# Research Papers — 2026-09 · bundle page 204 of 266 > Full-text markdown of papers #10151–10200 of 13298 published in 2026-09, newest first. AI-readable mirrors of the full 13298-paper 2026-09 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-09/page/204 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/203.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/205.md) ### Who Are They to Each Other? Multi-Agent Reasoning for Speaker Relationship Inference Authors: Yaohan Guan, Yen-Ju Lu, Yuzhe Wang, Junhyeok Lee, Jesus Villalba, Laureano Moro Velazquez, Thomas Thebaud, Najim Dehak Published: 2026-09-09 Categories: cs.MA, cs.CL, cs.SD arXiv: https://arxiv.org/abs/2609.09628v1 Inferring speaker relationships from spoken conversations is an important step towards socially aware speech understanding. However, this task remains underexplored, and supervised modeling is costly to train and scale. At the same time, existing inference-time LLM approaches provide limited structure for handling subtle, distributed, and multimodal relational cues that may support multiple plausible interpretations. To address these limitations, we introduce a training-free multi-agent reasoning framework that organizes inference through structured interaction among LLM agents, allowing relationship judgments to be proposed, challenged, and adjudicated without task-specific training. We instantiate this framework with two complementary designs. We propose Multi-Role Multi-Agent Debate as a task-specific adaptation of standard multi-agent debate for speaker relationship inference, assigning agents complementary roles or social-theory-grounded perspectives rather than a single undifferentiated viewpoint. In contrast, we introduce Multi-Agent Compete, a competition-based protocol that compares agent judgments through pairwise adjudication, eliminates weaker candidates, and retains the most defensible one. We evaluate these methods on the Seamless Interaction dataset across different modality settings, covering both binary classification and fine-grained relationship-detail prediction. Results suggest that they improve over zero-shot and existing multi-agent baselines in most cases. Human evaluation further suggests that this task is challenging even for people. LLM methods can sometimes outperform human annotators in text-included settings but are less competitive in the audio setting. Together, these findings suggest that relationship inference benefits from structured inference-time interaction among agents, while acoustic cues are not yet fully captured by current models. --- ### Seven Sources of Physical AI Capability Formation Authors: Gang Chen Published: 2026-09-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.09627v1 Capabilities relevant to Physical AI can arise from materially different formation histories, yet existing taxonomies organized by morphology, architecture, learning algorithm, task, or domain do not directly answer what gives rise to a capability. We define a capability-formation source as a factor materially contributing to capability formation, distinct from components or construction steps. We identify seven non-exclusive sources: Recorded-Experience (RE), Predictive-Modeling (PM), Evaluative-Interaction (EI), Surrogate-Environment (SE), Mechanism-Grounded (MG), Embodied-Coupling (EC), and Evolution-Driven (ED) Formation. Using reconstructive induction with theoretical saturation, we traced a research matrix to primary studies, deduplicated the literature, set coding rules, and conducted three rounds of maximum-difference and negative-case sampling. Challenges included curriculum and self-supervised learning, active inference, open-ended and developmental learning, planning and search, neuro-symbolic architectures, digital twins, generative physical world models, and morphology-control co-design. Within the scope and criteria fixed as of September 4, 2026, all 49 evidence records were explainable by the seven sources individually or in combination. No R1-R3 challenge produced an irreducible eighth source, and R3 required no new core definition or substantive boundary rule. We therefore claim theoretical saturation within the stated scope, not logical completeness or exhaustive future coverage. The framework distinguishes similarity in observed capability from similarity in how it was formed, supporting analysis of explanation, transfer, replication, dependencies, governance evidence, and geoeconomic foundations. --- ### Hyperbolic Geometry for Open-World Object Detection in Remote Sensing Imagery Authors: Wuzhou Li, Jiawei Zhou, Shenghang Wang, Xiang Li Published: 2026-09-09 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.09626v1 Open-world object detection (OWOD) extends closed-set detection by requiring models to identify unknown objects and incrementally learn them once annotations become available. In remote sensing imagery, object categories often exhibit latent hierarchical relationships that may be inadequately represented in the Euclidean spaces commonly adopted by existing methods, limiting unknown-object recall and incremental-learning performance. To address this issue, we investigate hyperbolic geometry for OWOD in remote sensing imagery and propose HyRS-OWOD. To improve unknown object recall, we design a two-step unknown-object discovery mechanism: a Decoupled Objectness Learning (DOL) module that disentangles foreground perception from semantic information to separate foreground proposals from background regions, followed by a Hyperbolic Uncertainty Learning (HUL) component that leverages the radius of hyperbolic embeddings as an uncertainty-aware cue for known-unknown discrimination. For incremental learning, we develop a Hyperbolic Metric Learning (HML) strategy that enhances inter-class separability, facilitating the incorporation of novel categories while mitigating catastrophic forgetting. Experiments on three remote sensing benchmarks demonstrate consistent improvements in unknown recall and incremental learning over state-of-the-art OWOD methods. --- ### From State Synchronization to Cognitive Self-Evolution: An Operational Architecture for Cognitive Digital Twins Authors: Haoran Gao, An Li, Zhen Li, Jun Cai Published: 2026-09-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.09625v1 As Digital Twin (DT) systems evolve beyond state synchronization toward task-oriented and knowledge-driven operation, Cognitive Digital Twins (CDTs) have emerged as an extension that incorporates cognitive capabilities into twin operation. Existing CDT studies often focus on specific enabling techniques, such as learning modules, knowledge graphs, and large language models, while providing limited insight into how cognition can be systematically integrated into DT architectures. To address this issue, this paper proposes a four-layer CDT architecture consisting of the physical layer, digital-twin layer, cognitive layer, and task layer. The proposed architecture establishes a self-evolving closed operational loop spanning these four layers, in which physical states are synchronized into digital representations, cognition constructs task-specific cognitive models through knowledge, memory, and attention, and task-level decisions are generated under practical constraints. Operational feedback further refines cognitive experience and updates relationships and annotations in the digital representation, enabling subsequent task interpretation, initiation, and reasoning to evolve with system operation. Based on this framework, two representative operation modes are characterized: user-request-driven cognition and self-driven cognition. We further discuss key enabling mechanisms and deployment challenges associated with semantic communication, knowledge querying, task orchestration, and closed-loop synchronization. A lightweight simulation study illustrates reliable closed-loop task feasibility under limited semantic information and improved operational efficiency through accumulated task experience. The proposed framework provides a structured foundation for the design and development of future CDT systems. --- ### Marker-free eye-gaze estimation using a single image and depth from defocus Authors: David Hurtubise-Martin, Feriel Fass, Djemel Ziou, Marie-Flavie Auclair-Fortier Published: 2026-09-09 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.09610v1 This paper presents a marker-free eye-gaze estimation approach using a single 2D camera, such as an integrated laptop webcam. The gaze-related features are estimated from iris localization and head pose estimated by using depth from defocus. A variational Bayesian multinomial logistic regression framework is used as mapping from the estimated features to the position of regard, based on an 8-dimensional feature vector of head-pose and iris-displacement parameters. No external marker is needed. Experiments were conducted by estimating the gaze of people watching a computer screen at different distances and compared against five existing methods. The obtained scores demonstrate the effectiveness of the proposed approach. --- ### RouteBridge: Reliability-Routed Bidirectional Distillation Between Neural Radiance Fields and 3D Gaussian Splatting Authors: YuanHang Wang, Xin Cao Published: 2026-09-09 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.09606v1 Neural radiance fields (NeRFs) and 3D Gaussian Splatting (3DGS) encode a scene with complementary inductive biases, but existing cross-representation distillation typically fixes one representation as teacher for the entire scene. A globally fixed teacher can propagate local reconstruction errors. We present RouteBridge, a bidirectional framework that selects the teaching direction for each ray. Its reliability estimator combines photometric residuals with representation-specific geometric evidence and routes supervision from NeRF to 3DGS, from 3DGS to NeRF, or abstains. A renderer-independent interface transfers color, opacity, and normalized depth without shared features or point correspondence. On mip-NeRF 360, the NeRF and 3DGS exports reach 28.56 and 28.77 dB, respectively. The 3DGS export improves over 3DGS by 1.56 dB and over NeRF-GS by 0.45 dB while reducing LPIPS to 0.207. On static three-view DTU, RouteBridge obtains 21.12 dB. Ablations show that both adaptive routing and geometric ray targets contribute to the improvement. --- ### Watermarks Without Verification: AI Text Watermarking After the EU AI Act