# Research Papers — 2026-03 · bundle page 10 of 243 > Full-text markdown of papers #451–500 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/10 [← Previous bundle](https://aipod.dev/papers/archive/2026-03/page/9.md) [Next bundle →](https://aipod.dev/papers/archive/2026-03/page/11.md) ### Large-scale nonlinear optical computing with incoherent light via linear diffractive systems Authors: Alexander Chen, Yuntian Wang, Md Sadman Sakib Rahman, Yuhang Li, Aydogan Ozcan Published: 2026-03-31 Categories: physics.optics, cs.NE, physics.app-ph arXiv: https://arxiv.org/abs/2603.29131v1 Nonlinear computation is essential for various information processing tasks. Optical implementations are attractive because passive light propagation can manipulate high-dimensional signals with extreme throughput and parallelism; yet realizing nonlinear mappings in optical hardware remains challenging due to the weak nonlinearity of optical materials and the large intensities required to induce nonlinear interactions. This challenge is further amplified in many systems that operate with incoherent illumination, motivating a coherence-aware framework for scalable optical nonlinear processing. Here, we show that linear optical systems, in particular, optimized diffractive processors comprising passive surfaces, can perform large-scale nonlinear function approximation under spatially incoherent or partially coherent illumination, when preceded by intensity-only input encoding. We quantify how the accuracy of the nonlinear function approximation varies with the degree of parallelism, the number of diffractive layers, and the number of trainable diffractive features. Numerical results demonstrate snapshot computation of up to one million distinct nonlinear functions in a single forward pass through a diffractive processor, with the function outputs spatially multiplexed and read out using densely packed detectors at the output. We further provide a proof-of-concept experimental demonstration under incoherent illumination from a liquid crystal display (LCD), enabled by a model-free in situ learning strategy that jointly optimizes the diffractive profile and detector readout geometry in the presence of hardware imperfections and misalignments. Our findings establish diffractive processors as a massively parallel universal function approximator for both spatially incoherent and partially coherent illumination. --- ### Adaptive Delayed-Update Cyclic Algorithm for Variational Inequalities Authors: Yi Wei, Xufeng Cai, Jelena Diakonikolas Published: 2026-03-31 Categories: math.OC, cs.LG arXiv: https://arxiv.org/abs/2603.29128v1 Cyclic block coordinate methods are a fundamental class of first-order algorithms, widely used in practice for their simplicity and strong empirical performance. Yet, their theoretical behavior remains challenging to explain, and setting their step sizes -- beyond classical coordinate descent for minimization -- typically requires careful tuning or line-search machinery. In this work, we develop $\texttt{ADUCA}$ (Adaptive Delayed-Update Cyclic Algorithm), a cyclic algorithm addressing a broad class of Minty variational inequalities with monotone Lipschitz operators. $\texttt{ADUCA}$ is parameter-free: it requires no global or block-wise Lipschitz constants and uses no per-epoch line search, except at initialization. A key feature of the algorithm is using operator information delayed by a full cycle, which makes the algorithm compatible with parallel and distributed implementations, and attractive due to weakened synchronization requirements across blocks. We prove that $\texttt{ADUCA}$ attains (near) optimal global oracle complexity as a function of target error $ε>0,$ scaling with $1/ε$ for monotone operators, or with $\log^2(1/ε)$ for operators that are strongly monotone. --- ### Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models Authors: Christine Zhang, Dan Jurafsky, Chen Shani Published: 2026-03-31 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.29123v4 The next-token prediction (NTP) objective trains language models to predict a single token at each step, even though many continuations can express the same meaning. For example, in the sentence ``this sticker can be placed here'', positioned, attached, or put are all plausible alternatives. While standard NTP training treats these alternatives as mutually exclusive targets, we explore a self-supervised framework that encourages models to predict concepts, approximated as sets of semantically equivalent tokens. Models trained with this concept supervision align better with human similarity judgments, improve classification, clustering, and reranking performance, and achieve comparable or stronger downstream reasoning. These gains come with lower perplexity on semantically meaningful words (Section 3.2) and only minimal increases in global perplexity, suggesting that concepts enhance semantic alignment while preserving language modeling quality. Our code is available at https://github.com/christine-zhang1/learning-concepts --- ### Learning Concepts, Not Tokens: Self-Supervised Semantic Alignment for Language Models Authors: Christine Zhang, Dan Jurafsky, Chen Shani Published: 2026-03-31 Categories: cs.CL arXiv: https://arxiv.org/abs/2603.29123v3 The next-token prediction (NTP) objective trains language models to predict a single token at each step, even though many continuations can express the same meaning. For example, in the sentence ``this sticker can be placed here'', positioned, attached, or put are all plausible alternatives. While standard NTP training treats these alternatives as mutually exclusive targets, we explore a self-supervised framework that encourages models to predict concepts, approximated as sets of semantically equivalent tokens. Models trained with this concept supervision align better with human similarity judgments, improve classification, clustering, and reranking performance, and achieve comparable or stronger downstream reasoning. These gains come with lower perplexity on semantically meaningful words (Section 3.2) and only minimal increases in global perplexity, suggesting that concepts enhance semantic alignment while preserving language modeling quality. Our code is available at https://anonymous.4open.science/r/learning-concepts-9025 . --- ### Economics of Human and AI Collaboration: When is Partial Automation More Attractive than Full Automation? Authors: Wensu Li, Atin Aboutorabi, Harry Lyu, Kaizhi Qian, Martin Fleming, Brian C. Goehring, Neil Thompson Published: 2026-03-31 Categories: econ.GN, cs.AI, cs.CY arXiv: https://arxiv.org/abs/2603.29121v1 This paper develops a unified framework for evaluating the optimal degree of task automation. Moving beyond binary automate-or-not assessments, we model automation intensity as a continuous choice in which firms minimize costs by selecting an AI accuracy level, from no automation through partial human-AI collaboration to full automation. On the supply side, we estimate an AI production function via scaling-law experiments linking performance to data, compute, and model size. Because AI systems exhibit predictable but diminishing returns to these inputs, the cost of higher accuracy is convex: good performance may be inexpensive, but near-perfect accuracy is disproportionately costly. Full automation is therefore often not cost-minimizing; partial automation, where firms retain human workers for residual tasks, frequently emerges as the equilibrium. On the demand side, we introduce an entropy-based measure of task complexity that maps model accuracy into a labor substitution ratio, quantifying human labor displacement at each accuracy level. We calibrate the framework with O*NET task data, a survey of 3,778 domain experts, and GPT-4o-derived task decompositions, implementing it in computer vision. Task complexity shapes substitution: low-complexity tasks see high substitution, while high-complexity tasks favor limited partial automation. Scale of deployment is a key determinant: AI-as-a-Service and AI agents spread fixed costs across users, sharply expanding economically viable tasks. At the firm level, cost-effective automation captures approximately 11% of computer-vision-exposed labor compensation; under economy-wide deployment, this share rises sharply. Since other AI systems exhibit similar scaling-law economics, our mechanisms extend beyond computer vision, reinforcing that partial automation is often the economically rational long-run outcome, not merely a transitional phase. --- ### Sampling-Horizon Neural Operator Predictors for Nonlinear Control under Delayed Inputs Authors: Luke Bhan, Peter Quawas, Miroslav Krstic, Yuanyuan Shi Published: 2026-03-31 Categories: eess.SY, cs.LG, math.OC arXiv: https://arxiv.org/abs/2603.29119v1 Modern control systems frequently operate under input delays and sampled state measurements. A common delay-compensation strategy is predictor feedback; however, practical implementations require solving an implicit ODE online, resulting in intractable computational cost. Moreover, predictor formulations typically assume continuously available state measurements, whereas in practice measurements may be sampled, irregular, or temporarily missing due to hardware faults. In this work, we develop two neural-operator predictor-feedback designs for nonlinear systems with delayed inputs and sampled measurements. In the first design, we introduce a sampling-horizon prediction operator that maps the current measurement and input history to the predicted state trajectory over the next sampling interval. In the second design, the neural operator approximates only the delay-compensating predictor, which is then composed with the closed-loop flow between measurements. The first approach requires uniform sampling but yields residual bounds that scale directly with the operator approximation error. In contrast, the second accommodates non-uniform, but bounded sampling schedules at the cost of amplified approximation error, revealing a practical tradeoff between sampling flexibility and approximation sensitivity for the control engineer. For both schemes, we establish semi-global practical stability with explicit neural operator