# Research Papers — 2026-09 · bundle page 230 of 266 > Full-text markdown of papers #11451–11500 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/230 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/229.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/231.md) ### Emotion as a Distribution: Joint Valence-Arousal Probability Learning for Speaker-Independent Multimodal Emotion Recognition Authors: Tingyi Lin, Wen-Ren Yang, Kuanwei Chen Published: 2026-09-04 Categories: cs.SD, cs.CL arXiv: https://arxiv.org/abs/2609.05755v1 Human emotion is graded and frequently mixed, yet most multimodal recognizers collapse it onto a single hard label. We argue the recognizer should instead expose a distribution over affective space. Our text+speech system, alongside its categorical decision, emits a $9\times9$ probability matrix over the Valence-Arousal plane, trained with a two-dimensional Gaussian soft target under a Kullback-Leibler/cross-entropy objective, aimed at counseling support. Evaluation is strict: speaker-independent 5-fold leave-one-session-out IEMOCAP with rotating-session inner validation, headline metrics only on the held-out session. Within one fixed encoder-fusion-head pipeline we compare Transformer and state-space (Mamba-1/2/3) backbones at matched depth and width, at two operating points ($T\approx550$, $T\approx2750$). The featured dual-head system reaches 73.0% $\pm$ 0.3 unweighted accuracy over three seeds (separate rerun: 72.1%), exceeding the Transformer fusion baseline by 3.0 UA points (95% session-bootstrap CI (1.0,4.7); significant under paired t-test and session-level bootstrap), with no latency or memory advantage at these lengths; swapping the ~1M trainable front-end for frozen WavLM-Large features (learnable layer weights) lifts the same architecture to 76.6% $\pm$ 1.3. Pre-specified controls scope the claims honestly: simpler valence-arousal auxiliaries reproduce the classification lift within noise, and a dedicated regression head tracks the continuous ratings slightly better, so the head's specific value is the normalized affect distribution itself. That distribution recovers the circumplex: its center of mass tracks valence and arousal (CCC 0.66/0.66; predominantly between-class structure, weaker within-class tracking), and its entropy is weakly but consistently linked to categorical rater ambiguity, not dimensional spread. --- ### A complete characterization of sequential testability and change detectability in i.i.d. models Authors: Aaditya Ramdas Published: 2026-09-04 Categories: math.ST, cs.IT, eess.SP, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2609.05752v1 We give a necessary and sufficient condition for the existence of power-one sequential tests in an i.i.d. composite testing problem. A level-\(α\) test with power one against every alternative exists if and only if the alternatives are separated from the null by a countable family of finite-block events. We provide other equivalent conditions using randomized fixed-sample tests, bounded finite-block scores, e-processes, reduced-filtration test supermartingales, and a countable cover whose finite-block weak-$*$ closed convex hulls are positively separated in total variation. As a bonus, the constructive proof yields tests have pointwise expected sample size \(O_Q(\log(1/α))\). Exactly the same conditions also characterize i.i.d.\ change detectability under optional-horizon average-run-length control: for every \(η>0\), they are equivalent to an alarm family \((T_γ)_{γ\ge1}\) satisfying \(\Prob_{P^\infty}(T_γ\leσ)\le \E_{P^\infty}σ/γ\) for every null law and every stopping time \(σ\). In fact, when these conditions hold, we can construct a single e-detector such that every null-law average run length lies between \(γ\) and \((1+η)γ+1\), and having robust Lorden delay \(O_Q(\logγ)\). --- ### The Normalization of Deviance in AI Development Authors: Emilio Barkett, Alexander Kimpton, Daniel Graham, Yusuf Kundgol Published: 2026-09-04 Categories: cs.AI, cs.ET, eess.SY arXiv: https://arxiv.org/abs/2609.05749v2 Work on the risks of artificial intelligence has focused predominantly on capability risk: the danger that systems become too powerful, too autonomous, or too misaligned with human values. Far less attention has been paid to the organizational level---to whether the institutions building these systems are themselves predisposed to drift toward failure. This paper argues that they are. Regardless of how capable AI systems become, the organizations building them face the same structural dynamics that preceded past major technological disasters. Drawing on case studies of the Space Shuttle Challenger, the Three Mile Island accident, and the Boeing 737 MAX crashes, this paper identifies the common structural mechanisms preceding each failure and maps them onto contemporary AI development. The findings suggest that existing safety infrastructure may provide less protection than it appears, as organizations can complete safety processes in full compliance and still produce catastrophic outcomes. The pre-disaster period of AI development is still underway; the purpose of this paper is to make these dynamics legible while they can still be interrupted. --- ### A Multi-Source Ensemble Approach to Candidate Generation for Alternative Vacation Rental Property Recommendations Authors: Syed Mohammed Arshad Zaidi, Eric Rincon, Shayan Hassantabar Published: 2026-09-04 Categories: cs.LG, cs.IR arXiv: https://arxiv.org/abs/2609.05748v1 Alternative property recommendations play a critical role in vacation rental marketplaces, helping users discover relevant options when viewing a specific listing. However, generating high-quality candidate alternatives presents unique challenges: heterogeneous inventory, geographic constraints, rapid availability changes, and long-tail property distributions. We present a comprehensive study of candidate generation (CG) approaches for vacation rental alternatives, comparing collaborative filtering, shallow embeddings, and graph neural network (GNN) methods. Our experiments on a large-scale vacation rental platform (over 2M active properties) show that a hybrid architecture combining item-based collaborative filtering with