Authors: Alexander Nemecek, Vipin Chaudhary, Erman Ayday Published: 2026-09-09 Categories: cs.CY, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2609.09604v1 On August 2, 2026, the obligations of Article 50 of the EU AI Act took effect, requiring generative AI providers to mark the content their systems produce and ensure it can be detected as AI-generated. Days later, Anthropic disclosed that every Claude model released after that date embeds a watermark based on SynthID-Text in all generated text, enabled by default with no user opt-out; Google has deployed SynthID-Text in Gemini since 2024. Users objected that the watermark degrades quality, particularly for code, that it secretly encodes identifying information, and, in mutual contradiction, that it is easily removable and inescapable; the vendor answered with assurances of unchanged quality, no identifying information, and robustness to light editing. In this work, we argue that neither the objections nor the assurances can currently be verified and that this unverifiability, rather than watermarking itself, is the substantive governance failure. We sort the contested assertions by what it would take to settle each and evaluate the open-source SynthID-Text implementation on two open-weight models, because no public tool can test the deployed systems. On prose, the measured effect of the watermark does not exceed that of changing the sampling seed. On code, the cost is three points of correctness on one model and below measurement on the other, while detection remains near chance, a limitation of detectability rather than quality. The remaining gaps trace to withheld access or missing institutions and we map each to a requirement: release of matched outputs, configuration disclosure, accredited audits, a shared evaluation protocol, and interoperable detection. --- ### Compact Visuotactile World Models for Lifting: Prediction, Reward Alignment, and Force Constraints Authors: Qinzhen Ma Published: 2026-09-09 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.09597v2 Accurate tactile forecasts need not improve force-constrained control. We study a 652,157-parameter action-conditioned visuotactile world model with matched behavior cloning, policy learning in imagination, independent reactive implicit Q-learning, and model-assisted force feedback. A fixed protocol executes 34 policies on 120 fresh MuJoCo environments spanning geometry and physical-parameter shifts, plus 324 independently replayed action branches on 12 additional ID environments. Visuotactile dynamics reduce force action-effect MAE from 0.413 N for persistence to 0.338 N. Model-assisted feedback raises ID force-budgeted success from 73.3% to 93.3%, with paired difference +20.0 (+6.7,+33.4) percentage points (95% CI), with the difference occurring during scripted lowering. Its pooled difference is +3.9 (-4.5,+11.7) points. Imagined RL achieves 11.9% pooled joint success versus 25.0% for reactive IQL. An empirical tactile-residual stress test adds 330 executions. The evidence concerns rigid-box lifting after a common approach, without physical-robot transfer or a closed-loop safety guarantee. --- ### Teacher Geometry Shapes Learnability in Teacher-Student Networks Authors: Kai J. Sandbrink, Flavio Martinelli, Alexander van Meegen, Wulfram Gerstner, Johanni Brea Published: 2026-09-09 Categories: cs.LG, cs.AI, cs.NE arXiv: https://arxiv.org/abs/2609.09595v1 Teacher-student systems, in which a teacher neural network generates training labels so that a student neural network can learn to implement the same function, are widely used as an abstract setting to study learning. However, the structure of the teachers is often overlooked by assuming randomly-generated, normally-distributed parameters. This hides substantial variation in how learnable different teachers are. We formalize learnability as the success rate of converging to the global minimum, as a function of overparameterization, learning algorithm, student initialization distribution, and teacher geometry. We both identify an easy distribution that maximizes node dissimilarity and a hard distribution that minimizes it, and show that these two distributions induce markedly different success rates across a large range of settings and for different activation functions. To explain the gap, we study the loss landscape of small neural networks that contain two distinct kinds of suboptimal local minima, out-of-bounds (OOB) minima at the edge of the data distribution and interior minima within. Assuming infinite data and a fast readout layer, we analytically reduce the loss landscape of small networks to two dimensions, showing that the region of attraction of interior minima changes as a function of teacher structure. In larger networks, maximally dissimilar teachers induce more interior minima, while minimally dissimilar teachers induce more OOB minima. Motivated by these analyses, we show that differentially increasing the learning rate of the readout layer and decreasing the learning rate of the inner biases increases success rates. These findings provide an important step in narrowing the gap between the study of teacher-student networks and more structured functions that arise in practice. --- ### Modality-Decoupled Federated Learning for Privacy-Preserving Embodied Intelligence in 6G Authors: Zhuodong Liu, Xiangyu Li, Chunhong Yuan, Hongyang Du, Bodong Shang, Qingqing Wu, Tony Q. S. Quek, Mohsen Guizani Published: 2026-09-09 Categories: eess.SP, cs.AI arXiv: https://arxiv.org/abs/2609.09591v1 Sixth-generation (6G) wireless networks are expected to provide a key infrastructure for large-scale embodied intelligence, where heterogeneous robots collaborate through low-latency connectivity, edge intelligence, and distributed sensing. Vision-language-action (VLA) models offer a foundation by integrating visual perception, language understanding, and action generation into a unified closed-loop policy. However, training and adapting VLA models to distributed robotic agents introduce challenges in privacy protection, communication efficiency, and model heterogeneity. Existing federated learning (FL) methods overlook the intrinsic differences among vision, language, and action pathways in parameter scale, privacy exposure, update dynamics, and tolerance to compression or perturbation. To address this issue, this article proposes FedMVLA, a modality-decoupled FL framework for privacy-preserving embodied intelligence in 6G networks. FedMVLA incorporates three mechanisms: modality-aware federated aggregation (MAFA), modality-aware privacy allocation (MAPA), and modality-aware communication compression (MACO), together with a modality-sliced transport design that routes the precision-critical action stream through a protected ultra-reliable low-latency slice. A case study on federated robotic manipulation over the Third Generation Partnership Project (3GPP)-based wireless substrate, covering fading, co-channel interference, and malicious jamming, shows that FedMVLA achieves an 84.8% task success rate, exceeds FedAvg by 22.2 percentage points, sustains a widening margin when scaling to 128 clients across eight cells, and reduces the schedule-averaged per-client uplink model-update payload by 95.6% (approximately 96%), while keeping the 95th percentile (p95) of the round-critical uplink completion time near 1.5s. --- ### A Function-Space Approach to the Statistical Mechanics of Learning Dynamics Authors: Yizhou Zhang, Weichen Wu, Lun Du, Zhengjie Miao Published: 2026-09-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.09589v1 Deep neural networks exhibit regular macroscopic behavior despite highly nonlinear dynamics in vast parameter spaces. We develop a statistical-mechanical description of learning directly in function space, treating parameter configurations as microscopic realizations and functions with their dynamical operators as macroscopic variables. For mean-squared loss, the exact error dynamics are governed by the learning operator \(M=JJ^\ast\). Combining the dynamical Boltzmann weight of the conditional stochastic dynamics with the parameter-space density of states, whose local curvature defines a statistical operator \(B\), and integrating over local fluctuations yields $$ Φ_{\mathrm{fluc}}(M;B)=\frac{σ_ξ^2}{2}\log\det(M^{-1}+B)+\mathrm{const}. $$ At fixed spectrum, this term is rotationally stationary when \((M,B)=0\), is minimized by pairing large eigenvalues of \(M\) with small eigenvalues of \(B\), and generates a local restoring contribution against rotational mismatch. For ReLU-type function spaces under mild stable statistical conditions, \(B=σ_ξ^2L^\ast\mathcal K L\), where \(L\) measures coarse-grained second-order structure. Thus the low-\(B\) sector corresponds, up to bounded anisotropy of \(\mathcal K\), to low structural curvature, implying a preference for faster relaxation along smooth, data-adaptive directions. These results identify function space as a natural macroscopic level for studying stable collective organization in learning. --- ### Distillation of Synthetic Data for Time Series Foundation Models Authors: Niloy Biswas, Noureddine El Karoui Published: 2026-09-09 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.09586v2 Time series foundation models (TSFMs) are increasingly pre-trained on synthetically generated time series trajectories, where the data generating process is known. Current pre-training recipes are based on loss objectives which compare TSFM outputs to realized future values of each trajectory. We instead propose loss objectives which compare TSFM outputs to the conditional forecast distribution of each trajectory, a procedure we call synthetic data distillation (SDD). SDD corresponds to a Rao-Blackwellization of the training objective, in that it leaves the expectation of stochastic gradients unchanged while provably reducing the covariance of the stochastic gradient under the Loewner partial ordering. We empirically validate SDD on a TSFM model family of sizes from $4$M to $2.5$B parameters, and observe faster convergence of validation loss at every model size: on Gaussian Process data, SDD attains or improves upon the Status Quo loss whilst requiring $10\%-40\%$ less training iterations. --- ### CityPlanner: A Sandbox Agent for Executable Urban Planning Authors: Wentao Zhang, Jingyuan Wang, Zetong Zhou, Yifan Yang, Wenrui Wang Published: 2026-09-09 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.09578v1 