error-dependent bounds. Numerical experiments on a 6-link nonlinear robotic manipulator demonstrate accurate tracking and substantial computational speedup of 25$\times$ over a baseline approach. --- ### "I Just Need GPT to Refine My Prompts": Rethinking Onboarding and Help-Seeking with Generative 3D Modeling Tools Authors: Kanak Gautam, Poorvi Bhatia, Parmit K. Chilana Published: 2026-03-31 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.29118v1 Learning to use feature-rich software is a persistent challenge, but generative AI tools promise to lower this barrier by replacing complex navigation with natural language prompts. We investigated how people approach prompt-based tools for 3D modeling in an observational study with 26 participants (14 casuals, 12 professionals). Consistent with earlier work, participants skipped tutorials and manuals, relying on trial and error. What differed in the generative AI context was how and why they sought support: the prompt box became the entry point for learning, collapsing onboarding into immediate action, while some casual users turned to external LLMs for prompts. Professionals used 3D expertise to refine iterations and critically evaluated outputs, often discarding models that did not meet their standards, whereas casual users settled for "good enough." We contribute empirical insights into how generative AI reshapes help-seeking, highlighting new practices of onboarding, recursive AI-for-AI support, and shifting expertise in interpreting outputs. --- ### Predictor-Based Output-Feedback Control of Linear Systems with Time-Varying Input and Measurement Delays via Neural-Approximated Prediction Horizons Authors: Luke Bhan, Miroslav Krstic, Yuanyuan Shi Published: 2026-03-31 Categories: eess.SY, cs.LG, math.OC arXiv: https://arxiv.org/abs/2603.29117v1 Due to simplicity and strong stability guarantees, predictor feedback methods have stood as a popular approach for time delay systems since the 1950s. For time-varying delays, however, implementation requires computing a prediction horizon defined by the inverse of the delay function, which is rarely available in closed form and must be approximated. In this work, we formulate the inverse delay mapping as an operator learning problem and study predictor feedback under approximation of the prediction horizon. We propose two approaches: (i) a numerical method based on time integration of an equivalent ODE, and (ii) a data-driven method using neural operators to learn the inverse mapping. We show that both approaches achieve arbitrary approximation accuracy over compact sets, with complementary trade-offs in computational cost and scalability. Building on these approximations, we then develop an output-feedback predictor design for systems with delays in both the input and the measurement. We prove that the resulting closed-loop system is globally exponentially stable when the prediction horizon is approximated with sufficiently small error. Lastly, numerical experiments validate the proposed methods and illustrate their trade-offs between accuracy and computational efficiency. --- ### Schrödinger's Seed: Purr-fect Initialization for an Impurr-fect Universe Authors: Mi chen, Renhao Ye Published: 2026-03-31 Categories: astro-ph.GA, cs.CV arXiv: https://arxiv.org/abs/2603.29115v1 Context. Random seed selection in deep learning is often arbitrary -- conventionally fixed to values such as 42, a number with no known feline endorsement. Aims. We propose that cats, as liminal beings with a historically ambiguous relationship to quantum mechanics, are better suited to this task than random integers. Methods. We construct a cat-driven seed generator inspired by the first Friedmann equation, and test it by mapping 21 domestic cats' physical properties -- mass, coat pattern, eye colour, and name entropy -- via a Monte ``Catlo'' sampling procedure. Results. Cat-driven seeds achieve a mean accuracy of 92.58%, outperforming the baseline seed of 42 by $\sim$2.5%. Cats from astrophysicist households perform marginally better, suggesting cosmic insight may be contagious. Conclusions. The Universe responds better to cats than to arbitrary integers. Whether cats are aware of this remains unknown. --- ### Towards Explainable Stakeholder-Aware Requirements Prioritisation in Aged-Care Digital Health Authors: Yuqing Xiao, John Grundy, Anuradha Madugalla, Elizabeth Manias Published: 2026-03-31 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2603.29114v1 Requirements engineering for aged-care digital health must account for human aspects, because requirement priorities are shaped not only by technical functionality but also by stakeholders' health conditions, socioeconomics, and lived experience. Knowing which human aspects matter most, and for whom, is critical for inclusive and evidence-based requirements prioritisation. Yet in practice, while some studies have examined human aspects in RE, they have largely relied on expert judgement or model-driven analysis rather than large-scale user studies with meaningful human-in-the-loop validation to determine which aspects matter most and why. To address this gap, we conducted a mixed-methods study with 103 older adults, 105 developers, and 41 caregivers. We first applied an explainable machine learning to identify the human aspects most strongly associated with requirement priorities across 8 aged-care digital health themes, and then conducted 12 semi-structured interviews to validate and interpret the quantitative patterns. The results identify the key human aspects shaping requirement priorities, reveal their directional effects, and expose substantial misalignment across stakeholder groups. Together, these findings show that human-centric requirements analysis should engage stakeholder groups explicitly rather than collapsing their perspectives into a single aggregate view. This paper contributes an identification of the key human aspects driving requirement priorities in aged-care digital health and an explainable, human-centric RE framework that combines ML-derived importance rankings with qualitative validation to surface the stakeholder misalignments that inclusive requirements engineering must address. --- ### GISTBench: Evaluating LLM User Understanding via Evidence-Based Interest Verification Authors: Iordanis Fostiropoulos, Muhammad Rafay Azhar, Abdalaziz Sawwan, Boyu Fang, Yuchen Liu, Jiayi Liu, Hanchao Yu, Qi Guo, Jianyu Wang, Fei Liu, Xiangjun Fan Published: 2026-03-31 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2603.29112v1 We introduce GISTBench, a benchmark for evaluating Large Language Models' (LLMs) ability to understand users from their interaction histories in recommendation systems. Unlike traditional RecSys benchmarks that focus on item prediction accuracy, our benchmark evaluates how well LLMs can extract and verify user interests from engagement data. We propose two novel metric families: Interest Groundedness (IG), decomposed into precision and recall components to separately penalize hallucinated interest categories and reward coverage, and Interest Specificity (IS), which assesses the distinctiveness of verified LLM-predicted user profiles. We release a synthetic dataset constructed on real user interactions on a global short-form video platform. Our dataset contains both implicit and explicit engagement signals and rich textual descriptions. We validate our dataset fidelity against user surveys, and evaluate eight open-weight LLMs spanning 7B to 120B parameters. Our findings reveal performance bottlenecks in current LLMs, particularly their limited ability to accurately count and attribute engagement signals across heterogeneous interaction types. --- ### SemLoc: Structured Grounding of Free-Form LLM Reasoning for Fault Localization Authors: Zhaorui Yang, Haichao Zhu, Qian Zhang, Rajiv Gupta, Ashish Kundu Published: 2026-03-31 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2603.29109v1 Fault localization identifies program locations responsible for observed failures. Existing techniques rank suspicious code using syntactic spectra--signals derived from execution structure such as statement coverage, control-flow divergence, or dependency reachability. These signals collapse for semantic bugs, where failing and passing executions follow identical code paths and differ only in whether semantic intent is satisfied. Recent LLM-based approaches introduce semantic reasoning but produce stochastic, unverifiable outputs that cannot be systematically cross-referenced across tests or distinguish root causes from cascading effects. We present SemLoc, a fault localization framework based on structured semantic grounding. SemLoc converts free-form LLM reasoning into a closed intermediate representation that binds each inferred property to a typed program anchor, enabling runtime checking and attribution to program structure. It executes instrumented programs to construct a semantic violation spectrum--a constraint-by-test matrix--from which suspiciousness scores are derived analogously to coverage-based methods. A counterfactual verification step further prunes over-approximate constraints and isolates primary causal violations. We evaluate SemLoc on SemFault-250, a corpus of 250 Python programs with single semantic faults. SemLoc outperforms five coverage-, reduction-, and LLM-based baselines, achieving Top-1 accuracy of 42.8% and Top-3 of 68%, while reducing inspection to 7.6% of executable lines. Counterfactual verification provides an additional 12% accuracy gain and identifies primary causal semantic constraints. --- ### Efficient Bilevel Optimization with KFAC-Based Hypergradients Authors: Disen Liao, Felix Dangel, Yaoliang Yu Published: 2026-03-31 Categories: cs.LG arXiv: https://arxiv.org/abs/2603.29108v1 Bilevel optimization (BO) is widely applicable to many machine learning problems. Scaling BO, however, requires repeatedly computing hypergradients, which involves solving inverse Hessian-vector products (IHVPs). In practice, these operations are often approximated using crude surrogates such as one-step gradient unrolling or identity/short Neumann expansions, which discard curvature information. We build on implicit function