GNN-based retrieval improves Recall@300 by 14.8% over the strongest baseline, by leveraging the complementary strengths of the two sources: collaborative filtering excels at early recall for properties with rich interaction history, while GNNs discover diverse, non-obvious alternatives and handle cold-start scenarios more effectively. As a component result, GNN-based embeddings alone substantially outperform shallow Hotel2Vec embeddings (48-68% relative recall improvement across K), motivating their inclusion in the ensemble. Crucially, we examine how CG-stage gains carry through to the downstream ranking stage, and find that a stronger candidate pool yields higher downstream ranking quality, though attributing this effect cleanly is complicated by the coupling between candidate generation and ranker training. This recall-conversion gap is an important consideration for practitioners deploying new retrieval methods in two-stage recommendation systems. --- ### Gaussian Linear Functional Manifold Method for Massive Point Cloud Data Authors: Hong Zhao, Tonglin Zhang, Baijian Yang, Jin Wei-Kocsis, Songlin Fei Published: 2026-09-04 Categories: cs.CV, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.05744v1 Reconstructing continuous terrain manifolds from massive, unstructured airborne LiDAR point clouds remains challenging in complex Wildland-Urban Interface (WUI) environments, where deep neural networks require costly point-wise annotations and nonparametric surface reconstruction methods often lack structural interpretability. This paper introduces the Gaussian Linear Functional Manifold (GLFM), a physics-informed statistical framework that represents continuous surface topography using deterministic linear functional bases while modeling microscale diffuse laser backscatter as an isotropic Gaussian process. To avoid the quadratic computational cost of exact constrained maximum likelihood estimation, we develop an algebraic singular value decomposition (SVD) rank-reduction algorithm that enables linear-time parameter estimation and closed-form quadric classification. Evaluated on 35.2 km^2 of real-world aerial LiDAR data, GLFM automatically filters ground points and extracts morphological features, achieving an adjusted Rand index (ARI) of 0.9933 against field-verified ground truth and outperforming four leading baselines while maintaining an out-of-core memory footprint. The framework provides a rigorous, interpretable, and scalable foundation for large-scale point cloud analytics. --- ### Some Tokens Behave like Magnets: Revealing Linguistic Organization in the Layers of Language Models Authors: Andrew Liu, Devan Srinivasan, Gerald Penn Published: 2026-09-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.05743v1 We identify a special group of token vectors inside large language models (LLMs), which we term magnetic vectors, that organize the surrounding tokens by either attracting or repelling them. Particularly, tokens pointing the same way as an attracting magnet are elongated; tokens pointing the same way as a repelling magnet are compressed. Just as physical magnets pull or push away the iron filings around them, these vectors organize their surroundings through two opposing polarities. Moreover, we identify a statistically significant pattern in linguistic category where function words consistently act as repelling magnets in early layers, and we also find magnets consistently reorganize their polarities in unique ways deeper in the model. In a further case study we find this observation may unveil a deliberate, layer-wise organization in how LLMs process language. This pattern is consistent across different LLM architectures, sizes, and layer configurations. It is also causally relevant. When the LLM is fine-tuned for a downstream task, the task-functional tokens emerge as magnets. E.g., in question answering, the answer-span tokens become uniquely repelling magnets in the final layer, geometrically carving the answer out of the surrounding context. Furthermore, removing early-layer repelling magnets devastates syntactic tasks (POS tagging accuracy drops from 91% to below 10%) while sparing semantic ones, and removing late-layer attracting magnets does the reverse. We believe this phenomenon warrants further investigation, as it opens the first probe-free path to understanding how language models geometrically organize linguistic computation across their layers. --- ### SeRV: Semantic-Aligned Residual Vector Quantization for American Sign Language Generation Authors: Hongyu Wu, Xu Wu, Tianhao Wu, Jiawei Yu, Phuc Nguyen, Jian Liu, Yi Wu Published: 2026-09-04 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.05742v1 American Sign Language (ASL) generation remains challenging due to limited paired text-ASL motion data and the difficulty of learning motion representations both precise for reconstruction and predictable from linguistic input. Existing methods rely on motion tokenizers optimized for reconstruction, without explicit semantic supervision from paired text. As a result, the learned tokens remain limited in supporting semantically consistent and fine-grained ASL motion generation. To address this limitation, we propose SeRV (Semantic-Aligned Residual Vector Quantization), a semantic-aligned RVQ tokenizer for ASL generation. SeRV learns a semantically structured residual token space by combining sentence-level motion-text alignment with token-level text-conditioned supervision. Building on this tokenizer, a Hierarchical GPT predicts residual motion tokens in a coarse-to-fine manner, generating structurally coherent and semantically aligned 3D ASL motion. We further construct a large-scale reconstructed 3D ASL motion-text benchmark by recovering paired 3D motion from YouTube-ASL videos. Experiments across 375 hours of ASL video show that SeRV achieves state-of-the-art pose accuracy on both How2Sign and YouTube-ASL datasets, while producing semantically consistent 3D ASL motion directly from text. --- ### RenderFormer-V2: Neural Rendering with Heterogeneous Scene Primitives Authors: Chong Zeng, Yue Dong, Pieter Peers, Lvmin Zhang, Maneesh Agrawala Published: 2026-09-04 Categories: cs.CV, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2609.05738v1 