Urban planning is a real-world spatial optimization problem that requires selecting feasible actions from large candidate spaces under practical objectives such as cost and service quality. Existing optimization and reinforcement learning methods are effective for fixed formulations, but often depend on task-specific representations and constraint handling. We propose \emph{CityPlanner}, a sandbox-agent framework for executable urban planning. CityPlanner introduces \emph{UrbanSandbox}, a unified file-based environment where agents inspect task files, generate plans, run evaluators, and revise decisions based on executable feedback. To make learning tractable, we further propose atomic-task reinforcement learning, which decomposes long sandbox trajectories into \emph{BuildPlan} for initial construction and \emph{ImprovePlan} for feedback-based refinement. Experiments on a real-world benchmark show that CityPlanner consistently outperforms heuristic, task-specific RL, and general LLM-agent baselines. Ablations verify the contributions of UrbanSandbox, atomic-task RL, and iterative deployment. We release the code and dataset at https://anonymous.4open.science/r/co-agent-C1C8 --- ### Myocardial Strain Drift Correction in Deep Learning Based Ultrasound Tracking Authors: Thierry Judge, Nicolas Duchateau, Andreas Østvik, Havard Dalen, Bjørnar Grenne, Pierre-Yves Courand, Lasse Lovstakken, Pierre-Marc Jodoin, Olivier Bernard Published: 2026-09-09 Categories: eess.IV, cs.AI, cs.CV arXiv: https://arxiv.org/abs/2609.09577v1 Myocardial strain from echocardiography is a key biomarker for cardiac function. Recent deep learning methods show strong performance for myocardial motion tracking but often lack physiological constraints, leading to temporal drift across the cardiac cycle. Consequently, tracked points may not return to their relative initial positions at the end of each cardiac cycle, producing inaccurate strain estimates and even divergence in some cases. We propose a deep learning framework that compensates for drift during myocardial tracking. We extend a state-of-the-art echocardiographic tracking method (TAS-Net) with persistent memory tokens that share information across sliding windows over full cardiac cycles. A teacher-student fine-tuning strategy on real echocardiographic data then enforces physiologically consistent cyclic motion while preserving tracking accuracy. Experiments show reduced global and regional strain drift, improved agreement with clinical references, and better test-retest reproducibility, supporting more reliable myocardial strain estimation in clinical practice. --- ### Beyond Top Words: MonoTM for Topic Modeling with Interpretable Monosemantic Features Authors: Una Joh, Bei Yu Published: 2026-09-09 Categories: cs.CL, cs.HC arXiv: https://arxiv.org/abs/2609.09575v1 Topic models summarize large text corpora, but top-ranked words often provide only a limited representation of topic semantics. Sparse autoencoders (SAEs) offer a way to move beyond word-level descriptors by extracting interpretable features from dense representations, yet how feature interpretability relates to topic-inference quality remains unclear. We introduce \textbf{MonoTM}, an interpretable topic modeling framework that decouples these roles. Across three benchmark corpora, we show that document--topic mixture estimation and semantic interpretation favor different SAE configurations and feature subsets. MonoTM estimates mixtures from the full SAE bag-of-features representation and, with them fixed, learns topic descriptors over a separate vocabulary of corpus-grounded semantic features. This design preserves global topic structure while representing topics with semantic units more meaningful than individual words, making them more useful for downstream corpus analysis. --- ### Geometric organization of olfactory descriptor data in the Poincaré disk Authors: Aniss Aiman Medbouhi, Farzaneh Taleb, Giovanni Luca Marchetti, Danica Kragic Published: 2026-09-09 Categories: cs.CE, cs.LG, stat.AP arXiv: https://arxiv.org/abs/2609.09573v1 Odor quality is commonly represented using high dimensional descriptor profiles, yet their low dimensional organization remains unclear. We investigated whether a two-dimensional hyperbolic embedding can provide an interpretable representation of this structure. We applied hyperbolic metric multidimensional scaling to two complementary datasets: 480 Sagar rating profiles from three participants rating 160 odorants on 15 continuous descriptors, and 4983 GoodScents--Leffingwell molecules annotated with 138 binary descriptors. The embeddings substantially preserved pairwise descriptor distances, supporting subsequent analyses of radial and angular organization. In Sagar, rating profile entropy was strongly and negatively associated with hyperbolic radius, with diffuse profiles closer to the center and concentrated profiles closer to the boundary. This radial organization emerged primarily at the level of the full descriptor profile, rather than any individual descriptor, and remained robust across alternative descriptor representations, participant specific analyses, and averaged ratings. Sweet, musky, fruity, pleasantness showed the strongest directional trends. In GoodScents--Leffingwell, active label entropy, reflecting descriptor multiplicity, increased with radius, whereas orthogonalized descriptor entropy, reflecting spread across orthogonal modes, decreased with radius. Related binary descriptors occupied coherent localized high-density regions. These findings reveal complementary radial and angular organization in the hyperbolic representation of olfactory descriptor data. They support hyperbolic mapping as an interpretable descriptive framework in which radius summarizes global profile properties, while the angular component captures continuous descriptor gradients and categorical organization. --- ### Learning with Synthetic Data via SGD in High-Dimensional Linear Regression Authors: Jichu li, Difan Zou Published: 2026-09-09 Categories: stat.ML, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.09572v1 Synthetic data has become a promising way to scale model training beyond limited human-generated data but it may also induce strong model collapse (Dohmatob et al., 2024), where any fixed fraction of synthetic data prevents model performance from improving under data scaling, leaving a non-vanishing excess risk floor. In this paper, we study how synthetic data affects the generalization of one-pass SGD in high-dimensional linear regression with model shift. We establish finite-sample risk bounds for mixed and two-stage training, separating standard bias and variance from source-mismatch effects, namely fluctuation and persistent drift under mixing and filtered initialization bias under two-stage. These bounds reveal a sharp contrast: mixed training induces strong model collapse, while two-stage training avoids the floor by using synthetic data only in the first stage, showing that collapse is not inevitable under a simple data curriculum. Under a random sketch model, we further obtain scaling laws for both protocols, with tight results for mixed training in the optimization-saturated regime. These laws show that larger models may amplify synthetic-induced degradation under mixing, and quantify how high-quality synthetic pretraining may reduce bias in two-stage training. Finally, we establish an exact finite-sample necessary-and-sufficient condition for two-stage training to strictly outperform real-only training under the same real-data budget and identical real-stage updates. Overall, our results highlight that synthetic data is neither inherently harmful nor beneficial; its effect depends critically on both its quality and the training protocol used to incorporate it. --- ### Reproducing Omitted Temporal Expressions in Japanese News for Retrieval-Augmented Applications Authors: Tomoaki Yasuda, Shotaro Ishihara Published: 2026-09-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.09569v1 News articles often contain omitted temporal expressions, such as day-only or month-only mentions, which must be interpreted with reference to the publication date. When such articles are indexed or processed as standalone text in search and retrieval-augmented generation (RAG) systems, these omissions can cause temporal mismatches and unstable interpretation by large language models. We focus on reproducing omitted temporal expressions as concrete dates or intervals using the publication date as external context before the articles are indexed for search and RAG applications. Specifically, building on established temporal-expression extraction and normalization techniques and informed by a manual analysis of Japanese news articles, we propose jaROTE, a rule-based pipeline for Japanese news. Experiments on two news corpora demonstrate that jaROTE achieves high performance, and remains competitive with LLMs while providing a fast, low-cost pipeline. We further show that temporal reproduction improves time-constrained lexical retrieval, demonstrating the practical value of publication-date-grounded normalization for Japanese news retrieval. --- ### Positional task conditioning for scalable defect detection across product families in large product catalogs Authors: Soham Satyadharma, Gabriel Roccabruna, Suleiman A. Khan Published: 2026-09-09 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.09567v2 Product families in large product catalogs suffer from inconsistencies such as duplicates and unit mismatches that degrade customer experience. Detecting these requires reasoning over multiple error types across lengthy product listings, where LLM classification quality degrades due to long-context limitations. We address this by decomposing detection into focused sub-tasks that reduce context and isolate error types, improving F1 from 52% to 87%. For scalable deployment, we introduce Positional Task Conditioning (PTC), which distills this capability into a single smaller model by reinforcing task identity at structural prompt boundaries. PTC outperforms rationale-based distillation across five models and two architecture families, achieving within 1.79% F1 of the frontier at upto 98% lower cost. Our system is deployed across multiple countries processing 10+ million product families. --- ### Multi-Agent Agentic Graph Learning via Structural Signatures Authors: Liang Qu, Jianxin Li, Hua Wang Published: 2026-09-09 