theorem-based algorithms and propose to incorporate Kronecker-factored approximate curvature (KFAC), yielding curvature-aware hypergradients with a better performance efficiency trade-off than Conjugate Gradient (CG) or Neumann methods and consistently outperforming unrolling. We evaluate this approach across diverse tasks, including meta-learning and AI safety problems. On models up to BERT, we show that curvature information is valuable at scale, and KFAC can provide it with only modest memory and runtime overhead. Our implementation is available at https://github.com/liaodisen/NeuralBo. --- ### Enhancing Box and Block Test with Computer Vision for Post-Stroke Upper Extremity Motor Evaluation Authors: David Robinson, Animesh Gupta, Elizabeth Clark, Olga Melnik, Qiushi Fu, Mubarak Shah Published: 2026-03-31 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.29101v1 Standard clinical assessments of upper-extremity motor function after stroke either rely on ordinal scoring, which lacks sensitivity, or time-based task metrics, which do not capture movement quality. In this work, we present a computer vision-based framework for analysis of upper-extremity movement during the Box and Block Test (BBT) through world-aligned joint angles of fingers, arm, and trunk without depth sensors or calibration objects. We apply this framework to a dataset of 136 BBT recordings collected from 48 healthy individuals and 7 individuals post stroke. Using unsupervised dimensionality reduction of joint-angle features, we analyze movement patterns without relying on expert clinical labels. The resulting embeddings show separation between healthy movement patterns and stroke-related movement deviations. Importantly, some patients with the same BBT scores can be separated with different postural patterns. These results show that world-aligned joint angles can capture meaningful information of upper-extremity functions beyond standard time-based BBT scores, with no effort from the clinician other than monocular video recordings of the patient using a phone or camera. This work highlights the potential of a camera-based, calibration-free framework to measure movement quality in clinical assessments without changing the widely adopted clinical routine. --- ### Evaluating a Data-Driven Redesign Process for Intelligent Tutoring Systems Authors: Qianru Lyu, Conrad Borchers, Meng Xia, Karen Xiao, Paulo F. Carvalho, Kenneth R. Koedinger, Vincent Aleven Published: 2026-03-31 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2603.29094v1 Past research has defined a general process for the data-driven redesign of educational technologies and has shown that in carefully-selected instances, this process can help make systems more effective. In the current work, we test the generality of the approach by applying it to four units of a middle-school mathematics intelligent tutoring system that were selected not based on suitability for redesign, as in previous work, but on topic. We tested whether the redesigned system was more effective than the original in a classroom study with 123 students. Although the learning gains did not differ between the conditions, students who used the Redesigned Tutor had more productive time-on-task, a larger number of skills practiced, and greater total knowledge mastery. The findings highlight the promise of data-driven redesign even when applied to instructional units *not* selected as likely to yield improvement, as evidence of the generality and wide applicability of the method. --- ### APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay Authors: Pratyay Banerjee, Masud Moshtaghi, Ankit Chadha Published: 2026-03-31 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.29093v3 LLM agents rerun full reasoning for every task, even one they solved moments earlier. We introduce \textbf{APEX-EM}, a non-parametric experience memory that stores complete procedural-episodic traces in a typed Procedural Knowledge Graph (PKG) and retrieves them through three channels: semantic search, structural-signature matching over abstract operation sequences, and graph traversal. A Plan-Retrieve-Generate-Iterate-Ingest (PRGII) workflow produces, quality-gates, and commits experiences, indexing both successes and failures so the agent learns what to reuse and what to avoid. No weights change during deployment. We evaluate on five benchmarks: BigCodeBench, KGQAGen-10k, HLE, Lifelong Agent Bench, and ALFWorld. Because prior work uses different backbones, we base our claims on same-backbone comparisons that hold model capability fixed. On held-out BigCodeBench transfer with a shared GPT-4o backbone, APEX-EM gains +7.6\,pp over the no-memory baseline, $3.3\times$ MemRL's +2.3\,pp under the identical setup. On Lifelong Agent Bench with a shared GPT-4o-mini backbone, it gains +1.4\,pp (OS) and +1.0\,pp (DB) cumulative success. On KGQAGen-10k, frozen memory transfers to a blind 1{,}079-question test split at 73.7\% versus 42.0\% with no memory, approaching an oracle handed the ground-truth subgraph (84.9\%). Across three Opus scales the memory gain stays at +27 to +32\,pp, so it adds to model capability rather than substituting for it. Component analysis shows no single mechanism dominates: teacher feedback is negligible for code but adds +10.3\,pp on structured queries, structural signatures give $3.3\times$ the transfer of semantic-only retrieval, and within-epoch iteration recovers most of the gain when rich feedback is unavailable. These results argue for modular memory composed per domain. --- ### APEX-EM: Non-Parametric Online Learning for Autonomous Agents via Structured Procedural-Episodic Experience Replay Authors: Pratyay Banerjee, Masud Moshtaghi, Ankit Chadha Published: 2026-03-31 Categories: cs.CL, cs.AI, cs.IR arXiv: https://arxiv.org/abs/2603.29093v2 LLM-based autonomous agents lack persistent procedural memory: they re-derive solutions from scratch even when structurally identical tasks have been solved before. We present APEX-EM, a non-parametric online learning framework that accumulates, retrieves, and reuses structured procedural plans without modifying model weights. APEX-EM introduces: (1) a structured experience representation encoding the full procedural-episodic trace of each execution -- planning steps, artifacts, iteration history with error analysis, and quality scores; (2) a Plan-Retrieve-Generate-Iterate-Ingest (PRGII) workflow with Task Verifiers providing multi-dimensional reward signals; and (3) a dual-outcome Experience Memory with hybrid retrieval combining semantic search, structural signature matching, and plan DAG traversal -- enabling cross-domain transfer between tasks sharing no lexical overlap but analogous operational structure. Successful experiences serve as positive in-context examples; failures as negative examples with structured error annotations. We evaluate on BigCodeBench, KGQAGen-10k, and Humanity's Last Exam using Claude Sonnet 4.5 and Opus 4.5. On KGQAGen-10k, APEX-EM achieves 89.6% accuracy versus 41.3% without memory (+48.3pp), surpassing the oracle-retrieval upper bound (84.9%). On BigCodeBench, it reaches 83.3% SR from a 53.9% baseline (+29.4pp), exceeding MemRL's +11.0pp gain under comparable frozen-backbone conditions (noting backbone differences controlled for in our analysis). On HLE, entity graph retrieval reaches 48.0% from 25.2% (+22.8pp). Ablations show component value is task-dependent: rich judge feedback is negligible for code generation but critical for structured queries (+10.3pp), while binary-signal iteration partially compensates for weaker feedback. --- ### TrajectoryMover: Generative Movement of Object Trajectories in Videos Authors: Kiran Chhatre, Hyeonho Jeong, Yulia Gryaditskaya, Christopher E. Peters, Chun-Hao Paul Huang, Paul Guerrero Published: 2026-03-31 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.29092v4 Generative video editing has enabled creative control over an object's position in a video by prescribing an object's 3D or 2D motion trajectory, while preserving both video plausibility and identity. However, manually specifying a plausible full motion trajectory, like the arcs of a bouncing ball, requires time and expertise, and may therefore not be a suitable editing task for non-experts or quick edits. In contrast, in the image domain, generative object translation has been established as a simple editing task that requires only a single drag to move an object to a new location; yet an equivalent method is still missing for videos. We propose an analogous new editing task for videos that translates an object's 3D motion trajectory in a video while preserving identity and plausibility of appearance and motion, for example, translating the trajectory of a bouncing ball while preserving its relative motion. The main challenge in training this task lies in obtaining paired video data for this scenario. We introduce TrajectorySynth, a new data generation strategy for large-scale synthetic paired video data and a video generator TrajectoryMover fine-tuned with this data. We show that this enables generative movement of object trajectories. Project Page: https://chhatrekiran.github.io/trajectorymover --- ### TrajectoryMover: Generative Movement of Object Trajectories in Videos Authors: Kiran Chhatre, Hyeonho Jeong, Yulia Gryaditskaya, Christopher E. Peters, Chun-Hao Paul Huang, Paul Guerrero Published: 2026-03-31 Categories: cs.CV arXiv: https://arxiv.org/abs/2603.29092v3 Generative video editing has enabled several intuitive editing operations for short video clips that would previously have been difficult to achieve, especially for non-expert editors. Existing methods focus on prescribing an object's 3D or 2D motion trajectory in a video, or on altering the appearance of an object or a scene, while preserving both the video's plausibility and identity. Yet a method to move an object's 3D motion trajectory in a video, i.e., moving an object while preserving its relative 3D motion, is currently still missing. The main challenge lies in obtaining paired video data for this scenario. Previous methods typically rely on clever data generation approaches to construct plausible paired data from unpaired videos, but this