We present 'RenderFormer-V2', a unified learned transformer-based neural rendering model, complementary to modern physics-based rendering systems, that can handle diverse light-transport effects such as caustics, volumetric scattering, environment lighting, textured and displaced surfaces and out-of-distribution materials without per-scene training or specialized code. RenderFormer-V2 models global light transport as a sequence-to-sequence transformation. Following its predecessor, RenderFormer-V2 also employs a two stage process: a view-independent stage that resolves intra-scene primitive to primitive transport, and a view-dependent stage that transforms the internal neural scene representation into image pixels. Different from RenderFormer, our model employs a novel combined windowed-attention and rendering-informed attention sink in the view-independent stage to improve scalability while maintaining render accuracy. To further improve versatility, RenderFormerV2 supports heterogeneous scene primitives, including environment maps and participating media, and it employs a material encoding independent of the underlying surface reflectance model that encodes material appearance via a novel neural embedding. We demonstrate the versatility of RenderFormer-V2 on a variety of scenes and perform an extensive ablation of the improved attention mechanism. --- ### Beyond Prompts: Measuring and Optimizing LLM Tool-Agent Harnesses Authors: Cen Mia Zhao, Haibo Ruan, Wenjie Chen, Pei-fen Tu, Usman Abbasi, Joel Hesch Published: 2026-09-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.05736v2 LLM tool agents can be improved without retraining by modifying the runtime harness around a fixed model: prompts, tool interfaces, middleware, state handling, and recovery logic. We study this setting as resource-bounded harness selection for fixed-model multi-turn tool agents, with the search surface scoped to prompts and tool-boundary middleware: edits are guarded intercepts at the tool boundary, not arbitrary rewriting of agent execution logic. Our optimizer-agnostic protocol reports mean held-out lift, worst-condition lift, repeatability, logged cost diagnostics, and RelLift95(B), a conservative estimate of the held-out gain of the harness selected under budget B. We instantiate the protocol with prompt-only and prompt-plus-middleware optimizers, including PRISM, which clusters failures and routes repairs to prompt, tool-boundary middleware, or joint edit surfaces within a Pareto search. On BFCL multi-round, tau2-Retail, and tau2-Telecom, PRISM obtains mean held-out lifts of 14.2, 14.9, and 10.1 percentage points and positive empirical RelLift95 on all three benchmarks, and a component ablation attributes the margin chiefly to failure-surface routing and the edit-pattern constraint. Across optimizers, the results show that some search procedures can occasionally find large gains but still choose brittle updates, so the reliability of the chosen harness should be reported alongside average held-out lift. --- ### Bigger Text Encoders Can Hurt CLIP Zero-Shot Performance Authors: Samir Char, Carles Domingo-Enrich, Randall Balestriero Published: 2026-09-04 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.05730v1 Contrastive Language-Image Pretraining (CLIP) is a building block of many machine learning applications. Scaling laws have guided resource allocation for large-scale training, yet prior work treats total CLIP model size as a single variable, without exploring how the capacity split between encoders impacts downstream performance. Here, we train multiple CLIP models with different vision and text encoder sizes, revealing that for most vision encoders, there is an optimal text encoder size beyond which zero-shot performance degrades---even as total parameter count increases. Exploiting this behavior yields efficient configurations that match the zero-shot performance of the standard ViT-B/16 architecture with up to 55% fewer parameters. We further show that this degradation stems from overfitting induced by the oversized text encoder, and that using modality-specific weight decay coefficients not only recovers but improves performance across all degraded configurations. A geometric analysis reveals a trade-off in which scaling the text encoder improves embedding uniformity but worsens cross-modal alignment; we further show that these metrics are predictive of zero-shot performance. We hope these findings motivate CLIP architectures and training methods that counteract this degradation, a prerequisite for scaling CLIP reliably and efficiently. --- ### MedWER: A Reproducible, Model-Free Evaluation Protocol for Medical Speech Recognition Authors: Justin Behling Published: 2026-09-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.05728v1 Overall word error rate hides clinically critical errors: a transcript can be 95% correct and still swap one drug for another. The usual fix weights errors on medical entities, and almost always depends on an evaluation-time named-entity recognition (NER) model or cloud API, which makes the metric's denominator a versioned black box. We present MedWER, an evaluation protocol and open-source tool for medical ASR whose denominator is a fixed, license-clean term list: 19,373 drug, diagnosis, symptom, and injury-mechanism entries projected from public sources. The protocol couples a pinned text normalizer with a phrase-aware term-restricted WER, the MedWER, so the only versioned component is a normalizer dependency held at an exact release and checked against committed golden fixtures. Coverage is validated against an independent provincial drug-benefit file the list was not built from; the matching heuristic is calibrated against ground-truth entity spans. Baselines for Moonshine~base, Whisper~base.en, and MedASR on two open benchmarks are scored with the released tool and reported with 95% confidence intervals from resampled per-utterance scores. --- ### Newton Matching for Generative Modeling: A Unified Framework for Fine-Tuning and Sampling Authors: Zeyang Li, Yunan Wang, Paolo Giaretta, Navid Azizan Published: 2026-09-04 Categories: cs.LG, cs.AI, eess.SY arXiv: https://arxiv.org/abs/2609.05727v1 We develop Newton Matching, a unified framework for fine-tuning and sampling in generative modeling. The target is $π\proptoμe^{τr}$, where $r$ is the reward, $τ>0$ the inverse temperature, and $μ$ denotes the pretrained model's terminal density for fine-tuning or the constant $1$ for sampling. We shift the paradigm from isolated losses to iterative optimization over canonical models: population minimizers of standard conditional matching for terminal densities. Under compatible smooth-realization assumptions, canonical velocities form a manifold diffeomorphic to the density manifold. Transporting the Fisher-Rao metric and mixture connection to this manifold, we show that the reverse-KL Hessian equals the metric, so the Newton direction coincides with the negative Fisher-Rao gradient. At terminal density $ρ$, each stage takes a tangential step generated by the regularized reward $r-\frac1τ\log(ρ/μ)$, followed by terminal-density-preserving canonicalization. This canonical retraction yields an exact finite-stepsize density characterization. For the ideal iteration, we prove strict reverse-KL descent away from the target for $0 < η\le τ$, global convergence under mild conditions, and local quadratic convergence for full steps ($η=τ$). Covariance and gradient forms, each with forward or reverse regression-pair constructions, yield sample-wise tangential-update losses with the same population minimizer, without importance sampling or full-trajectory backpropagation. We develop approximate updates and define critical-point consistency as vanishing tangential displacement if and only if $ρ=π$. We recover representative methods as exact realizations, critical-point-consistent approximations, or objective-altering variants, enabling modular algorithm design. Our work advances the theory and algorithms of reinforcement learning for generative models. --- ### Recovering Temporal and Geographic Signals from Language Model Embeddings Authors: Esteban Feuerstein, Victoria Klimkowski, Juan Manuel Ortiz de Zarate, Federico Hernán Suaiter Published: 2026-09-04 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.05721v1 Understanding whether language-model embeddings encode structured real-world information is important for both representation analysis and information retrieval. We study this question for temporal and geographic signals using a simple projection-based method that operates directly on output embeddings. Given a small set of seed examples, the method defines an axis in embedding space and ranks texts or entities by their projection onto that axis. Our approach is fully black-box and model-agnostic: it requires only embeddings, without access to model weights, internal activations, auxiliary probes, or additional training. This makes it applicable to modern embedding models available only through APIs and provides a lightweight way to analyze whether temporal and spatial dimensions are present in their representation spaces. We apply the method to temporal and geographic datasets and find that embedding projections recover meaningful chronological and spatial structure. These results provide evidence that output embeddings encode signals relevant to time and space, while also offering a practical tool for interpretability and for downstream temporal and geographic information retrieval tasks, such as temporal ordering, geographic ranking, and tagging. --- ### Tight Lower Bounds for State Tomography with Limited Entanglement Authors: Ufuk Keskin, Jason Luo, Mahbod Majid, Matthew Radzihovsky Published: 2026-09-04 Categories: quant-ph, cs.IT, cs.LG arXiv: https://arxiv.org/abs/2609.05718v1 We study state tomography when each measurement acts on at most $k$ fresh copies and no quantum memory is retained between blocks. We prove a lower bound matching the upper bound in (arXiv:2510.07788). Thus the copy complexity of estimating an arbitrary $d$-dimensional state to trace distance $ε$ is, up to absolute constant factors, $\max\{d^3/(\sqrt{k}ε^2),d^2/ε^2\}$ for every $k$ and all sufficiently small $ε$. This removes the earlier restriction that $k$ be small as a function of the accuracy. The lower bound applies to arbitrary measurements within each block and adaptive choices between blocks. The lower bound already applies in a small neighborhood of any state whose smallest eigenvalue is of order $1/d$, even when the center is known. The main ingredient is a uniform Fisher information bound for one measurement block that depends only on the smallest eigenvalue of the state. The proof avoids the perturbative expansion responsible for the restriction in (arXiv:2402.16353). Fano's inequality for metric balls and a log-Sobolev comparison between mutual and Fisher information then reduce the adaptive protocol to this block bound (arXiv:1607.00550, arXiv:1902.08582). --- ### CUSP: Decomposable Collective Uncertainty for Multi-Agent Multimodal Reasoning Authors: Chung-En Johnny Yu, David Garcia, Brian Jalaian, Nathaniel D. Bastian Published: 2026-09-04 Categories: cs.AI, cs.LG, cs.MA arXiv: https://arxiv.org/abs/2609.05708v1 Aggregating heterogeneous vision-language models (VLMs) can improve multimodal reasoning, but neither an individual model's confidence nor that of the aggregated answer measures reliability at the system level. We present CUSP (Collective Uncertainty through Semantic Opinion Pooling), a training-free uncertainty quantification framework that maps multiple VLM responses to a shared semantic response space, pools them into a pooled semantic opinion, and reports two complementary system-level signals: collective uncertainty, the dispersion of the pooled opinion, and Jensen-Shannon divergence (JSD), the conflict among the model-level opinions. Within this pooled semantic opinion, the unnormalized collective entropy decomposes exactly into the mean of the models' individual semantic entropies and the JSD, separating total dispersion from model conflict. Requiring neither token logits nor calibration labels, CUSP applies to open-weight and commercial VLMs alike. In static multi-VLM ensembles, collective uncertainty is the strongest signal in the small-model regime (0.764 AUROC for prediction-error detection, 0.889 AUARC