCEL substantially improves bit error tolerance, up to 15 % in accuracy for an error rate of 1 %. Our proposed MCEL method is simple to implement, efficient, and can be integrated as a drop-in replacement for standard CEL. It provides a scalable and principled alternative to training-time bit flip injection, offering new insights into the origins of NN robustness and enabling more efficient deployment on approximate computing and memory systems.

---

### NeuronMoE: Neuron-Guided Mixture-of-Experts for Efficient Multilingual LLM Extension

Authors: Rongzhi Li, Hitomi Yanaka
Published: 2026-03-05
Categories: cs.CL
arXiv: https://arxiv.org/abs/2603.05046v1

Extending large language models to low-resource languages is essential for global accessibility, but training separate models per language is prohibitively expensive. Mixture-of-Experts (MoE) architectures address this by adding sparse language-specific parameters, but determining how many experts each layer needs remains an open question. Current approaches allocate experts based on layer-level similarity, yet language processing exhibits fine-grained specialization at individual neurons. We propose $\textbf{NeuronMoE}$, a method that analyzes language-specific neurons across all transformer components to guide expert allocation per layer based on empirically measured cross-lingual neuron diversity. Applied to Llama-3.2-3B for low-resource languages (Greek, Turkish, and Hungarian), this approach achieves approximately 40% average parameter reduction while matching the performance of the LayerMoE baseline. We find that low-resource language experts independently develop neuron specialization patterns mirroring the high-resource language, which are concentrated in early and late layers. This reveals potential universal architectural principles in how multilingual models organize linguistic knowledge.

---

### WebFactory: Automated Compression of Foundational Language Intelligence into Grounded Web Agents

Authors: Sicheng Fan, Qingyun Shi, Shengze Xu, Shengbo Cai, Tieyong Zeng, Li Ling, Yanyi Shang, Dehan Kong
Published: 2026-03-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.05044v2

Current paradigms for training GUI agents are fundamentally limited by a reliance on either unsafe, non-reproducible live web interactions or costly, scarce human-crafted data and environments. We argue this focus on data volume overlooks a more critical factor: the efficiency of compressing a large language model's (LLM) latent knowledge into actionable agent behavior. We introduce WebFactory, a novel, fully automated closed-loop reinforcement learning pipeline for GUI agents, systematically compressing LLM-encoded internet intelligence into efficient, grounded actions. Our pipeline features a process of scalable environment synthesis, knowledge-aware task generation, LLM-powered trajectory collection, decomposed reward RL training, and systematic agent evaluation. Remarkably, our agent demonstrates exceptional data efficiency and generalization. Trained on synthetic data from only 10 websites within WebFactory, it achieves performance comparable to GUI agents trained on the same amount of human-annotated data from a much larger set of environments. This superior performance is consistent across our internal offline and online transfer benchmarks, where our agent also significantly outperforms the base foundation model. We further provide critical insights into the "embodiment potential" of different LLM foundations, offering a new axis for model evaluation. This work presents a scalable and cost-effective paradigm for transforming passive internet knowledge into active, grounded intelligence, marking a critical step towards general-purpose interactive agents.

---

### CoIn3D: Revisiting Configuration-Invariant Multi-Camera 3D Object Detection

Authors: Zhaonian Kuang, Rui Ding, Haotian Wang, Xinhu Zheng, Meng Yang, Gang Hua
Published: 2026-03-05
Categories: cs.CV, cs.RO
arXiv: https://arxiv.org/abs/2603.05042v2

Multi-camera 3D object detection (MC3D) has attracted increasing attention with the growing deployment of multi-sensor physical agents, such as robots and autonomous vehicles. However, MC3D models still struggle to generalize to unseen platforms with new multi-camera configurations. Current solutions simply employ a meta-camera for unified representation but lack comprehensive consideration. In this paper, we revisit this issue and identify that the devil lies in spatial prior discrepancies across source and target configurations, including different intrinsics, extrinsics, and array layouts. To address this, we propose CoIn3D, a generalizable MC3D framework that enables strong transferability from source configurations to unseen target ones. CoIn3D explicitly incorporates all identified spatial priors into both feature embedding and image observation through spatial-aware feature modulation (SFM) and camera-aware data augmentation (CDA), respectively. SFM enriches feature space by integrating four spatial representations, such as focal length, ground depth, ground gradient, and Plücker coordinate. CDA improves observation diversity under various configurations via a training-free dynamic novel-view image synthesis scheme. Extensive experiments demonstrate that CoIn3D achieves strong cross-configuration performance on landmark datasets such as NuScenes, Waymo, and Lyft, under three dominant MC3D paradigms represented by BEVDepth, BEVFormer, and PETR.