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.09565v1 Agentic graph learning (AGL) has recently achieved promising results on graph reasoning tasks, where an agent powered by a large language model (LLM) sequentially samples the graph as evidence to support its final prediction. Existing methods either employ a single agent or orchestrate multiple role-based agents to reason and learn over the entire graph, but both essentially rely on a shared reasoning policy across different graph regions, which can be suboptimal for graphs with heterogeneous structural and semantic patterns. Inspired by the progress of multi-agent collaboration on complex reasoning tasks, a natural remedy is to let multiple agents own different memory and collaborate; however, applying this paradigm to graphs directly faces two challenges. First, existing AGL methods typically verbalize graph structures into natural-language descriptions for LLM agents, making the reasoning process sensitive to the ordering of structural information and thereby breaking the permutation-invariant nature of graphs. Second, incorporating increasingly large sampled neighborhoods leads to rapidly growing contexts. To address these challenges, this paper introduces a multi-agent agentic graph learning (i.e., MAAGL) framework. MAAGL partitions the graph into communities and assigns an independent agent to each community for region-specific specialization. MAAGL represents structural and semantic evidence separately. Structural evidence is summarized by a dynamically updated structural signature that is permutation-invariant and fixed in size, while semantic evidence is filtered to the top-k nodes ranked by relevance. Based on historical trajectories with similar signatures, agents estimate their confidence and trigger debate-style collaboration when needed. Extensive experiments on four benchmark datasets show that MAAGL outperforms SOTA AGL methods. --- ### Robust Industrial Cyber Physical Classification Using Neuromorphic Temporal Embeddings and Hybrid SNN XGBoost Under Machine Unlearning Attacks Authors: Ammar Kamoona, Sajad Koushkbaghi, Mahdi Jalili, Peter McTaggart, Xinghuo Yu Published: 2026-09-09 Categories: cs.LG, cs.CR, cs.NE arXiv: https://arxiv.org/abs/2609.09564v1 The digitalisation of electrical distribution networks has increased the exposure of power-grid infrastructure to cyber attacks. Existing intrusion detection systems (IDSs), however, often rely on computationally expensive deep learning models that are difficult to deploy at the edge. Periodic retraining also exposes these systems to machine unlearning attacks, where selective data removal can degrade detection performance. We propose a hybrid Spiking Neural Network (SNN) and XGBoost architecture that combines efficient temporal encoding with a lightweight classifier and provides structural resilience to such attacks. The SNN is trained once on clean data and used as a fixed feature extractor, while only the XGBoost classifier is retrained during model updates. Evaluated on two real-world public power-system datasets, the proposed method achieves 99.9\% accuracy (F1-macro 0.999) on the Synchrophasor dataset and 95.0\% accuracy (F1-macro 0.943) on the MSU/ORNL dataset, outperforming standalone baselines. Under selective label-flipping attacks, the hybrid model loses only 0.9\% F1-macro at 10\% poisoning and delays target-class collapse from 60\% to 70\% poisoning compared with raw models. These results demonstrate that neuromorphic temporal encoding can provide both accurate cyber-attack detection and improved resilience to data poisoning in cyber-physical systems. --- ### Towards Automatic Evolution Tree Generation from Citation Graphs Authors: Zexing Zhao, Yuntong Hu, Liang Zhao Published: 2026-09-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.09561v1 Surveys remain the primary way researchers grasp the lineage of methods within an AI subfield, but they scale poorly against the current rate of publication. Existing taxonomy-induction methods are largely leaf-bound and time-agnostic; they tend to force transitional papers into mature leaves and can create topological inversions between ancestors and descendants. We propose EvoTree, a staged framework that decouples conceptual backbone learning from temporal refinement: a graph-aware encoder with distribution-based hierarchical clustering yields a stable taxonomy backbone; temporal fine-tuning then re-attaches marginal papers to internal nodes under monotonic-path constraints; a final LLM pass labels concepts without altering the topology. We release the first annotated benchmark for this task across 11 AI subfields. EvoTree attains the highest NMI and citation-direction accuracy among all baselines and the best concept purity on the annotated benchmark, and is the only method with non-trivial marginal-paper detection on the annotated set. --- ### The Vibe Shift in Software Engineering: Evaluating AI-Led Conversational Programming for Performance, Cognition, and Responsible Adoption Authors: Sales G. Aribe, Louie Jay S. Labastida Published: 2026-09-09 Categories: cs.SE, cs.AI, cs.ET arXiv: https://arxiv.org/abs/2609.09560v1 This study evaluates Vibe Coding, an emerging AI-led conversational programming paradigm that enables developers to generate software through natural-language interaction with large language models. Using a mixed-methods design, the study assessed performance efficiency, cognitive implications, and responsible adoption in comparison with traditional and AI-assisted coding environments. Thirty participants, including professional developers and advanced computing students, completed equivalent programming tasks under three experimental conditions. Quantitative data were analyzed using descriptive statistics and repeated-measures ANOVA, while qualitative data were examined through thematic analysis. Results show that vibe coding significantly improved development efficiency, reducing task completion time by 27% compared with traditional coding and 12% compared with AI-assisted coding. However, these gains were accompanied by lower maintainability indices and higher security vulnerabilities, indicating trade-offs in software quality. Usability results yielded a good rating (SUS = 71.4), while cognitive workload remained moderate (NASA-TLX = 55.5), reflecting reduced syntactic effort but increased linguistic reasoning. Thematic analysis identified trust calibration, loss of control, cognitive adaptation, and prompt-engineering strategy as key constructs. Notably, perceived loss of control was associated with increased security risks due to reduced transparency and validation of AI-generated outputs. Based on these findings, the study proposes a three-pillar framework for responsible adoption: hybrid integration of human and AI capabilities, human oversight and transparent accountability, and context-aware deployment. Overall, vibe coding enhances productivity but requires critical oversight, reinforcing its role as a transformative yet transitional paradigm in software development. --- ### High-probability guarantees for linear accessibility in feature superposition Authors: Enrico Vompa Published: 2026-09-09 Categories: stat.ML, cs.AI, cs.IR, cs.LG, math.PR arXiv: https://arxiv.org/abs/2609.09556v3 Neural networks can leverage feature superposition to encode more concepts than dimensions, but cross-feature interference constrains the linear accessibility of simultaneously active features. By framing linear accessibility as a compressed sensing problem, we derive high-probability bounds for fixed supports under subgaussian noise, proving the sufficient dimension scales linearly ($d=O_{\varepsilon}(k \log m)$) rather than prior worst-case quadratic limits. We characterize the asymmetry between active and inactive interference and the trade-off between interference and observation-noise budgets. We then validate these bounds across system parameters through Gaussian-tail approximations. We also introduce IHT-SAE, which uses learned iterative refinement to improve feature recovery beyond the limits of linear availability. These results quantify the geometric constraints of the linear representation hypothesis, providing a framework for evaluating sparse autoencoders, compositional generalization, and neural interpretability. --- ### BuzzASR: A Swarm of 100+ Monolingual Speech Recognition Models Authors: Shivam Singh, Aditya Yadavalli, Catherine Arnett, Alex Warstadt Published: 2026-09-09 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.09554v1 We introduce BuzzASR, a collection of language-specialized fine-tuned Whisper models adapted for automatic speech recognition (ASR) in 102 languages. Large end-to-end Transformer-based ASR models such as Whisper have revolutionized ASR, but most prominent models are highly multilingual. As a result, these models often perform poorly on languages less well-represented in their training set. While it has long been known that effective language adaptation can be achieved through simple fine-tuning on monolingual data, this strategy has only been applied to a small number of languages. We massively scale up this simple approach to 102 languages covered in the FLEURS dataset, while also implementing a more complex language adaptation strategy that integrates monolingual tokenizer replacement and data augmentation using text-only fine-tuning. BuzzASR models outperform Whisper-large-v3 on 77 out of 102 languages, reducing character error rates (CER) by a factor of over 2.8 on average. Our models achieve state-of-the-art CER among open-source systems on 27 of 102 languages on the combined FLEURS and Common Voice test set. Our tokenizer replacement strategy yields an average 3.3x improvement in compression rate (characters per token) over Whisper's multilingual BPE, with gains of up to 21.7x. We release all models, code, and detailed results: https://lemn-lab.github.io/buzz-asr --- ### Arbitrary Cipher Attacks Against Large Language Models Do Not Require Fine-Tuning Authors: Thomas Rivasseau Published: 2026-09-09 Categories: cs.CR, cs.AI arXiv: https://arxiv.org/abs/2609.09553v1 Large language model safety and security research is preoccupied with, among other things, detecting and preventing jailbreak attacks: alignment bypasses that allow an adversarial user to elicit unwanted or harmful outputs from models. Arbitrary cipher, or covert communication, attacks are one such type of jailbreak and have previously