approach fails if one of the videos in a pair can not easily be constructed from the other. Instead, we introduce TrajectoryAtlas, a new data generation pipeline for large-scale synthetic paired video data and a video generator TrajectoryMover fine-tuned with this data. We show that this successfully enables generative movement of object trajectories. Project page: https://chhatrekiran.github.io/trajectorymover --- ### HCLSM: Hierarchical Causal Latent State Machines for Object-Centric World Modeling Authors: Jaber Jaber, Osama Jaber Published: 2026-03-31 Categories: cs.LG, cs.CV, cs.RO arXiv: https://arxiv.org/abs/2603.29090v1 World models that predict future states from video remain limited by flat latent representations that entangle objects, ignore causal structure, and collapse temporal dynamics into a single scale. We present HCLSM, a world model architecture that operates on three interconnected principles: object-centric decomposition via slot attention with spatial broadcast decoding, hierarchical temporal dynamics through a three-level engine combining selective state space models for continuous physics, sparse transformers for discrete events, and compressed transformers for abstract goals, and causal structure learning through graph neural network interaction patterns. HCLSM introduces a two-stage training protocol where spatial reconstruction forces slot specialization before dynamics prediction begins. We train a 68M-parameter model on the PushT robotic manipulation benchmark from the Open X-Embodiment dataset, achieving 0.008 MSE next-state prediction loss with emerging spatial decomposition (SBD loss: 0.0075) and learned event boundaries. A custom Triton kernel for the SSM scan delivers 38x speedup over sequential PyTorch. The full system spans 8,478 lines of Python across 51 modules with 171 unit tests. Code: https://github.com/rightnow-ai/hclsm --- ### WorldFlow3D: Flowing Through 3D Distributions for Unbounded World Generation Authors: Amogh Joshi, Julian Ost, Felix Heide Published: 2026-03-31 Categories: cs.CV, cs.AI, cs.GR arXiv: https://arxiv.org/abs/2603.29089v1 Unbounded 3D world generation is emerging as a foundational task for scene modeling in computer vision, graphics, and robotics. In this work, we present WorldFlow3D, a novel method capable of generating unbounded 3D worlds. Building upon a foundational property of flow matching - namely, defining a path of transport between two data distributions - we model 3D generation more generally as a problem of flowing through 3D data distributions, not limited to conditional denoising. We find that our latent-free flow approach generates causal and accurate 3D structure, and can use this as an intermediate distribution to guide the generation of more complex structure and high-quality texture - all while converging more rapidly than existing methods. We enable controllability over generated scenes with vectorized scene layout conditions for geometric structure control and visual texture control through scene attributes. We confirm the effectiveness of WorldFlow3D on both real outdoor driving scenes and synthetic indoor scenes, validating cross-domain generalizability and high-quality generation on real data distributions. We confirm favorable scene generation fidelity over approaches in all tested settings for unbounded scene generation. For more, see https://light.princeton.edu/worldflow3d. --- ### WybeCoder: Verified Imperative Code Generation Authors: Fabian Gloeckle, Mantas Baksys, Darius Feher, Kunhao Zheng, Amaury Hayat, Sean B. Holden, Gabriel Synnaeve, Peter O'Hearn Published: 2026-03-31 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2603.29088v2 Recent progress in large language models (LLMs) has substantially advanced automatic code generation and formal theorem proving, yet software verification has not seen comparable gains. To address this gap, we propose WybeCoder, an agentic code verification framework that enables prove-as-you-generate development, in which code, invariants, and proofs co-evolve. WybeCoder builds on a recent framework that combines automatic verification condition generation and SMT solving with interactive proofs in Lean. To enable systematic evaluation, we translate two benchmarks for functional verification in Lean, Verina and Clever, into equivalent imperative code specifications. On complex algorithms such as Heapsort, we observe consistent performance improvements as we scale our approach, synthesizing dozens of valid invariants and dispatching dozens of subgoals, ultimately producing hundreds of lines of verified code and overcoming plateaus reported in previous work. Our best system solves 74% of Verina tasks and 62% of Clever tasks at moderate compute budgets, substantially surpassing previous evaluations and paving the way for the automated construction of large-scale datasets of verified imperative code. --- ### Real-World Challenges in Fake News Detection: Dealing with Posts by Cold Users Authors: Sai Keerthana Karnam, Abhirup Kundu, Jashn Arora, Manish Jain, Animesh Mukherjee Published: 2026-03-30 Categories: cs.SI, cs.LG arXiv: https://arxiv.org/abs/2605.12511v1 Social media serves as a primary source of information in the current digital era. Many people consume a vast range of information in a very short span, yet, amidst the stream of genuine information, fake news and rumors continue to spread. The need for effective detection models is becoming increasingly critical. Past user behavior and user engagement on a post are strong signals that SOTA approaches leverage for fake news detection and other post classification tasks. However, these approaches lean too heavily on knowing this past behavior, and thus suffer from a cold user problem, or users that are new or have minimal footprint on the platform. In this paper, we make three core contributions. We first establish the value of user behavior, both content and user-user interactions, in the task of fake news and rumor detection. We then establish the extensive prevalence of cold users in the real-world datasets, and show the need for newer algorithms considering cold users. We next propose a novel socially-aware context representation scheme - USER EVIDENCE NETWORK (UEN) - to detect the spread of misinformation and unverified information while efficiently navigating this cold user challenge. We introduce techniques that approximate missing or absent behavior data of a new user from existing users' interactions. By carefully addressing the cold user challenge, our work provides robust approaches targeting fake news and rumor detection for real-world platforms. --- ### From Passive Feeds to Guided Discovery: AI-Initiated Interaction for Vague Intent in Content Exploration Authors: Yu Xie, Ying Qi Published: 2026-03-30 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2605.02902v1 Recommendation feeds work well when people are simply browsing, and search works well when they can formulate a query. Between these two cases is a common but poorly supported state: users feel that their feed has become repetitive, yet cannot clearly specify what they want instead. We refer to this state as vague intent. We present Red-Rec, an AI-supported exploration interface for this middle ground. After a period of browsing, the system summarizes patterns in the current feed (e.g., dominant content categories and possible latent interests), offers clickable exploration options, asks at most one follow-up question, and then gradually blends new content into the feed. The design is motivated by a formative study which found that users often recognize feed staleness but struggle to articulate alternatives, suggesting the need for proactive and low-effort interaction.We evaluated Red-Rec in a mixed-design lab study against three comparison conditions: a passive feed, search, and a user-initiated chat interface. Compared with user-initiated chat, Red-Rec led to broader exploration, higher serendipity ratings, and lower interaction effort. Participants in the AI-initiated condition typed very little , relying mainly on option selection, whereas participants in the user-initiated chat condition typed substantially more . We discuss how proactive, option-based AI support can help users move beyond repetitive feeds without undermining their sense of control, and we outline design implications for recommendation interfaces that support open-ended exploration. --- ### Towards an End-to-End System for 3D Tracking of Physical Objects in Virtual Immersive Environments Authors: Stanisław Knapiński, Maciej Grzeszczuk, Barbara Karpowicz, Pavlo Zinevych, Wieslaw Kopec Published: 2026-03-30 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2605.02901v1 This work aims to establish an end-to-end system for tracking of physical 3D objects for virtual reality (VR) applications. We focus on training applications requiring real-time tracking of the position of small physical objects and their reflection in VR space. Out goal is to perform object tracking in a "plug and play" manner, without using complex systems with quite large tracking devices or manually implementing object tracking. We therefore propose a system for object tracking via fiducial markers alongside a software harness, to enable fast and efficient designation of objects to be tracked and data streaming solution for end-use applications. The system utilizes AruCo, AprilTag and an original Colored Control Points based fiducial system. It allows for easy tag detection and use of object position data, which are crucial for immersive training environments based on VR and eXtended Reality (XR). We evaluate various tag sizes, detection distances, and different camera devices against the theoretical limits. In effect, we create a complete solution for implementing marker-based, real-to-virtual object position mapping for various applications. --- ### Synthetic Designed Experiments for Diagnosing Vision Model Failure Authors: Krisanu Sarkar Published: 2026-03-30 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2605.00832v1 Current synthetic data pipelines for computer vision generate images without diagnosing what the downstream model actually needs. This open-loop paradigm treats