---

### Exploiting Intermediate Reconstructions in Optical Coherence Tomography for Test-Time Adaption of Medical Image Segmentation

Authors: Thomas Pinetz, Veit Hucke, Hrvoje Bogunovic
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05041v1

Primary health care frequently relies on low-cost imaging devices, which are commonly used for screening purposes. To ensure accurate diagnosis, these systems depend on advanced reconstruction algorithms designed to approximate the performance of high-quality counterparts. Such algorithms typically employ iterative reconstruction methods that incorporate domain-specific prior knowledge. However, downstream task performance is generally assessed using only the final reconstructed image, thereby disregarding the informative intermediate representations generated throughout the reconstruction process. In this work, we propose IRTTA to exploit these intermediate representations at test-time by adapting the normalization-layer parameters of a frozen downstream network via a modulator network that conditions on the current reconstruction timescale. The modulator network is learned during test-time using an averaged entropy loss across all individual timesteps. Variation among the timestep-wise segmentations additionally provides uncertainty estimates at no extra cost. This approach enhances segmentation performance and enables semantically meaningful uncertainty estimation, all without modifying either the reconstruction process or the downstream model.

---

### Enhancing Zero-shot Commonsense Reasoning by Integrating Visual Knowledge via Machine Imagination

Authors: Hyuntae Park, Yeachan Kim, SangKeun Lee
Published: 2026-03-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.05040v1

Recent advancements in zero-shot commonsense reasoning have empowered Pre-trained Language Models (PLMs) to acquire extensive commonsense knowledge without requiring task-specific fine-tuning. Despite this progress, these models frequently suffer from limitations caused by human reporting biases inherent in textual knowledge, leading to understanding discrepancies between machines and humans. To bridge this gap, we introduce an additional modality to enrich the reasoning capabilities of PLMs. We propose Imagine (Machine Imagination-based Reasoning), a novel zero-shot commonsense reasoning framework that supplements textual inputs with visual signals from machine-generated images. Specifically, we enhance PLMs with the ability to imagine by embedding an image generator directly into the reasoning pipeline. To facilitate effective utilization of this imagined visual context, we construct synthetic datasets designed to emulate visual question-answering scenarios. Through comprehensive evaluations on multiple commonsense reasoning benchmarks, we demonstrate that Imagine substantially outperforms existing zero-shot approaches and even surpasses advanced large language models. These results underscore the capability of machine imagination to mitigate reporting bias and significantly enhance the generalization ability of commonsense reasoning models

---

### Generalizable Multiscale Segmentation of Heterogeneous Map Collections

Authors: Remi Petitpierre
Published: 2026-03-05
Categories: cs.CV
arXiv: https://arxiv.org/abs/2603.05037v1

Historical map collections are highly diverse in style, scale, and geographic focus, often consisting of many single-sheet documents. Yet most work in map recognition focuses on specialist models tailored to homogeneous map series. In contrast, this article aims to develop generalizable semantic segmentation models and ontology. First, we introduce Semap, a new open benchmark dataset comprising 1,439 manually annotated patches designed to reflect the variety of historical map documents. Second, we present a segmentation framework that combines procedural data synthesis with multiscale integration to improve robustness and transferability. This framework achieves state-of-the-art performance on both the HCMSSD and Semap datasets, showing that a diversity-driven approach to map recognition is not only viable but also beneficial. The results indicate that segmentation performance remains largely stable across map collections, scales, geographic regions, and publication contexts. By proposing benchmark datasets and methods for the generic segmentation of historical maps, this work opens the way to integrating the long tail of cartographic archives to historical geographic studies.