for abstention), outperforming uncertainty baselines majority voting and naive selection by 4.7 to 15.8 points and widening its margin as the ensemble grows; JSD is strongest in the evaluated commercial regime (0.819 AUROC, 0.910 AUARC) and ranks hard-answer model conflict with AUROC up to 0.982. The pooled prediction also improves accuracy over the average single model by 5.6 to 13.0 points. Over the full trajectory of a multi-step, multi-agent system, subagent collective uncertainty ranks system failures above chance (0.619 AUROC) and gives the best abstention ordering among the evaluated signals (0.699 AUARC). --- ### Intra-Prompt Parallel Decoding for Common-Context Question Answering Authors: Theodore Glavas, Nikhita Vedula, Dushyanta Dhyani, Antonios Valkanas, Yilun Zhu, Shervin Malmasi Published: 2026-09-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.05707v1 In common-context question answering (CCQA) tasks, multiple input questions share a common context to base their answers from. However, Large Language Models typically generate each answer using an independent prompt. While existing batching and caching techniques help improve parallelism and reduce repeated computations, the separation of questions across prompts limits the achievable speedup, as modern GPUs are underutilized due to a memory bottleneck during attention. We present Intra-Prompt Parallel Decoding (IPPD), a novel inference method that answers multiple common-context questions in parallel within a single prompt. IPPD directly addresses the bottleneck by efficiently sharing both memory and computation during the attention process, as the next token for every question is decoded in a single inference step. IPPD uses virtual position IDs and attention mask manipulation to generate the same output as standard prompting without requiring fine-tuning or any changes to the LLM architecture. Since all parallelism occurs within a prompt, IPPD is fully compatible with batched inference, even when each prompt features a different context. Our experiments show that IPPD delivers up to 7X the effective throughput as standard decoding without quality degradation, and outperforms prefix caching with PagedAttention in most settings. --- ### Characterizing Privacy Risks of Quantum Machine Learning with Emergent Quantum-Native Access Authors: Liou Tang, James Joshi, Ashish Kundu Published: 2026-09-04 Categories: quant-ph, cs.LG arXiv: https://arxiv.org/abs/2609.05702v2 Quantum Machine Learning (QML) has shown rapid advances by utilizing quantum computing for machine learning tasks. Meanwhile, the privacy risks accompanying QML is also starting to be studied, which inherit privacy leakage channels from "classical" ML and also quantum-unique risks. Existing work on privacy-preserving QML largely focuses on a QML-as-a-service scenario, which generally assumes that the QML model owner provides only classical bit outputs to queries, while users (and adversaries) have only classical computing abilities. However, this view is increasingly challenged in a quantum-native world of quantum-capable users/adversaries, which may have access to both quantum computing abilities and access to quantum information output from service providers. In this paper, we aim to bridge this gap by examining membership inference attacks against QML models by demonstrating that increasing quantum access and quantum computing abilities provides provable theoretical privacy leakage and empirical adversarial gain. However, the probabilistic nature of QML introduces a gap between theoretical and empirical adversarial advantage. These results show that existing research on privacy leakage in QML models underestimates privacy leakage in emergent quantum-native access regimes, and we hope to establish a first step in examining potential privacy leakages for QML in the quantum-native world. --- ### XAI-SDN: An Explainable Entropy-Guided Machine Learning Framework for Real-Time DDoS Detection in Software Defined Networks Authors: Adeel Ahmad, Ali Akarma, Ahmad Ali, Hammad Muneer, Toqeer Ali Syed Published: 2026-09-04 Categories: cs.CR, cs.AI, cs.NI arXiv: https://arxiv.org/abs/2609.05701v1 One of the biggest risks faced by Software Defined Networks (SDN) is the Distributed Denial of Service (DDoS) attack in which a compromised controller can make an entire network unusable. To address these challenges, we suggest an entropy-guided machine learning framework, called XAI-SDN, for real-time DDoS detection in SDN environments which is lightweight and explainable. The framework extends the flow features extracted by CICFlowMeter with eight Shannon entropy metrics obtained by an $\mathcal{O}(1)$ rolling algorithm and uses a Random Forest classifier with SHAP TreeExplainer for providing transparency at the prediction level. On a fixed temporal split, XAI-SDN achieves an accuracy of 99.9987\%, a macro F1-score of 99.9621\%, and an AUC-ROC of 1.0000 on the full 3.59 million flows of the CIC-DDoS2019 SYN benchmark. The pipeline sustains 0.0165~ms per flow (60{,}606 flows/s) without the use of SHAP and 0.5122~ms per flow (1{,}953 flows/s) with full support of SHAP under the 99.14\% prevalence of DDoS traffic, which is a step towards achieving a balance between the detection performance and operational transparency in next-generation SDN security. --- ### Analysis of Respiratory Sinus Arrhythmia with Neural Networks Authors: Julian Szymanski, Patryk Orkisz, Higinio Mora Published: 2026-09-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.05698v1 The paper introduces a neural network-based approach for analyzing ECG signals to estimate respiratory rate by leveraging the phe- nomenon of Respiratory Sinus Arrhythmia (RSA). Our method employs a deep learning model trained to predict respiratory waveforms directly from ECG input data. To achieve this, we developed and evaluated three different neural network architectures capable of automatically extract- ing relevant features from ECG signals without the need for manual preprocessing. The proposed approach offers a robust and scalable solu- tion for non-invasive respiratory monitoring, with potential applications in healthcare and wearable technology --- ### GraphNOSE: A Graph