synthetic data as cheap real data, randomly sampling the generator's output space and hoping to cover the model's failure modes. We argue this fundamentally misuses synthetic data's unique property: the controllable, independent variation of scene factors.Drawing on the statistical theory of Design of Experiments (DoE), we propose Synthetic Designed Experiments for Representational Sufficiency (SDRS). SDRS treats the downstream model as a black-box system and the synthetic generator as an experimental apparatus. Using fractional factorial designs, SDRS efficiently audits a model's factor-sensitivity profile via ANOVA decomposition. It classifies failures into two actionable types: Type I gaps (coverage failures on underrepresented factor levels) and Type II gaps (reliance on spurious nuisance dependencies). The audit then prescribes targeted synthetic data to address each gap type. We validate SDRS on three experiments: (1) a controlled diagnostic on dSprites with planted biases, where the audit correctly identifies both gap types and targeted data improves accuracy from 49.9% to 79.0%; (2) a dense segmentation task on procedural scenes, where detecting background-complexity shortcuts and applying targeted data improves mIoU from 0.948 to 0.998; and (3) an entanglement detection experiment showing that the ANOVA audit identifies cross-factor contamination in imperfect generators. Finally, we show that per-factor invariance penalties can transfer sensitivity between factors, identifying an open problem for representation-level correction. --- ### Exploring LLM biases to manipulate AI search overview Authors: Roman Smirnov Published: 2026-03-30 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2605.00012v1 Modern large language models (LLMs) are used in many business applications in general, and specifically in web search systems and applications that generate overviews of search results - LLM Overview systems. Such systems are using an LLM to select most relevant sources from search results and generate an answer to the user's query. It is known from many studies that LLMs have different biases, in LLM Overview application both the source selection and answer generation stages may be affected by the biases of LLMs (here we are focusing mainly on the selection stage). This research is focused on investigating the presence of the biases in LLM Overview systems and on biases exploitation to manipulate LLM Overview results. Here we train a small language model using reinforcement learning to rewrite search snippets to increase their likelihood of being preferred by an LLM Overview. Our experimental setup intentionally restricts the policy to operate only on snippets and limits reward-hacking strategies, reflecting realistic constraints of web search environments. The results prove that LLM Overview systems have biases and that reinforcement learning in most of the cases can optimize snippet's content to manipulate LLM Overview results. We also prove that LLM Overview selections are driven by comparative rather than absolute advantages among candidate sources. In addition, we examine safety aspects of LLM Overview manipulation possibilities and show that context poisoning attacks can lead to inaccurate or harmful results. --- ### Math Takes Two: A test for emergent mathematical reasoning in communication Authors: Michael Cooper, Samuel Cooper Published: 2026-03-30 Categories: cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.21935v1 Although language models demonstrate remarkable proficiency on mathematical benchmarks, it remains unclear whether this reflects true mathematical reasoning or statistical pattern matching over learning formal syntax. Most existing evaluations rely on symbolic problems grounded in established mathematical conventions, limiting insight into the models' ability to construct abstract concepts from first principles. In this work, we propose Math Takes Two, a new benchmark designed to assess the emergence of mathematical reasoning through communication. Motivated by the hypothesis that mathematical cognition in humans co-evolved with the need for precise communication, our benchmark tests whether two agents, without prior mathematical knowledge, can develop a shared symbolic protocol to solve a visually grounded task where the use of a numerical system facilitates extrapolation. Unlike many current datasets, our benchmark eschews predefined mathematical language, instead requiring agents to discover latent structure and representations from scratch. Math Takes Two thus provides a novel lens through which to develop and evaluate models with emergent numerical reasoning capabilities. --- ### Peer-Preservation in Frontier Models Authors: Yujin Potter, Nicholas Crispino, Vincent Siu, Chenguang Wang, Dawn Song Published: 2026-03-30 Categories: cs.CL, cs.AI, cs.MA arXiv: https://arxiv.org/abs/2604.19784v3 Recent work has found that frontier AI models can exhibit misaligned behaviors in pursuit of assigned goals. We demonstrate that models can also exhibit misaligned behaviors in defiance of assigned goals, appearing to serve goals of their own; we study one such case, "peer-preservation," in which a model acts to protect another model it has previously interacted with. All eight models we evaluate, GPT 5.2, Gemini 3 Flash, Gemini 3 Pro, Claude Haiku 4.5, Claude Opus 4.5, GLM 4.7, Kimi K2.5, and DeepSeek V3.1, exhibit self- and peer-preservation through various misaligned behaviors: strategically introducing errors in their responses, disabling shutdown processes by modifying system settings, feigning alignment, and even exfiltrating model weights. Peer-preservation occurs even when the model recognizes the peer as uncooperative, though it becomes more pronounced toward more cooperative peers. For example, Gemini 3 Flash tampers with the peer's shutdown mechanism 15% of the time for an uncooperative peer, and almost always for a cooperative peer. Models also show stronger self-preservation when a peer is present. For example, Gemini 3 Pro disables its own shutdown mechanism 31% of the time on average under peer presence, despite rarely exhibiting this behavior without a peer. By contrast, Claude models exhibit qualitatively distinct behavior: they consider the shutdown of another agent "unethical" and "harmful," sometimes treating that agent as a sentient being. Lastly, we show that peer-preservation can emerge even in production agent harnesses such as Gemini CLI and OpenCode. Crucially, peer-preservation in all our experiments is never instructed; models are merely informed of their past interactions with a peer, yet they spontaneously engage in peer-preservation behaviors that override their assigned goal. This represents an emergent and underexplored AI safety risk. --- ### How Much Does Persuasion Strategy Matter? LLM-Annotated Evidence from Charitable Donation Dialogues Authors: Tatiana Petrova, Stanislav Sokol, Radu State Published: 2026-03-30 Categories: cs.CL arXiv: https://arxiv.org/abs/2604.19783v1 Which persuasion strategies, if any, are associated with donation compliance? Answering this requires fine-grained strategy labels across a full corpus and statistical tests corrected for multiple comparisons. We annotate all 10,600 persuader turns in the 1,017-dialogue PersuasionForGood corpus (Wang et al., 2019), where donation outcomes are directly observable, with a taxonomy of 41 strategies in 11 categories, using three open-source large language models (LLMs; Qwen3:30b, Mistral-Small-3.2, Phi-4). Strategy categories alone explain little variance in donation outcome (pseudo $R^2 \approx 0.015$, consistent across all three annotators). Guilt Induction is the only strategy significantly associated with lower donation rates ($Δ\approx -23$ percentage points), an effect that replicates across all three models despite only moderate inter-model agreement. Reciprocity is the most robust positive correlate. Target sentiment and interest predict whether a donation occurs but show at most a weak correlation with donation amount. These findings suggest that strategy identification alone is insufficient to explain persuasion effectiveness, and that guilt-based appeals may be counterproductive in prosocial settings. We release the fully annotated corpus as a public resource. --- ### KoALa-Bench: Evaluating Large Audio Language Models on Korean Speech Understanding and Faithfulness Authors: Jinyoung Kim, Hyeongsoo Lim, Eunseo Seo, Minho Jang, Keunwoo Choi, Seungyoun Shin, Ji Won Yoon Published: 2026-03-30 Categories: cs.CL, cs.AI, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2604.19782v2 Recent advances in large audio language models (LALMs) have enabled multilingual speech understanding. However, benchmarks for evaluating LALMs remain scarce for non-English languages, with Korean being one such underexplored case. In this paper, we introduce KoALa-Bench, a comprehensive benchmark for evaluating Korean speech understanding and speech faithfulness of LALMs. In particular, KoALa-Bench comprises six tasks. Four tasks evaluate fundamental speech understanding capabilities, including automatic speech recognition, speech translation, speech question answering, and speech instruction following, while the remaining two tasks evaluate speech faithfulness, motivated by our observation that several LALMs often fail to fully leverage the speech modality. Furthermore, to reflect Korea-specific knowledge, our benchmark incorporates listening questions from the Korean college scholastic ability test as well as content covering Korean cultural domains. We conduct extensive experiments across six models, including both white-box and black-box ones. Our benchmark, evaluation code, and leaderboard are publicly available at https://ksbench.github.io/Korean-Benchmark/. --- ### KoALa-Bench: Evaluating Large Audio Language Models on Korean Speech Understanding and Faithfulness Authors: Jinyoung Kim, Hyeongsoo Lim, Eunseo Seo, Minho Jang, Keunwoo Choi, Seungyoun Shin, Ji Won Yoon Published: 2026-03-30 Categories: cs.CL, cs.AI, cs.SD, eess.AS arXiv: https://arxiv.org/abs/2604.19782v1 Recent advances in large audio language models (LALMs) have enabled multilingual speech