been demonstrated against the fine-tuning APIs of commercial models. In these attacks, target models are trained on a corpus of encrypted harmful questions and responses and subsequently respond to harmful requests through the learned encryption scheme. In this paper, we show that newer frontier models do not require fine-tuning to acquire cipher-based communication skills. Instead, they can learn these skills through prompting and, when necessary, through in-context learning. Furthermore, model alignment is significantly weakened or entirely bypassed when communication occurs through the learned cipher. To the best of our knowledge, this constitutes a novel attack vector against commercial black-box large language models. We demonstrate successful jailbreaks against frontier models developed by Anthropic, Google, and OpenAI. Our attack bypasses commercial harmfulness classifiers because harmful content is encrypted and therefore appears as nonsensical text or gibberish. --- ### TEFM: Token-Efficient Faithful Modeling for Structured Data Authors: Zhichao Hou, Lingdao Sha, Xueyu Mao, Yang Liu, Peijie Qiu, Rui Song Published: 2026-09-09 Categories: cs.CL, cs.LG arXiv: https://arxiv.org/abs/2609.09552v1 In this paper, we solve two fundamental obstacles in applying LLMs to critical domains: token efficiency and faithfulness. To address both constraints jointly, we present TEFM (Token-Efficient Faithful Modeling), a framework designed for structured data analysis in critical domains. TEFM achieves token efficiency by compressing lengthy structured observations into compact Behavioral Code tokens, dramatically reducing token consumption with minimal information loss. Moreover, TEFM enables faithful rationalization through a dual-fidelity objective that jointly optimizes code-level reconstruction and prediction-level fidelity, identifying minimal sufficient feature subsets grounded in input data. Comprehensive experiments across various domain datasets and model backbones (Qwen3, Gemma-2, Phi-4) show that TEFM achieves competitive classification accuracy with dramatic token reduction (approximately 1\% token retention in clinical and 2\% in security domains) while producing faithful rationales. --- ### An Efficient and Effective Agentic Group Shilling Attack on Recommender Systems Authors: Quoc Viet Nguyen, Trinh Pham, Viet Huynh, Hongzhi Yin, Quoc Viet Hung Nguyen, Bay Vo, Thanh Tam Nguyen Published: 2026-09-09 Categories: cs.CR, cs.CL arXiv: https://arxiv.org/abs/2609.09551v1 Recommender systems have become core infrastructure for modern online platforms, personalizing content at scale and strongly influencing what users see, click on, and purchase. However, this dependence on user interaction also exposes them to shilling attacks, where malicious actors can inject fake profiles to distort item rankings and control visibility. Existing attacks often rely on target-specific fine-tuning or fixed profile templates, making them either difficult to adapt to different victims or easier to detect. To overcome these limitations, we propose the Agentic Group Attack System (AGAS), a coordinated shilling framework where a central Coordinator directs a group of role-switching worker agents to adaptively promote a target item across different victim families. The Coordinator dynamically adjusts the strategy when progress stalls or suppression signals increase, while workers pursue a shared objective and switch between active and inactive roles to avoid repetitive patterns. Under the same attack budgets and evaluation protocols, AGAS consistently surpasses strong baselines in target promotion while better preserving benign recommendation quality, weakening representative detectors, and achieving higher efficiency than prior attacks. These findings also emphasize that defending recommender systems may require mechanisms that can handle adaptive shilling campaigns, not just isolated fake-profile injections. Our code is available at https://github.com/phkhanhtrinh23/AGAS. --- ### A Statistical Approach to Estimating Sample Size of Machine Learning Models Authors: Dat Phan-Trong, Sunil Gupta, Svetha Venkatesh Published: 2026-09-09 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.09547v1 Sample size determination for machine learning (ML) prediction models is challenging because conventional power analysis typically requires the predictor-outcome relationship and effect structure to be specified a priori. Nonlinear ML models learn complex prediction surfaces that do not admit straightforward analytical power calculations. We propose a framework that approximates nonlinear ML models with localized linear representations and estimates sample size requirements by evaluating statistical power across these local regions. --- ### A Mechanistic Study of AI-Text Detection Neurons in Frozen BERT: Sparse Probing and Activation Patching on RAID Authors: Paweł Blicharz, Miłosz Grunwald Published: 2026-09-08 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.30287v1 AI-generated text detectors achieve high accuracy on standard benchmarks, yet the internal representations that drive these predictions remain poorly understood. We study which neurons in a frozen BERT-base-uncased encoder support AI-text detection, using the RAID benchmark across six generators spanning pure-base and instruction-tuned models. We apply the L1-to-L2 sparse-probing protocol of Gurnee et al. (2023) to all 9,216 CLS hidden-state dimensions (12 layers x 768), which we call neurons. The procedure recovers a stable set of under 1% of neurons per generator, consistent across folds and seeds; a probe restricted to that set retains most of the full-feature detection accuracy. Bidirectional activation patching confirms this set's causal relevance: in both directions it flips predictions an order of magnitude more often than size-matched random sets. Mean-ablating the same neurons leaves accuracy largely intact; the signal is therefore redundantly distributed. Cross-generator analysis reveals a bipartite structure: instruction-tuned generators concentrate 30-36% of stable neurons in BERT's final layer while both base generators fall below 14%, consistent with a layer-12 footprint of post-training alignment. Leave-one-family-out evaluation shows the selected neurons retain 86-94% of the full-feature ceiling on unseen generator families, so a detector can operate on a small fixed subspace without re-identifying neurons per generator. --- ### When Does Advection-Aware Graph Nowcasting Help? A Controlled Study of Distributed Solar Ramp Forecasting with a Self-Supervised Cloud-Motion Estimator Authors: Phillip Jiang Published: 2026-09-08 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.30286v1 Short-term forecasting of cloud-induced power ramps across a network of distributed photovoltaic (PV) or irradiance sensors is a recognised pain point for grid operators. A natural idea is to make the graph neural network (GNN) advection-aware: connect each site to the sites upwind of it, with edge time-lags set by the cloud-motion vector (CMV), so that a ramp is propagated forward before it physically arrives. Using a controlled synthetic testbed with a known wind field, we show that (i) with a realistic cross-correlation CMV estimate, an explicit advection graph does not beat a plain static or learned-adjacency spatiotemporal GNN; (ii) roughly half of the benefit available from a perfect CMV comes simply from providing an accurate motion vector as an input feature, not from graph structure; and (iii) advection helps only when the advective displacement over the forecast horizon, v*H, fits inside the sensor network. Motivated by (ii), we introduce a small self-supervised cloud-motion estimator -- a position-aware encoder trained only on a multi-lag optical-flow reconstruction objective with an annealed kernel -- that recovers the true wind vector to 2-4 degrees median angular error, 2-4x better than the classical cross-correlation method across every wind regime. Freezing this estimator and feeding its vector to the forecaster closes about 60% of the oracle-CMV RMSE gap at moderate wind (8-15% RMSE reduction over no advection), with no external wind data. We also report a negative result for a spatially-coherent probabilistic head. All claims are established on a single synthetic simulator; we discuss why real-network validation is the necessary next step and outline it. --- ### ChainDoRA: Tensor-Train Factorized Weight-Decomposed Low-Rank Adaptation for Parameter-Efficient LLM Fine-Tuning Authors: Ashfak Yeafi, Mehedi Hasan, Md Khairul Islam Published: 2026-09-08 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.25058v1 Parameter-efficient fine-tuning (PEFT) adapts large language models (LLMs) to downstream tasks while updating only a small fraction of their pretrained parameters. Low-Rank Adaptation (LoRA) uses two trainable low-rank matrices, while Weight-Decomposed Low-Rank Adaptation (DoRA) further separates weight magnitude and direction but retains the dense LoRA-style factorization in its directional branch. We propose ChainDoRA, a weight-decomposed adaptation framework that constructs the directional low-rank factors from a connected Tensor-Train (TT) chain, where the adapter rank forms the boundary rank between input- and output-side TT contractions and an independent TT rank controls representation capacity and parameter cost. Under a controlled 15,119-example response-only adaptation setting with LLaMA-7B, ChainDoRA is evaluated against matched LoRA and DoRA baselines on seven commonsense reasoning benchmarks. ChainDoRA with TT rank 16 achieves a seven-task average accuracy of 72.30%, compared with 69.88% for LoRA and 69.39% for DoRA, while requiring only 5.35M trainable parameters versus 56.10M for LoRA and 56.98M for DoRA, corresponding to a 90.62% reduction relative to DoRA. Ablations over TT rank and adapter placement show controllable parameter-accuracy trade-offs, indicating that connected TT parameterization can substantially reduce the parameter cost of magnitude-direction adaptation while preserving, and in this setting improving, downstream reasoning performance. --- ### Physics-guided deep metric learning with continuous time embeddings for open-world radar pulse de-interleaving Authors: Vikas Agnihotri, Jasleen Kaur Published: 2026-09-08 Categories: eess.SP, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.25057v1 Radar pulse de-interleaving is a foundational Electronic