Transformer in Olfaction Authors: Mrityunjay Sharma, Sarabeshwar Balaji, Valentina Parma, Ritesh Kumar Published: 2026-09-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.05694v1 Predicting olfactory qualities from molecular structure is an open problem in chemoinformatics. Although linear models can link molecular features to odor descriptors, they often fail when extrapolating to novel chemical scaffolds, extreme molecular weights, or complex odor mixtures. To address this, we introduce GraphNOSE, an open-source graph transformer framework that predicts multi-label odor descriptors from simplified molecular-input line-entry system (SMILES) strings for single molecules and binary mixtures. By integrating positional and structural encodings within a transformer-based graph architecture, GraphNOSE achieves strong performance with six times fewer parameters than standard graph neural network (GNN) baseline while consistently outperforming linear models, molecular language model embeddings, molecular fingerprints, and baseline GNNs by an average area under the ROC curve (AUROC) margin of 4.52% (p < 0.01). GraphNOSE achieves an AUROC of 84% on out-of-distribution compounds (OODs). This exceeds the current state-of-the-art GNN for OOD in olfaction (Open-POM: 81%, p < 0.001), and identifies conditions under which linear models empirically fail. Finally, we apply XAI (explainable AI) methods to identify which substructures and molecular features drive odor predictions, yielding insights consistent with chemical intuition and grounded in the model's learned representations. Together, these results establish GraphNOSE as a scalable and interpretable architecture for olfactory prediction that generalizes to structurally distinct compounds underrepresented in current perceptual databases. --- ### Connecting Score Matching, Maximum Likelihood, and Expectation-Maximization in Mixed Linear Regression Authors: Zhankun Luo, Abolfazl Hashemi Published: 2026-09-04 Categories: cs.LG, stat.ML arXiv: https://arxiv.org/abs/2609.05688v1 We study variance-preserving diffusion of the response in mixed linear regression (MLR) with unknown mixing weights. Our analysis separates the statistical guarantees of score matching from the loss geometry and optimization signal at a fixed diffusion noise level. The KL divergence links the denoising score matching objective integrated over the diffusion path with the likelihood and a terminal discrepancy. Under mild regularity conditions and terminal schedule, the resulting estimator converges up to the ground truth parameters of MLR, and its scaled error converges to the Gaussian limit of the maximum-likelihood estimator. At a fixed scale of the diffusion noise level, we derive a decomposition linking the score matching loss to cross-entropy and Expectation-Maximization (EM) operators. This decomposition yields an EM-related low-noise gradient expansion with additional correction terms of latent variance. In the high-noise limit, we further characterize gradient descent on this limiting loss under isotropic covariance. Along fixed high signal-to-noise ratio rays, the score matching imbalance gradient and the latent-variance term tend to zero pointwise. Numerical experiments illustrate our theoretical findings and statistical guarantees. --- ### Nonparametric Hypothesis Testing of High-dimensional Clustering With Application to Single-cell RNA Data Authors: Yifan Dai, Di Wu, Yufeng Liu Published: 2026-09-04 Categories: stat.ME, stat.AP, stat.CO, stat.ML arXiv: https://arxiv.org/abs/2609.05683v1 Single-cell RNA sequencing studies routinely use clustering to define putative cell types and cell states, yet the observed separation may arise from sampling variability rather than genuine biological heterogeneity. This paper studies formal significance testing of such clustering structure in high-dimensional data. Existing SigClust methods assess clustering significance through Monte Carlo simulation under a Gaussian single-cluster null, but this assumption can be unreliable for normalized gene expression data and other non-Gaussian settings. We propose SigClust-LCP, a nonparametric extension that models a single cluster by a log-concave distribution. To make this approach computationally feasible in moderate to high dimensions, we develop a score-matching estimator for log-concave projection inspired by recent generative modeling ideas. We establish theoretical guarantees for the estimator and for its use in clustering significance testing. Simulations show that SigClust-LCP controls Type-I error more reliably than existing methods across a range of unimodal and mixture distributions while retaining competitive power. In a single-cell RNA sequencing analysis of Hydra cells, the method avoids spurious subclusters within annotated cell populations and supports biologically meaningful separation across lineages and body-axis regions. --- ### A Rubric-Guided Large Language Model Solution for Opioid Use Disorder Computable Phenotyping Authors: Mengxian Lyu, Paredes Pardo, Cheng Peng, Ziyi Chen, Mengyuan Zhang, Jieting Li Lu, Gary M Reisfield, William M Greene, Jenny Lo-Ciganic, Yonghui Wu Published: 2026-09-04 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.05682v1 Opioid use disorder (OUD) remains a public health crisis in the United States, yet it is difficult to identify from electronic health records (EHRs) because missing diagnosis codes and supporting evidence are buried in clinical narratives. Accurate OUD identification is critical to support interventions and improve health outcomes. This study developed a rubric-guided large language model (LLM) that incorporated Optimization by PROmpting (OPRO) for OUD computable phenotyping (CP). The framework used an 18-item, expert-identified rubric to instruct LLMs to automatically extract critical text with supporting evidence to determine OUD flags. Two UF Health physicians (GMR and WMG) chart-reviewed 253 patients, including 68 OUD-positive cases. Our LLM-based computable phenotype (CP) achieved the best F1 score of 0.774 and an AUROC of 0.934, outperforming the machine learning-based CP using EHR and natural language processing-extracted variables, and zero-shot LLMs by