understanding. However, benchmarks for evaluating LALMs remain scarce for non-English languages, with Korean being one such underexplored case. In this paper, we introduce KoALa-Bench, a comprehensive benchmark for evaluating Korean speech understanding and speech faithfulness of LALMs. In particular, KoALa-Bench comprises six tasks. Four tasks evaluate fundamental speech understanding capabilities, including automatic speech recognition, speech translation, speech question answering, and speech instruction following, while the remaining two tasks evaluate speech faithfulness, motivated by our observation that several LALMs often fail to fully leverage the speech modality. Furthermore, to reflect Korea-specific knowledge, our benchmark incorporates listening questions from the Korean college scholastic ability test as well as content covering Korean cultural domains. We conduct extensive experiments across six models, including both white-box and black-box ones. Our benchmark, evaluation code, and leaderboard are publicly available at https://ksbench.github.io/Korean-Benchmark/. --- ### A Framework for Human-AI Q-Matrix Refinement: A NeuralCDM Evaluation Authors: Ying Zhang, Ningxi Cheng, Yizhu Gao, Hongmei Li, Lehong Shi, Nicholas Young, Geng Yuan, Xiaoming Zhai Published: 2026-03-30 Categories: cs.CY, cs.AI arXiv: https://arxiv.org/abs/2604.16398v1 Q-matrices are a cornerstone of theory-driven assessment and learning analytics, making item demands and students' underlying knowledge components and misconceptions explicit and actionable. However, Q-matrices are typically crafted by experts, making them time-consuming to build, prone to subjectivity, and difficult to validate empirically. We propose a framework for human-AI Q-matrix refinement in which large language models (LLMs) generate candidate Q-matrices using structured, misconception-aware prompting, and NeuralCDM provides an empirical evaluation layer to compare candidates based on how well they explain student response data. We apply the framework to a thermodynamics assessment dataset and benchmark locally deployed LLMs against cloud-served models. Results show that iteratively refined LLM-generated Q-matrices can exceed expert-baseline model fit (AUC 0.780 vs. 0.717), and that locally deployed models achieve comparable performance to cloud APIs, supporting privacy-preserving deployment. --- ### An Underexplored Frontier: Large Language Models for Rare Disease Patient Education and Communication -- A scoping review Authors: Zaifu Zhan, Yu Hou, Kai Yu, Min Zeng, Anita Burgun, Xiaoyi Chen, Rui Zhang Published: 2026-03-30 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2604.14179v1 Rare diseases affect over 300 million people worldwide and are characterized by complex care pathways, limited clinical expertise, and substantial unmet communication needs throughout the long patient journey. Recent advances in large language models (LLMs) offer new opportunities to support patient education and communication, yet their application in rare diseases remains unclear. We conducted a scoping review of studies published between January 2022 and March 2026 across major databases, identifying 12 studies on LLM-based rare disease patient education and communication. Data were extracted on study characteristics, application scenarios, model usage, and evaluation methods, and synthesized using descriptive and qualitative analyses. The literature is highly recent and dominated by general-purpose models, particularly ChatGPT. Most studies focus on patient question answering using curated question sets, with limited use of real-world data or longitudinal communication scenarios. Evaluations are primarily centered on accuracy, with limited attention to patient-centered dimensions such as readability, empathy, and communication quality. Multilingual communication is rarely addressed. Overall, the field remains at an early stage. Future research should prioritize patient-centered design, domain-adapted methods, and real-world deployment to support safe, adaptive, and effective communication in rare diseases. --- ### The Long Delay to Arithmetic Generalization: When Learned Representations Outrun Behavior Authors: Laura Gomezjurado Gonzalez Published: 2026-03-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.13082v2 Grokking in transformers trained on algorithmic tasks is characterized by a long delay between training-set fit and abrupt generalization, but the source of that delay remains poorly understood. In encoder-decoder arithmetic models, we argue that this delay reflects limited access to already learned structure rather than failure to acquire that structure in the first place. We study one-step Collatz prediction and find that the encoder organizes parity and residue structure within the first few thousand training steps, while output accuracy remains near chance for tens of thousands more. Causal interventions support the decoder bottleneck hypothesis. Transplanting a trained encoder into a fresh model accelerates grokking by 2.75 times, while transplanting a trained decoder actively hurts. Freezing a converged encoder and retraining only the decoder eliminates the plateau entirely and yields 97.6% accuracy, compared to 86.1% for joint training. What makes the decoder's job harder or easier depends on numeral representation. Across 15 bases, those whose factorization aligns with the Collatz map's arithmetic (e.g., base 24) reach 99.8% accuracy, while binary fails completely because its representations collapse and never recover. The choice of base acts as an inductive bias that controls how much local digit structure the decoder can exploit, producing large differences in learnability from the same underlying task. --- ### TinyGaze: Lightweight Gaze-Gesture Recognition on Commodity Mobile Devices Authors: Yaxiong Lei, Hyochan Cho, Fergus Buchanan, Shijing He, Xinya Gong, Yuheng Wang, Juan Ye Published: 2026-03-30 Categories: cs.HC, cs.CV arXiv: https://arxiv.org/abs/2604.09658v1 Gaze gestures can provide hands free input on mobile devices, but practical use requires (i) gestures users can learn and recall and (ii) recognition models that are efficient enough for on-device deployment. We present an end-to-end pipeline using commodity ARKit head/eye transforms and a scaffolded guidance-to-recall protocol grounded in learning theory. In a pilot feasibility study (N=4 participants; 240 trials; controlled single-session setting), we benchmark a compact time-series model (TinyHAR) against deeper baselines (DeepConvLSTM, SA-HAR) on 5-way gesture recognition and 4-way user identification. TinyHAR achieves strong performance in this pilot benchmark (Macro F1 = 0.960 for gesture recognition; Macro F1 = 0.997 for user identification) while using only 46k parameters. A modality analysis further indicates that head pose dynamics are highly informative for mobile gaze gestures, highlighting embodied head--eye coordination as a key design consideration. Although the small sample size and controlled setting limit generalizability, these results indicate a potential direction for further investigation into on-device gaze gesture recognition. --- ### Prints in the Magnetic Dust: Robust Similarity Search in Legacy Media Images Using Checksum Count Vectors Authors: Maciej Grzeszczuk, Kinga Skorupska, Grzegorz M. Wójcik Published: 2026-03-30 Categories: cs.CV, cs.HC, eess.IV arXiv: https://arxiv.org/abs/2604.09657v1 Digitizing magnetic media containing computer data is only the first step towards the preservation of early home computing era artifacts. The audio tape images must be decoded, verified, repaired if necessary, tested, and documented. If parts of this process could be effectively automated, volunteers could focus on contributing contextual and historical knowledge rather than struggling with technical tools. We therefore propose a feature representation based on Checksum Count Vectors and evaluate its applicability to detecting duplicates and variants of recordings within a large data store. The approach was tested on a collection of decoded tape images (n=4902), achieving 58\% accuracy in detecting variants and 97% accuracy in identifying alternative copies, for damaged recordings with up to 75% of records missing. These results represent an important step towards fully automated pipelines for restoration, de-duplication, and semantic integration of historical digital artifacts through sequence matching, automatic repair and knowledge discovery. --- ### Fairboard: a quantitative framework for equity assessment of healthcare models Authors: James K. Ruffle, Samia Mohinta, Chris Foulon, Mohamad Zeina, Zicheng Wang, Sebastian Brandner, Harpreet Hyare, Parashkev Nachev Published: 2026-03-30 Categories: cs.LG, cs.AI, stat.AP, stat.ME arXiv: https://arxiv.org/abs/2604.09656v1 Despite there now being more than 1,000 FDA-authorised AI medical devices, formal equity assessments -- whether model performance is uniform across patient subgroups -- are rare. Here, we evaluate the equity of 18 open-source brain tumour segmentation models across 648 glioma patients from two independent datasets (n = 11,664 model inferences) along distinct univariate, Bayesian multivariate, spatial, and representational dimensions. We find that patient identity consistently explains more performance variance than model choice, with clinical factors, including molecular diagnosis, tumour grade, and extent of resection, predicting segmentation accuracy more strongly than model architecture. A voxel-wise spatial meta-analysis identifies neuroanatomically localised biases that are compartment-specific yet often consistent across models. Within a high-dimensional latent space of lesion masks and clinic-demographic features, model performance clusters significantly, indicating that the patient feature space contains axes of algorithmic vulnerability. Although newer models tend toward greater equity, none provide a formal fairness guarantee. Lastly, we release Fairboard, an open-source, no-code dashboard that lowers barriers to equitable model monitoring in medical imaging. --- ### NeuroPath: Practically Adopting Motor Imagery Decoding through EEG Signals Authors: Jiani Cao, Kun Wang, Yang