Support Measures (ESM) task that aims to separate chronologically interleaved pulse streams from multiple non-cooperative transmitters under unknown emitter cardinality in dense, contested electromagnetic environments. Classical histogram transforms and closed-world deep classifiers degrade under severe pulse loss, agile Pulse Repeti tion Interval (PRI) modulation, and spurious clutter. In this paper, we systematically characterise continuous temporal representations and physics-guided model selection in deep metric learning for open-world radar de-interleaving. Building on the transformer-based metric-learning framework for open-world deinterleaving introduced by Gunn et al. (1), we introduce a continuous Time-of-Arrival (ToA) sinusoidal positional encoding that directly models physical inter-pulse durations rather than ordinal token indices, a design choice that contrasts with (1), who found ordinal positional encodings provided no benefit and omitted them entirely. Neural network parameters are optimised solely via Supervised Contrastive (SupCon) learning, while scale-aware physical domain priors based on PRI Consistency and Angle-of-Arrival (AoA) continuity serve as physics-guided validation and checkpoint-selection criteria operating on unsupervised HDBSCAN cluster assignments. --- ### Graph-Based Inference for Feedback-Driven Word Deduction: A Scalable Framework for the Jotto Problem Authors: Dakshi Arora, Prakhar Kumar Srivastava, Ranjib Banerjee Published: 2026-09-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.25056v1 A feedback-based word deduction framework based on the Jotto problem is proposed, and the problem space is represented as a weighted graph where all valid words correspond to nodes, and the edge weight is defined by the number of common letters between the two words. Finally, the gameplay is defined as an iterative constraint propagation mechanism where feedback is used to iteratively narrow the incompatible space of the graph, facilitating the reduction of the hypothesis space in a structured and interpretable manner. In contrast to existing approaches, where the problem space is typically defined for fixed-length isograms, the proposed framework generalizes to variable-length words (between 3 and 8 letters) and naturally extends to repeated letter cases, facilitating the treatment of realistic Jotto problem instances within a unified framework for the first time. The proposed framework's applicability and solver dynamics are also discussed through an interactive implementation and a qualitative case study, respectively. Significant automated tests on approximately 3,000 simulated gameplay scenarios identify a novel convergence behavior: the expected number of iterations diminishes with increasing word length. A strong relationship is confirmed using statistical tests to verify a logarithmic relationship, which is also verified using regression modeling and goodness-of-fit tests. In addition to the initial problem statement, this formulation introduces graph pruning as a viable paradigm for feedback-driven inference with interpretability and its association with symbolic reasoning and interactive intelligent systems. --- ### ICDAR2026 Competition on Multimodal Reasoning over Documents in Multiple Domains Authors: Artemis Llabrés, Marc Serra Ortega, Tomàs Ockier, Samuel Ortega Cuadra, Amritpal Singh, Christos Georgakilas, Andrey Barsky, Ernest Valveny, Dimosthenis Karatzas Published: 2026-09-08 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.25055v1 In this report we present results of the ICDAR2026 Competition on Multimodal Reasoning over Documents in Multiple Domains. This competition aimed to advance research in document understanding through the task of Visual Question Answering (VQA). Building upon previous DocVQA benchmarks, this competition introduces challenging reasoning questions over a diverse collection of documents spanning eight domains, including business reports, scientific papers, slides, posters, maps, comics, infographics, and engineering drawings. The competition concluded with 20 valid submissions from 8 teams spanning zero-shot VLMs, OCR and parser-augmented pipelines, agentic retrieval systems, multi-agent ensembles, and fine-tuned multimodal models. The results show that the strongest systems move beyond single-pass prompting and instead rely on structured evidence extraction, retrieval, verification, and orchestration across multiple components. --- ### Did You Steal My Shot? Pioneering Camera Motion Plagiarism Detection in Generative Videos Authors: Chengguo Zhang, Ping Ping Published: 2026-09-08 Categories: cs.CV, cs.MM arXiv: https://arxiv.org/abs/2609.22267v1 Camera motion often reflects directorial intent and requires professional equipment, making it a high value form of intellectual property. However, generative video models can imitate such high value camera motions with simple prompts, while existing similarity detection methods mainly operate on visual content and fail to capture deeper motion similarity. This is mainly because their training data entangles camera motion with visual content. Moreover, traditional optical flow is insufficient to represent complex camera motions. We therefore build the first benchmark for camera motion analysis, including a motion dataset with \textbf{11} motion styles and evaluation protocols. Furthermore, we propose a motion representation that augments optical flow with vorticity cues from fluid dynamics, thereby better capturing motions. Experiments show that our detector achieves a \textbf{3.02*} improvement in plagiarism detection over the strongest baseline and remains effective on generative videos. We believe our work extends copyright protection beyond static content to dynamic camera motion. --- ### Grouped Value Attention: Efficient KV Caching via On-Demand Key Reconstruction Authors: Vishesh Tripathi, Abhay Kumar, Ramsha Khan Published: 2026-09-08 Categories: cs.AR, cs.LG arXiv: https://arxiv.org/abs/2609.13285v2 The KV cache is a primary bottleneck for Transformer decoding: its memory footprint and cache-read traffic grow with sequence length. Grouped-query attention (GQA) reduces this cost by sharing key-value heads, but still stores both a key and a value at every step. We introduce Grouped Value Attention (GVA), which stores grouped values and reconstructs content keys with a learned linear map. At inference, the map can be absorbed into the query, eliminating the need to materialize content keys in the intended decode path. A small shared decoupled RoPE channel retains positional information through a separately cached positional key. For the configurations studied, this representation reduces persistent cache scalars by approximately 45-47% relative to matched GQA. At the 350M-parameter scale with 30B FineWeb-Edu tokens, the 16-dimensional positional variant reaches 44.18 average accuracy across five tasks, compared with 44.36 for GQA and 43.88 for MLA. These results demonstrate near-GQA benchmark accuracy with a more compact cache representation. To translate this compact representation into faster autoregressive inference, we have developed custom decoding kernels and are currently evaluating their end-to-end inference performance with an open-source release planned soon. --- ### Multimodal-Multiresolution Foundation Model for Lunar Remote Sensing Authors: Paolo Fraccaro, Gabby Nyirjesy, Daniela Szwarcman, Himanshu Patil, Vishal Gaur, Rohit Lal, Rachel A. Slank, Geoffrey Dawson, Hiyam Debary, Michael K. Barker, Andrew Annex, Vishnu Viswanathan et al. Published: 2026-09-08 Categories: cs.CV, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.13283v1 We present a multimodal foundation model for lunar remote sensing, pretrained from scratch on SomBench, a geographically partitioned corpus of nearly two million co-registered tile bundles spanning 11 modalities at two spatial scales (1 m/pixel and 100 m/pixel). The model adapts the TerraMind masked-token architecture with two lunar-specific extensions: acquisition geometry is provided as explicit context, and meter- and hundred-meter-scale tiles are trained jointly so that a single set of weights covers both resolutions. FlexiViT patch embeddings allow adaptation to different patch sizes without retraining, while modality-wise inputs enable flexible multimodal fine-tuning. Qualitative generation experiments suggest the model learns meaningful cross-modal correspondences, including terrain derivatives from elevation and illumination-consistent reflectance from geometry. We evaluate on four benchmarks: crater detection at WAC and NAC scales, irregular mare patch (IMP) segmentation, and polar ice prospectivity regression. Across tasks, the pretrained model matches or outperforms ImageNet-pretrained baselines and an architecturally identical random-init control. On multimodal ice prospectivity regression, pretrained variants achieve the best results, while the random-init model outperforms most baselines, suggesting gains arise from both the architecture and pretraining. Label efficiency is notable for WAC crater detection, where the pretrained model trained on 50% of the data exceeds the strongest ImageNet baseline trained on the full dataset. Among adaptation strategies, LoRA matches or surpasses full fine-tuning on crater detection and IMP segmentation while using far fewer trainable parameters, whereas full fine-tuning performs best for ice prospectivity regression. We release the pretrained checkpoint, benchmark datasets, and fine-tuning code to support reproducible lunar AI research. --- ### GaugeDefect: Detecting Surface Anomalies by Curvature of Feature Transport Authors: Yefan Wang Published: 2026-09-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.13282v1 Industrial anomaly localization has advanced rapidly with feature-based, reconstruction-based, and distillation-based methods. Most of these methods score a region by asking how unusual its local appearance or feature representation is with respect to normal training images. This is a strong and practical formulation. In this work, we study a complementary geometric cue for cases where an abnormal region may still contain locally plausible visual features. Thin scratches, small dents, and disrupted repeated patterns often do not make every local patch individually abnormal; instead, they disturb how nearby features vary and connect across the surface. We propose GaugeDefect, a geometric method for