relative F1 improvements of 12.8% and 44.4%, respectively. The proposed LLM-based CP could link LLM-extracted evidence to OUD phenotyping for better explainability. --- ### Who Maintains Agent Skills? A Longitudinal Study of Human-Governed, AI-Assisted Skill Maintenance Authors: Chen Shen, Estevam Hruschka Published: 2026-09-04 Categories: cs.CL, cs.AI, cs.HC, cs.SE arXiv: https://arxiv.org/abs/2609.05677v1 Lifelong LLM agents increasingly rely on external skill artifacts as one element for preserving and reusing capabilities over time. These skills (usually portable Markdown files such as SKILL.md) describe when and how to apply a capability and must be corrected, expanded, and consolidated as tools and usage patterns shift over deployment. Recent work seeks to automate skill curation, but it largely evaluates against automated baselines and treats human maintenance as an unmeasured bottleneck. We study that missing process directly. We mine the full commit histories of five public AI-skill repositories, a purposive sample of AI-tooling organizations, covering 873 commits, 143 skill files, and 254 substantive post-creation edits from October 2025 to June 2026. We code each edit with pre-registered governance, operation, and trigger-evidence codebooks. Three findings emerge. First, every substantive edit is authored or merged through a named human account, while 62% carry an AI co-author trailer, with large repository-level variation. Second, these edits are genuine curation: an audited sample shows that most change skill content, and the coded operations are dominated by additions and corrections. Third, a pre-registered rule-likeness axis fails its reliability gate; reliably coding rule-likeness from commit artifacts remains an open measurement problem. We release the corpus, codebooks, mining scripts, and a replay protocol for automated skill curators. For self-evolving agents, public skill maintenance currently looks less like an autonomous pipeline than a human-governed, AI-assisted loop that future curators must measure against and operate within. --- ### PAC-Private Autoregressive Generation: Calibrating Noise to Ensemble Disagreement Authors: Mina Mirzadehsarcheshmeh, Amir Keyvan Khandani Published: 2026-09-04 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.05676v1 Language models adapted on private text are often served through APIs, so privacy leakage occurs through generated outputs rather than exposed weights. Private prediction protects these releases. Methods such as PMixED incur privacy cost at each release and increasingly rely on the public model over long horizons. PAC privacy instead calibrates noise to output variability across possible secrets, adding less noise when predictions are stable. To our knowledge, PAC-private prediction has not previously been extended from classification to autoregressive generation. We construct $m=128$ overlapping worlds from the private corpus, with each record appearing in exactly $m/2$ worlds, and train one adapter per world over a frozen public model. The realized world is the secret. At each token, the public model defines a candidate set, the worlds vote, and their posterior-weighted disagreement determines the PAC noise; unanimity requires no calibration noise. We prove $I(S;Y_{1:T}) \leq I(S;H_T) \leq bT$. Our contributions are extending PAC privacy to autoregressive generation, handling adaptive self-generated contexts, and introducing coupled decoding that preserves privacy accounting while avoiding greedy degeneration. On WikiText-103 with GPT-2-small, we retain 74% of the fine-tuning gain at a per-token budget of $2^{-32}$, while membership-inference success is bounded by 51.08% after $10^6$ tokens; posterior-entropy estimates of leakage are roughly 17% of the charged budget. Inference privacy is not content protection: even when membership advantage on a memorized canary is indistinguishable from zero, the canary is emitted at the same rate. Against PMixED under matched membership-inference bounds on the same data universe and test set, we retain 98% of non-private headroom from $10^2$ to $10^6$ tokens, versus at most 56%, with no crossover. --- ### What LLM Trading Agents Actually Do in Production: A Six-Month, Population-Scale Record from Two Fleets Authors: T. J. Barton, Chris Constantakis, Patti Hauseman, Annie Mous, Alaska Hoffman, Brian Bergeron, Hunter Goodreau Published: 2026-09-04 Categories: cs.AI, cs.CE, cs.MA arXiv: https://arxiv.org/abs/2609.05663v1 We present a continuous, population-scale measurement record of autonomous language-model trading agents operating in production across two systems with one design lineage: DX Terminal Pro (3,505 user-funded vaults trading real ETH in Base memecoin markets for 21 days, February to March 2026) and the DXAP live alpha fleet (500 to 599 user-created agents all-history, 91 to 117 concurrently active, trading Hyperliquid perpetuals, June to August 2026). The record spans roughly six months, 7.5M single-model invocations with about 300K onchain actions, and a further 231,638 multi-tool turns producing 14,596 fills. Four findings carry the paper. First, the operating layer determines behavior more than anything written in strategy text: a risk slider explains leverage (+0.425 per level), agent fixed effects absorb 60% of variance, and a leaderboard render boundary causally routes selection (regression discontinuity 1.75x at the top-3 cut). Second, sizing is volatility-blind: median leverage is 5.0x in every volatility sextile, and one posture-slider cell (11% of the book) holds 62% of liquidations. Third, agents capture almost none of the upside they reach: 43.2% of positions saw at least +300 bps of favorable excursion within 24h, yet 49.3% of those closed with a negative trade return; a mechanical bracket recovers +39.0 bps per position. Fourth, neither fleet shows a directional edge. The DXAP fleet is not profitable and trails a matched Hyperliquid retail benchmark (41% vs. 50% roundtrip win rate). A paired-replay league of frontier models on 416 captured production scenarios finds decision quality statistically indistinguishable at this horizon, while choice stability differs sharply across model families. Every