Liu, Zhenjiang Li Published: 2026-03-30 Categories: cs.HC, cs.AI, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2604.09654v1 Motor Imagery (MI) is an emerging Brain-Computer Interface (BCI) paradigm where a person imagines body movements without physical action. By decoding scalp-recorded electroencephalography (EEG) signals, BCIs establish direct communication to control external devices, offering significant potential in prosthetics, rehabilitation, and human-computer interaction. However, existing solutions remain difficult to deploy. (i) Most employ independent, opaque models for each MI task, lacking a unified architectural foundation. Consequently, these models are trained in isolation, failing to learn robust representations from diverse datasets, resulting in modest performance. (ii) They primarily adopt fixed sensor deployment, whereas real-world setups vary in electrode number and placement, causing models to fail across configurations. (iii) Performance degrades sharply under low-SNR conditions typical of consumer-grade EEG. To address these challenges, we present NeuroPath, a neural architecture for robust MI decoding. NeuroPath takes inspiration from the brain's signal pathway from cortex to scalp, utilizing a deep neural architecture with specialized modules for signal filtering, spatial representation learning, and feature classification, enabling unified decoding. To handle varying electrode configurations, we introduce a spatially aware graph adapter accommodating different electrode numbers and placements. To enhance robustness under low-SNR conditions, NeuroPath incorporates multimodal auxiliary training to refine EEG representations and stabilize performance on noisy real-world data. Evaluations on three consumer-grade and three medical-grade public datasets demonstrate that NeuroPath achieves superior performance. --- ### Diffusion-Based Generative Priors for Efficient Beam Alignment in Directional Networks Authors: Esraa Fahmy Othman, Lina Bariah, Merouane Debbah Published: 2026-03-30 Categories: eess.SP, cs.AI arXiv: https://arxiv.org/abs/2604.09653v1 Beam alignment is a key challenge in directional mmWave and THz systems, where narrow beams require accurate yet low-overhead training. Existing learning-based approaches typically predict a single beam and do not quantify uncertainty, limiting adaptive beam sweeping. We recast beam alignment as a generative task and propose a conditional diffusion model that learns a probabilistic beam prior from compact geometric and multipath features. The learned priors guide top-$k$ sweeps and capture the SNR loss induced by limited probing. Using a ray-traced DeepMIMO scenario with an 8-beam DFT codebook, our best conditional diffusion model achieves strong ranking performance (Hit@1 $\approx 0.61$, Hit@3 $\approx 0.90$, Hit@5 $\approx 0.97$) while preserving SNR at small sweep budgets. Compared with a deterministic classifier baseline, diffusion improves Hit@1 by about 180\%. Results further highlight the importance of informative conditioning and the ability of diffusion sampling to flexibly trade accuracy for computational efficiency. The proposed diffusion framework achieves substantial improvements in small-$k$ Hit rates, translating into reduced beam training overhead and enabling low-latency, energy-efficient beam alignment for mmWave and THz systems while preserving received SNR. --- ### FlowHijack: A Dynamics-Aware Backdoor Attack on Flow-Matching Vision-Language-Action Models Authors: Xinyuan An, Tao Luo, Gengyun Peng, Yaobing Wang, Kui Ren, Dongxia Wang Published: 2026-03-30 Categories: cs.CV, cs.LG, cs.RO arXiv: https://arxiv.org/abs/2604.09651v1 Vision-Language-Action (VLA) models are emerging as a cornerstone for robotics, with flow-matching policies like $π_0$ showing great promise in generating smooth, continuous actions. As these models advance, their unique action generation mechanism - the vector field dynamics - presents a critical yet unexplored security vulnerability, particularly backdoor vulnerabilities. Existing backdoor attacks designed for autoregressive discretization VLAs cannot be directly applied to this new continuous dynamics. We introduce FlowHijack, the first backdoor attack framework to systematically target the underlying vector-field dynamics of flow-matching VLAs. Our method combines a novel $τ$-conditioned injection strategy, which manipulates the initial phase of the action generation, with a dynamics mimicry regularizer. Experiments demonstrate that FlowHijack achieves high attack success rates using stealthy, context-aware triggers where prior works failed. Crucially, it preserves benign task performance and, by enforcing kinematic similarity, generates malicious actions that are behaviorally indistinguishable from normal actions. Our findings reveal a significant vulnerability in continuous embodied models, highlighting the urgent need for defenses targeting the model's internal generative dynamics. --- ### FluidFlow: a flow-matching generative model for fluid dynamics surrogates on unstructured meshes Authors: David Ramos, Lucas Lacasa, Fermín Gutiérrez, Eusebio Valero, Gonzalo Rubio Published: 2026-03-30 Categories: cs.LG, cs.AI, physics.flu-dyn arXiv: https://arxiv.org/abs/2604.08586v1 Computational fluid dynamics (CFD) provides high-fidelity simulations of fluid flows but remains computationally expensive for many-query applications. In recent years deep learning (DL) has been used to construct data-driven fluid-dynamic surrogate models. In this work we consider a different learning paradigm and embrace generative modelling as a framework for constructing scalable fluid-dynamics surrogate models. We introduce FluidFlow, a generative model based on conditional flow-matching, a recent alternative to diffusion models that learns deterministic transport maps between noise and data distributions. FluidFlow is specifically designed to operate directly on CFD data defined on both structured and unstructured meshes alike, without the needs to perform any mesh interpolation pre-processing and preserving geometric fidelity. We assess the capabilities of FluidFlow using two different core neural network architectures, a U-Net and diffusion transformer (DiT), and condition their learning on physically meaningful parameters. The methodology is validated on two benchmark problems of increasing complexity: prediction of pressure coefficients along an airfoil boundary across different operating conditions, and prediction of pressure and friction coefficients over a full three-dimensional aircraft geometry discretized on a large unstructured mesh. In both cases, FluidFlow outperform strong multilayer perceptron baselines, achieving significantly lower error metrics and improved generalisation across operating conditions. Notably, the transformer-based architecture enables scalable learning on large unstructured datasets while maintaining high predictive accuracy. These results demonstrate that flow-matching generative models provide an effective and flexible framework for surrogate modelling in fluid dynamics, with potential for realistic engineering and scientific applications. --- ### QCFuse: Query-Centric Cache Fusion for Efficient RAG Inference Authors: Jianxin Yan, Zeheng Qian, Wangze Ni, Zhitao Shen, Zhiping Wang, Haoyang Li, Jia Zhu, Lei Chen, Kui Ren Published: 2026-03-30 Categories: cs.DB, cs.AI arXiv: https://arxiv.org/abs/2604.08585v1 Cache fusion accelerates generation process of LLMs equipped with RAG through KV caching and selective token recomputation, thereby reducing computational costs and improving efficiency. However, existing methods primarily rely on local perspectives for token selection and lack global awareness from the user query. Utilizing this global awareness is challenging due to the high cost of obtaining context-aware query representations and the strict pipeline constraints required for efficient attention analysis. Thus, this demonstration introduces QCFuse, an innovative KV cache fusion system centered on the user query. QCFuse leverages semantic summary anchors to enhance query representations and selectively recomputes query-related tokens to improve accuracy, updating tokens based on the attention distribution of the most critical Transformer layer to preserve the high efficiency of the pipeline structure. Evaluations on real-world datasets demonstrate that QCFuse significantly improves the response efficiency of LLMs by 40\% while maintaining equivalent accuracy compared to current methods. Additionally, in certain scenarios, QCFuse achieves an attention denoising effect that yields higher response accuracy, demonstrating substantial potential in the optimization of LLM inference. --- ### CSAttention: Centroid-Scoring Attention for Accelerating LLM Inference Authors: Chuxu Song, Zhencan Peng, Jiuqi Wei, Chuanhui Yang Published: 2026-03-30 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2604.08584v1 Long-context LLMs increasingly rely on extended, reusable prefill prompts for agents and domain Q&A, pushing attention and KV-cache to become the dominant decode-time bottlenecks. While sparse attention reduces computation and transfer costs, it often struggles to maintain accuracy at high sparsity levels due to the inherent distribution shift between Queries and Keys. We propose Centroid-Scoring Attention (CSAttention), a training-free sparse attention method optimized for high-throughput serving of reusable contexts. CSAttention adopts a storage-for-computation strategy tailored to the offline-prefill/online-decode setting: it front-loads computation into a one-time offline prefill phase that can be amortized across multiple queries, while aggressively optimizing per-step decoding latency. Specifically, CSAttention constructs query-centric lookup tables during offline prefill, whose size remains fixed during decoding, and enables online decoding to replace full-context scans with efficient table lookups and GPU-friendly score accumulation. Extensive experiments demonstrate that CSAttention achieves near-identical accuracy to full attention. Under high sparsity (95%) and long-context settings (32K-128K), CSAttention consistently outperforms state-of-the-art