---

### The Trilingual Triad Framework: Integrating Design, AI, and Domain Knowledge in No-code AI Smart City Course

Authors: Qian Huang, King Wang Poon
Published: 2026-03-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.05036v1

This paper introduces the "Trilingual Triad" framework, a model that explains how students learn to design with generative artificial intelligence (AI) through the integration of Design, AI, and Domain Knowledge. As generative AI rapidly enters higher education, students often engage with these systems as passive users of generated outputs rather than active creators of AI-enabled knowledge tools. This study investigates how students can transition from using AI as a tool to designing AI as a collaborative teammate. The research examines a graduate course, Creating the Frontier of No-code Smart Cities at the Singapore University of Technology and Design (SUTD), in which students developed domain-specific custom GPT systems without coding. Using a qualitative multi-case study approach, three projects - the Interview Companion GPT, the Urban Observer GPT, and Buddy Buddy - were analyzed across three dimensions: design, AI architecture, and domain expertise. The findings show that effective human-AI collaboration emerges when these three "languages" are orchestrated together: domain knowledge structures the AI's logic, design mediates human-AI interaction, and AI extends learners' cognitive capacity. The Trilingual Triad framework highlights how building AI systems can serve as a constructionist learning process that strengthens AI literacy, metacognition, and learner agency.

---

### Good-Enough LLM Obfuscation (GELO)

Authors: Anatoly Belikov, Ilya Fedotov
Published: 2026-03-05
Categories: cs.CR, cs.LG
arXiv: https://arxiv.org/abs/2603.05035v3

Large Language Models (LLMs) are increasingly served on shared accelerators where an adversary with read access to device memory can observe KV caches and hidden states, threatening prompt privacy for open-source models. Cryptographic protections such as MPC and FHE offer strong guarantees but remain one to two orders of magnitude too slow for interactive inference, while static obfuscation schemes break under multi-run statistical attacks once the model is known. We present GELO (Good-Enough LLM Obfuscation), a lightweight protocol for privacy-preserving inference that limits information leakage from untrusted accelerator observations by hiding hidden states with fresh, per-batch invertible mixing. For each offloaded projection, the TEE samples a random matrix $A$, forms $U = AH$, offloads $U$ and weights W to the accelerator, and then applies $A^{-1}$ on return, so that $A^{-1}((AH)W ) = HW$ and outputs are unchanged. Because mixing is never reused across batches, the attacker faces only a single-batch blind source separation problem. We analyse information leakage and introduce two practical defences: (i) non-orthogonal mixing to mask Gram matrices, and (ii) orthogonal mixing augmented with a small fraction of high-energy "shield" vectors that pollute higher-order statistics. On Llama-2 7B, GELO preserves float32 outputs exactly, closely matches low-precision baselines, and shows about $20$--$30\%$ compute-side overhead in a controlled offload microbenchmark; an unoptimized remote prototype is dominated by transport overhead, motivating deeper serving-engine integration. GELO resists ICA/BSS and anchor-assisted attacks; a 60M-parameter transformer-based unmixing attack also fails under strong mixing and shielding.

---

### AegisUI: Behavioral Anomaly Detection for Structured User Interface Protocols in AI Agent Systems

Authors: Mohd Safwan Uddin, Saba Hajira
Published: 2026-03-05
Categories: cs.AI
arXiv: https://arxiv.org/abs/2603.05031v1

AI agents that build user interfaces on the fly assembling buttons, forms, and data displays from structured protocol payloads are becoming common in production systems. The trouble is that a payload can pass every schema check and still trick a user: a button might say "View invoice" while its hidden action wipes an account, or a display widget might quietly bind to an internal salary field. Current defenses stop at syntax; they were never built to catch this kind of behavioral mismatch. We built AegisUI to study exactly this gap. The framework generates structured UI payloads, injects realistic attacks into them, extracts numeric features, and benchmarks anomaly detectors end-to-end. We produced 4000 labeled payloads (3000 benign, 1000 malicious) spanning five application domains and five attack families: phishing interfaces, data leakage, layout abuse, manipulative UI, and workflow anomalies. From each payload we extracted 18 features covering structural, semantic, binding, and session dimensions, then compared three detectors: Isolation Forest (unsupervised), a benign-trained autoencoder (semi-supervised), and Random Forest (supervised). On a stratified 80/20 split, Random Forest scored best overall (accuracy 0.931, precision 0.980, recall 0.740, F1 0.843, ROC-AUC 0.952). The autoencoder came second (F1 0.762, ROC-AUC 0.863) and needs no malicious labels at training time, which matters when deploying a new system that lacks attack history. Per-attack-type analysis showed that layout abuse is easiest to catch while manipulative UI payloads are hardest. All code, data, and configurations are released for full reproducibility.

---