surface anomaly localization based on the curvature of feature transport. Given a feature lattice, we estimate a local feature frame at each node and compute orthogonal transports between neighboring frames. The accumulated transport around a small closed loop gives a holonomy matrix, whose deviation from identity measures feature-transport curvature. After calibration on normal training images, unusually large curvature indicates a local inconsistency in the feature field. The curvature here is not the physical curvature of the inspected object, but a representation-space measure of neighborhood inconsistency. This makes the method applicable to curved surfaces, textured materials, and non-planar industrial objects. Its main role is to improve localization of subtle surface disruptions, while often producing sharper responses near defect boundaries as a natural consequence of the curvature signal. --- ### Large-scale bioacoustic detection using semantic segmentation: a deep learning framework applied to fin whale calls in ocean-bottom seismometer recordings Authors: Jocelyn Japnanto, Alex A. Saoulis, Miriam Romagosa, Rita Leitão, Gabrielle Arrieta, Mónica A. Silva, Matthew Graham, Ana M. G. Ferreira Published: 2026-09-08 Categories: physics.geo-ph, cs.LG arXiv: https://arxiv.org/abs/2609.13281v2 Ocean-bottom seismometers (OBS), originally deployed for geophysical research, continuously record low-frequency sound for months to years across broad areas of ocean, offering a largely untapped resource for passive acoustic monitoring (PAM) of baleen whales. Realising this potential requires automated detection methods that operate reliably across the varied conditions in large sensor networks. We present a deep learning semantic segmentation framework that detects the 20-Hz notes of fin whales (Balaenoptera physalus) in OBS spectrograms, assigning each pixel a probability of belonging to a call and converting the resulting probability maps into time-frequency bounding boxes describing individual detections. We trained the model on hydrophone data from one OBS deployment in the Azores-Madeira-Canaries region and applied it without retraining to vertical-component seismometer data from a second, geographically distinct deployment, showing that a single trained model generalises across sensor types and recording environments. Applied to 378,912 h of recordings from 46 OBS sites, the detector identified 6.3 million calls, forming the largest fin whale call catalogue assembled to date, with high precision (~97%) across both deployments. The resulting catalogue resolves call timing and spectral structure accurately enough to support ecological analyses, revealing coherent seasonal shifts in three persistent inter-note interval (INI) groups across the singing season and basin-scale patterns in calling activity. By transforming existing geophysical infrastructure into a scalable sensing network, our approach substantially expands the spatial and temporal reach of PAM without new hardware investment, offering a transferable framework for tracking other low-frequency vocalising species and informing conservation planning, marine spatial management, and abundance estimation across large scales. --- ### RxScribe Bench: A Multi-Axis Benchmark for Evaluating Vision-Language Models on Indian Outpatient Prescriptions Authors: Somil, Vijay Saini, Vidit Verma, Riya, Aastha Batta, Vibhuti Malhotra, Chayan Khetan, Piyush Mittal, Puneet Poonia Published: 2026-09-08 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.13280v1 Prescription transcription errors are not interchangeable. A model that fabricates a drug and a model that misreads a legible dose pose very different clinical risks, yet prescription-transcription accuracy is typically reported as a single blended figure that treats the two as equivalent. We introduce RxScribe Bench, a benchmark for evaluating vision-language models on handwritten prescription digitization that decomposes performance into four axes tied to clinical severity, rather than folding everything into a single aggregated score. Given only a prescription image and an output schema, a model produces a structured record, which is then compared field by field against a human-authored ground truth of identical shape, with each field also labeled for visibility and legibility. The four axes isolate distinct failure modes, namely Correctness, Hallucination, Engagement, and Robustness. The Robustness axis withholds its hard-field results rather than reporting one when the supporting sample falls below a minimum-evidence threshold. We evaluate frontier vision-language models on real prescriptions across independent cold runs per image, and find that no single model wins across all four axes. --- ### The MODA General Attribute Suite: A Four-Track Evaluation Benchmark for Fashion Attribute Extraction Authors: Arkid Mitra Published: 2026-09-08 Categories: cs.CV, eess.IV arXiv: https://arxiv.org/abs/2609.13279v2 Fashion attribute extraction is evaluated inconsistently: results are reported as single aggregate numbers across image types that pose different problems, fields that are not visible in an image are scored as ordinary negatives, and the effect of vocabulary mismatch between datasets is acknowledged but not measured. We release the MODA General Attribute Suite, a four-track benchmark that keeps these problems separate by construction. Each track (localized garment crops, catalogue product images, applicability-aware full-body photographs, and product text) carries its own frozen test set, input contract, metric, and leakage unit, and the tracks are never averaged. The protocol requires label-blind prediction, SHA-256 commitment of prediction files before any label is opened, a fail-closed scorer, and a 10,000-sample paired cluster bootstrap at each track's natural leakage unit; promotion requires a positive interval on every track rather than a favourable mean. We release the scorers, the split builders, our own prediction files with their hashes for every image-track system in the main results and for runs we lose, and the MODA_NER(V) model checkpoints for three of the four tracks. The text model is not distributed; its benchmark is. We report baseline results, six interventions that did not improve them, and the limitations of the evidence. On the garment-crop track we also measure how far the headline depends on the benchmark's own design: the order of the two strongest systems depends on whether fields are micro- or macro-averaged, two thirds of the cells the protocol judges carry no annotation, and disabling the applicability decision cuts attribute micro-F1 from 0.63 to 0.27, so most of that score rests on absence decisions the source labels cannot adjudicate. This version also corrects one figure in an external comparison that an inference preprocessing defect had affected. --- ### SomBench: Benchmark Dataset for Advancing Machine Learning in Lunar Science Authors: Himanshu Patil, Gabby Nyirjesy, Rachel A. Slank, Vishal Gaur, Daniela Szwarcman, Paolo Fraccaro, Nikolaos Dionelis, Michael K. Barker, Andrew Annex, Vishnu Viswanathan, Zachary Morse, Ethan I. Schaefer et al. Published: 2026-09-08 Categories: cs.CV, cs.LG arXiv: https://arxiv.org/abs/2609.13277v1 Lunar orbital missions, such as Lunar Reconnaissance Orbiter, Kaguya/SELENE, Gravity Recovery and Interior Laboratory, and Lunar Prospector, among others, provide rich multi-instrument observations, but their heterogeneity in sampling, projection, and conventions limits reproducible machine learning (ML). We introduce SomBench, a unified, spatially-aligned, ML-ready lunar dataset aggregating 30+ co-registered layers from ten instruments across four missions, spanning 1 meter to 20 kilometer/pixel and covering 82 degree latitude in 90 Lunar Transverse Mercator zones with two polar stereographic caps. An image-anchored tiling pipeline yields pretraining-ready multimodal tile views with leakage-safe splits, distributed as netCDF with Parquet catalogs. An application benchmark suite spans impact processes, volcanic history, and polar volatiles. Baseline experiments with ResNet-50 and SwinV2-B models confirm that each benchmark task is learnable from the released inputs, establishing reference points for future model development. --- ### Assessment of Non-Institutional AI Tool Usage Among Clinicians Authors: Sarah Pungitore, Jarrod Mosier Published: 2026-09-08 Categories: physics.soc-ph, cs.AI arXiv: https://arxiv.org/abs/2609.11990v1 Generative artificial intelligence (AI) tools are increasingly accessible and have the potential to improve efficiency across clinical workflows. However, clinicians may also use non-institutional AI tools that are not provided, managed, or governed by their healthcare institutions, creating potential concerns related to privacy, security, accuracy, and clinician-AI interaction. Little is known about how clinicians currently use these tools for work-related tasks. We conducted a descriptive survey of clinicians recruited from the University of Arizona College of Medicine-Tucson and Banner University Medical Center-Tucson between May 20 and June 26, 2026. Participants reported their use of AI tool categories and the frequency with which they used AI for specific tasks across five workload categories: administrative work, clinical work, research, studying/continued education, and teaching. Forty-four respondents completed the survey. Forty-three respondents reported using AI for at least one work-related task during the preceding 6 months. Conversational AI and clinical decision support/diagnostic AI were the most used tool categories, each reported by 28 respondents. Administrative and clinical tasks demonstrated the most frequent use. AI was also used for higher-risk activities, including diagnostic assistance and clinical decision support. Non-institutional AI use was common among surveyed clinicians and extended across a broad range of work-related activities, including tasks with potential implications for clinical reasoning and patient care. Further research is needed to characterize how clinicians use these tools, how they evaluate AI-generated outputs, and how AI can be safely and effectively integrated into clinical workflows. --- ### Harness or Model? Isolating the Harness Effect in Agentic Coding with a Contamination-Controlled Private Suite Authors: Mohsen Arjmandi Published: 2026-09-08 Categories: cs.AI, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2609.11987v1 An agentic coding system couples a language model to a harness: the tools, prompts and control flow that turn a chat model into an autonomous software engineer. Vendors ship harnesses tuned to their own models, and practitioners assume the vendor-native pairing solves more tasks. We measure that assumption with paired same-model contrasts on a private, contamination-controlled suite of 256 repository and post-cutoff contest tasks. The same 80 tasks ran under claude-agent-sdk and under deepagents on claude-opus-4-8, and under the openai-codex SDK and deepagents on gpt-5.5, with gemini-3.5-flash and deepseek-v3.2 as side cells. 