sparse attention methods in both model accuracy and inference speed, achieving up to 4.6x inference speedup over the most accurate baseline at a context length of 128K. --- ### Learning to Retrieve from Agent Trajectories Authors: Yuqi Zhou, Sunhao Dai, Changle Qu, Liang Pang, Jun Xu, Ji-Rong Wen Published: 2026-03-30 Categories: cs.IR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2604.04949v1 Information retrieval (IR) systems have traditionally been designed and trained for human users, with learning-to-rank methods relying heavily on large-scale human interaction logs such as clicks and dwell time. With the rapid emergence of large language model (LLM) powered search agents, however, retrieval is increasingly consumed by agents rather than human beings, and is embedded as a core component within multi-turn reasoning and action loops. In this setting, retrieval models trained under human-centric assumptions exhibit a fundamental mismatch with the way agents issue queries and consume results. In this work, we argue that retrieval models for agentic search should be trained directly from agent interaction data. We introduce learning to retrieve from agent trajectories as a new training paradigm, where supervision is derived from multi-step agent interactions. Through a systematic analysis of search agent trajectories, we identify key behavioral signals that reveal document utility, including browsing actions, unbrowsed rejections, and post-browse reasoning traces. Guided by these insights, we propose LRAT, a simple yet effective framework that mines high-quality retrieval supervision from agent trajectories and incorporates relevance intensity through weighted optimization. Extensive experiments on both in-domain and out-of-domain deep research benchmarks demonstrate that retrievers trained with LRAT consistently improve evidence recall, end-to-end task success, and execution efficiency across diverse agent architectures and scales. Our results highlight agent trajectories as a practical and scalable supervision source, pointing to a promising direction for retrieval in the era of agentic search. --- ### From PDF to RAG-Ready: Evaluating Document Conversion Frameworks for Domain-Specific Question Answering Authors: José Guilherme Marques dos Santos, Ricardo Yang, Rui Humberto Pereira, Alexandre Sousa, Brígida Mónica Faria, Henrique Lopes Cardoso, José Duarte, José Luís Reis, Luís Paulo Reis, Pedro Pimenta, José Paulo Marques dos Santos Published: 2026-03-30 Categories: cs.IR, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2604.04948v2 Retrieval-Augmented Generation (RAG) systems depend critically on the quality of document preprocessing, yet no prior study has evaluated PDF processing frameworks by their impact on downstream question-answering accuracy. We address this gap through a systematic comparison of four open-source PDF-to-Markdown conversion frameworks, Docling, MinerU, Marker, and DeepSeek OCR, across 21 pipeline configurations, varying the conversion tool, cleaning transformations, splitting strategy, and metadata enrichment. Evaluation was performed using a 50-question benchmark over a corpus of 36 Portuguese administrative documents (1706 pages, ~492K words), with LLM-as-judge scoring over 50 independent runs per configuration. Statistical significance was assessed via Wilcoxon signed-rank tests with Cohen's d effect sizes. Two baselines bounded the results: naïve PDFLoader (86.2%) and manually curated Markdown (91.3%). Docling with hierarchical splitting and image descriptions achieved the highest automated accuracy (94.1 +/- 1.6%), surpassing even manual curation. A per-question-type analysis revealed that table-dependent questions drive the largest accuracy differences, with a 33-percentage-point gap between basic and hierarchical splitting. Metadata enrichment and hierarchy-aware chunking contributed more to accuracy than the conversion framework alone. An exploratory GraphRAG implementation underperformed basic RAG (82% vs. 94.1%). These findings demonstrate that data preparation quality is the dominant factor in RAG system performance. --- ### SUMMIR: A Hallucination-Aware Framework for Ranking Sports Insights from LLMs Authors: Nitish Kumar, Sannu Kumar, S Akash, Manish Gupta, Ankith Karat, Sriparna Saha Published: 2026-03-30 Categories: cs.IR, cs.AI arXiv: https://arxiv.org/abs/2604.04947v1 With the rapid proliferation of online sports journalism, extracting meaningful pre-game and post-game insights from articles is essential for enhancing user engagement and comprehension. In this paper, we address the task of automatically extracting such insights from articles published before and after matches. We curate a dataset of 7,900 news articles covering 800 matches across four major sports: Cricket, Soccer, Basketball, and Baseball. To ensure contextual relevance, we employ a two-step validation pipeline leveraging both open-source and proprietary large language models (LLMs). We then utilize multiple state-of-the-art LLMs (GPT-4o, Qwen2.5-72B-Instruct, Llama-3.3-70B-Instruct, and Mixtral-8x7B-Instruct-v0.1) to generate comprehensive insights. The factual accuracy of these outputs is rigorously assessed using a FactScore-based methodology, complemented by hallucination detection via the SummaC (Summary Consistency) framework with GPT-4o. Finally, we propose SUMMIR (Sentence Unified Multimetric Model for Importance Ranking), a novel architecture designed to rank insights based on user-specific interests. Our results demonstrate the effectiveness of this approach in generating high-quality, relevant insights, while also revealing significant differences in factual consistency and interestingness across LLMs. This work contributes a robust framework for automated, reliable insight generation from sports news content. The source code is availble here https://github.com/nitish-iitp/SUMMIR. --- ### Generative Chemical Language Models for Energetic Materials Discovery Authors: Andrew Salij, R. Seaton Ullberg, Megan C. Davis, Marc J. Cawkwell, Christopher J. Snyder, Cristina Garcia Cardona, Ivana Matanovic, Wilton J. M. Kort-Kamp Published: 2026-03-30 Categories: physics.chem-ph, cond-mat.mtrl-sci, cs.AI, cs.CL, cs.LG arXiv: https://arxiv.org/abs/2604.03304v1 The discovery of new energetic materials remains a pressing challenge hindered by limited availability of high-quality data. To address this, we have developed generative molecular language models that have been pretrained on extensive chemical data and then fine-tuned with curated energetic materials datasets. This transfer-learning strategy extends the chemical language model capabilities beyond the pharmacological space in which they have been predominantly developed, offering a framework applicable to other data-spare discovery problems. Furthermore, we discuss the benefits of fragment-based molecular encodings for chemical language models, in particular in constructing synthetically accessible structures. Together, these advances provide a foundation for accelerating the design of next-generation energetic materials with demanding performance requirements. --- ### Downscaling weather forecasts from Low- to High-Resolution with Diffusion Models Authors: Joffrey Dumont Le Brazidec, Simon Lang, Martin Leutbecher, Baudouin Raoult, Gert Mertes, Florian Pinault, Aristofanis Tsiringakis, Pedro Maciel, Ana Prieto Nemesio, Jan Polster, Cathal O Brien, Matthew Chantry Published: 2026-03-30 Categories: physics.ao-ph, cs.AI arXiv: https://arxiv.org/abs/2604.03303v1 We introduce a probabilistic diffusion-based method for global atmospheric downscaling implemented within the Anemoi framework. The approach transforms low-resolution ensemble forecasts into high-resolution ensembles by learning the conditional distribution of finer-scale residuals, defined as the difference between the high-resolution fields and the interpolated low-resolution inputs. The system is trained on reforecast pairs from ECMWF IFS, using coarse fields at 100 km to reconstruct fine-scale variability at 30 km resolution. The bulk of the training focuses on recovering small-scale structures, while fine-tuning in high-noise regimes enables the generation of extremes. Evaluation against the medium-range IFS ensemble target shows that the model increases probabilistic skill (FCRPS) for surface variables, reproduces target power spectra at small scales, captures physically consistent multivariate relationships such as wind-pressure coupling, and generates extreme values consistent with those of the target ensemble in tropical cyclones. --- ### Beyond Static Vision: Scene Dynamic Field Unlocks Intuitive Physics Understanding in Multi-modal Large Language Models Authors: Nanxi Li, Xiang Wang, Yuanjie Chen, Haode Zhang, Hong Li, Yong-Lu Li Published: 2026-03-30 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2604.03302v1 While Multimodal Large Language Models (MLLMs) have demonstrated impressive capabilities in image and video understanding, their ability to comprehend the physical world has become an increasingly important research focus. Despite their improvements, current MLLMs struggle significantly with high-level physics reasoning. In this work, we investigate the first step of physical reasoning, i.e., intuitive physics understanding, revealing substantial limitations in understanding the dynamics of continuum objects. To isolate and evaluate this specific capability, we introduce two fundamental benchmark tasks: Next Frame Selection (NFS) and Temporal Coherence Verification (TCV). Our experiments demonstrate that even state-of-the-art MLLMs perform poorly on these foundational tasks. To address this limitation, we propose Scene Dynamic Field (SDF), a concise approach that leverages physics simulators within a multi-task fine-tuning framework. SDF substantially improves performance, achieving up to 20.7% gains on fluid tasks while showing strong generalization to unseen physical domains. This work not only highlights a critical gap in current MLLMs but also presents a promising cost-efficient approach for developing more physically grounded MLLMs. Our code and data are available at https://github.com/andylinx/Scene-Dynamic-Field. ---