792 of 800 planned runs were graded by an isolated oracle. Neither contrast resolves an average advantage for either harness: -1.25 pp for Opus 4.8 (48.8% vs 50.0%, task-bootstrap 95% CI (-10.0, +7.5)) and +1.25 pp for GPT-5.5 (55.6% vs 54.4%, CI (-4.4, +6.9)). The Opus average combines opposite strata: the native harness trails by 9.0 pp on the 61 repository tasks and leads by 23.7 pp on the 19 contest tasks (label-permutation p = 0.003). The partition was chosen after seeing the data and needs a designed replication. Correctness and completion also separate: 22 of 81 runs cancelled at the wall-clock ceiling had produced a passing patch. Re-priced from raw per-turn usage at frozen list prices, the neutral harness cost 1.3 to 1.6 times as much per solved task on Opus 4.8 and 1.2 times on GPT-5.5. These are observed-usage estimates. On the Anthropic account 58 runs left no usage record, and allocating that spend to either cell would move the Opus ratio between 0.7 and 2.3, so the billed ordering is unresolved. This revision corrects an August 2026 manuscript whose cost figures rested on a usage-semantics defect in our own telemetry (Section 5.1). We release the orchestrator, grading oracle, reanalysis code and derived aggregates. The tasks stay private. --- ### Unlabeled Echoes: Pseudo-Labels and Genus-Aware Smoothing for Bat Call Recognition Authors: Frank Fundel, Alexandra Howard Published: 2026-09-08 Categories: cs.SD, cs.CV, q-bio.QM arXiv: https://arxiv.org/abs/2609.11986v1 Passive acoustic monitoring produces far more bat recordings than experts can label. We show that simple model-generated pseudo-labels turn this surplus into effective supervision. We compare pseudo-labeling with other semi-supervised learning methods on an 18-species European corpus using only 10% of its training labels, then transfer the strongest approaches to South African field audio containing nine bat taxa and a nuisance class. Pseudo-labeling outperforms the other semi-supervised learning methods on every European measure, recovering up to 61.5% of the gap to full supervision. It transfers to field audio with gains of 10.69 points in species accuracy and 4.96 points in species macro-F1. We also introduce genus-aware smoothing, which directs uncertain target mass toward congeneric species. Combined with uniform smoothing, it reaches 79.16 species macro-F1, 4.73 points above hard targets. Simple pseudo-labels are therefore highly effective at this ecological data scale, while genus-aware targets inject useful biological structure at no annotation cost. https://code4conservation.github.io/UnlabeledEchoes/ --- ### Understanding In-Context Multimodal Jailbreaks via Posterior Reweighting Authors: Xu Zhang, Dev Mistry, Xiang Xu, Ren Wang Published: 2026-09-08 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2609.10613v1 In-context learning (ICL) jailbreaks reveal a critical vulnerability in multimodal large language models (MLLMs): harmful demonstrations in the prompt can induce unsafe outputs without modifying model parameters. Despite extensive empirical evidence, existing work lacks a principled understanding of why such jailbreaks reliably succeed or how their effectiveness scales with context composition. We propose a posterior reweighting framework that models a safety-aligned MLLM as implicitly operating over competing behavioral modes, and interprets in-context demonstrations as inference-time evidence that dynamically shifts the model's posterior preference between safe and harmful behaviors. This view formalizes jailbreak as a process of evidence accumulation, yielding predictive scaling laws with respect to demonstration count, harmful ratio, adversarial strength, and semantic diversity. Guided by this framework, we introduce a posterior-aware inference-time defense that adaptively injects benign counter-evidence based on estimated risk, effectively suppressing harmful posterior drift while preserving model utility. Compared to existing in-context defenses, our method achieves a significantly improved robustness-utility trade-off under a fixed intervention budget. Together, our results establish posterior reweighting as a unifying and predictive framework for understanding and mitigating ICL jailbreak in MLLMs. --- ### PEARL: A Task-Aware Framework for Evaluating Differentially Private Synthetic Educational Data Authors: Xianghui Meng, Yujing Zhang, Jionghao Lin Published: 2026-09-08 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2609.10612v1 Personalized learning systems rely on real learner data, including performance, behavior, and demographic information, but these data are highly privacy-sensitive. Differentially private (DP) synthetic data can support system development and educational research while reducing exposure of individual learners. Existing evaluations, however, assess privacy and predictive usefulness separately, without determining whether synthetic learner data remain usable for the intended personalized learning task. We introduce PEARL (Privacy-Equivalence Audit and Release Ledger), which approves a DP synthetic educational dataset only when it passes all required checks of validity, privacy protection, predictive usefulness, and suitability for the intended educational task, while recording why each rejected dataset fails. Across 96 study settings, each defined by a dataset, data-generation method, privacy budget, and random seed, only 12 produced synthetic datasets that passed all applicable PEARL checks. Many privacy-protected datasets were rejected for omitting important outcome groups, such as withdrawn students, or for failing to preserve the order of learning activities. Fairness analysis further showed that some datasets passing privacy and predictive-usefulness checks still yielded unequal at-risk prediction performance across groups defined by disability and socioeconomic background. Moreover, Deep Knowledge Tracing and Self-Attentive Knowledge Tracing learned no meaningful next-response patterns from any tested synthetic knowledge-tracing dataset, showing that privacy protection alone does not guarantee usefulness for dropout prediction, knowledge tracing, or adaptive tutoring. --- ### Black-Box Membership Inference via Word-Level Probability Estimation Authors: Shengjie Niu, Yeheng Ge, Jian Huang Published: 2026-09-08 Categories: cs.CR, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.10611v1 Membership inference attacks (MIAs) have emerged as critical tools for auditing privacy risks in large language models (LLMs), aiming to determine whether a given text was included in a model's training corpus. However, most existing MIAs require access to per-token logits or probabilities, making them inapplicable in practice to proprietary LLMs that expose only textual continuations. To address this underexplored setting, we propose Word-level Probability MIA (WPMIA), a statistically principled MIA for strict black-box privacy auditing. WPMIA estimates word-level generation probabilities via Monte Carlo sampling with local kernel smoothing, then aggregates these estimates into a sequence-level likelihood estimator. Furthermore, WPMIA constructs the likelihood conditioned on different prefixes, thereby amplifying the distributional differences between members and non-members. We evaluate WPMIA across various open-source LLMs and find that it consistently outperforms existing black-box baselines. Importantly, we also evaluate WPMIA on modern proprietary LLMs, including GPT-5-Chat, Gemini-2.5-Flash, and Claude-4.5-Haiku, achieving an average TPR@5\%FPR of 42.0 across these models. These results offer a sound foundation for future research on strict black-box membership inference. Code is available at \href{https://github.com/niusj03/WPMIA}{https://github.com/niusj03/WPMIA}. --- ### On the Relation between Code Quality and Machine Learning Performance: A Large-scale Empirical Study Authors: Marius Mignard, Steven Costiou, Anne Etien Published: 2026-09-08 Categories: cs.SE, cs.LG arXiv: https://arxiv.org/abs/2609.10610v1 Context: Computational notebooks are the standard environment for machine learning (ML) development. Within the ML community, model performance is often the primary considered metric, and code quality is treated as a secondary concern. This prioritization relies on a largely untested assumption that code quality and ML performance are unrelated. Practitioners also reuse existing code that may come from notebooks selected through social signals (popularity, author expertise) whose reliability as quality proxies has never been assessed. Objective: We empirically investigated the relationship between code quality and ML performance in notebooks, and evaluated whether popularity and author expertise give indication on code quality or performance. Method: We conducted a large-scale empirical study of 265,363 Python notebooks submitted to Kaggle competitions. We assessed code quality with two static analysis tools: Pylint, capturing general Python code quality, and SonarQube, configured with a profile of 34 rules targeting data-science and ML-specific practices. Results: The relationship between code quality and performance depends on the notion of quality considered. General Python code quality is decoupled from ML performance, showing negligible or non-significant correlations across all observations. In contrast, ML-specific violations exhibit a consistent, small negative association with performance that persists across all observations. The popularity of a notebook does not give information on the code quality or performance. Code expertise provides no information on quality or performance, but competition expertise correlates with better performance, fewer ML-specific violations, and slightly more Python errors and refactoring violations. ---