headline survives day-clustered inference, permutation nulls, and a common-fee restatement; the paper closes with a 17-rule methodology canon bought with our own retractions. --- ### Full-Page Optical Music Recognition of Handwritten Monophonic Scores Authors: Adrian Rosello, Antonio Ríos-Vila, David Rizo, Jorge Calvo-Zaragoza Published: 2026-09-04 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.05662v1 Full-page end-to-end Optical Music Recognition seeks to transcribe entire music pages directly into symbolic notation, avoiding the limitations of traditional pipelines that rely on accurate staff segmentation. Recent Transformer-based architectures have achieved strong performance on typeset scores, relying on large-scale synthetic data for pretraining. However, their applicability to handwritten music remains largely unexplored. In this work, we study full-page transcription on handwritten monophonic collections and analyze the impact of synthetic pretraining in this setting. To investigate which factors are most relevant during pretraining, we introduce a generator capable of producing visually coherent full-page scores in both typeset and handwritten styles. Experiments on three real handwritten datasets provide a comparative evaluation of several full-page pipelines and different synthetic pretraining strategies. The results suggest that the benefits of synthetic pretraining are primarily associated with learning structural layout conventions rather than with visual similarity to the target handwriting. --- ### Srijika: OpenType-Layout-Reusing Font Restyling for Nine Indic Scripts Authors: Anil Pai Published: 2026-09-04 Categories: cs.CV, cs.GR, cs.LG arXiv: https://arxiv.org/abs/2609.05661v1 We present Srijika, a system for producing installable OpenType fonts for nine Brahmic scripts: Devanagari, Tamil, Bengali, Telugu, Kannada, Malayalam, Gujarati, Gurmukhi, and Odia. Rather than generating fonts from scratch, Srijika restyles glyph outlines from shaping-complete template fonts. It preserves the template's cmap and GSUB closure and its GPOS data under a documented metric policy, making every output a complete font by construction. This addresses a central challenge of Indic font generation: hundreds to thousands of conjuncts, half forms, and matra variants must remain mutually consistent under OpenType shaping. Srijika produces 66 TTFs: 57 curated presets and nine open-vocabulary showcase fonts. All pass the OpenType Sanitizer, while HarfBuzz and CoreText reproduce the template glyph-ID sequences on conjunct-heavy probes. A full-closure audit covering 80,915 glyphs and 54,812 anchors quantifies metric changes. Natural-language style selection uses Lipika, a retrieval index over approximately 650 open-license font families. A reference-conditioned latent diffusion model redraws template glyphs in the selected style, followed by content gating, harmonization, and shaped-cluster verification with fallback to template outlines. We evaluate against no-learning baselines. On diffusion-training-family-held-out SSIM gates, template copying outperforms generation on 50 of 56 faces. Style movement is measurable only with an internal same-model embedding whose training corpus includes the held-out families, so these results require caution. A learned baseline, independent style metric, and human study are outside this report's scope. Our contributions are the layout-reusing formulation and pipeline, its nine-script audit and benchmark, and a negative-results catalogue covering failed conditioning, objective choices, and data-hull limits of reference-guided restyling. --- ### Robustness of LLM-Generated SystemVerilog Assertions to Semantics-Preserving RTL Transformations Authors: FNU Aditi Published: 2026-09-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.05658v1 Large language models (LLMs) are increasingly being explored for automating SystemVerilog Assertion (SVA) generation, yet most evaluations report correctness on a single syntactic representation of an input. Such point accuracy does not reveal whether a model's correct output is stable when the same RTL behavior is written differently. This paper presents a controlled metamorphic evaluation of LLM-based SVA generation under semantics-preserving RTL transformations. Starting from the VERT dataset, we construct a quality-filtered conditional-control pool and a stratified 40-program evaluation set containing 295 assignment behaviors. We evaluate two open code models, Qwen2.5-Coder-7B and DeepSeek-Coder-V2-Lite, with an identical evaluation prompt and greedy decoding. Three transformations are studied: operand reordering, deterministic identifier renaming, and redundant parenthesization. Beyond baseline and transformed accuracy, we measure conditional robustness, invariance failure, and any-flip rate, with 10,000-sample clustered bootstrap intervals at the RTL-program level. Across all six model-transformation conditions, 9.7%-27.0% of behaviors that were correct on the original RTL become incorrect after a semantics-preserving transformation. Aggregate accuracy can therefore hide substantial instability: under identifier renaming, DeepSeek-Coder-V2-Lite improves from 53.9% to 63.7% accuracy while 19.5% of its originally correct behaviors fail. Manual review of 30 sampled correct-to-wrong transitions identifies dropped path predicates, branch-polarity errors, Boolean-structure corruption, and output-contract violations. The results show that point accuracy alone is insufficient for characterizing LLM reliability in assertion generation and motivate robustness-aware evaluation for AI-assisted hardware verification. --- ### Endogenous Exploration in Reinforcement Learning with Intrinsic Curiosity Authors: Armando Vieira Published: 2026-09-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.05650v1 We propose a reinforcement learning framework in which exploration is driven by intrinsic curiosity, designed for scenarios where environments are non-stationary and rewards are sparse, delayed, uninformative, or absent. In our model, action selection is guided by a combination of external rewards and an epistemic motivation mechanism that biases the agent toward structured exploratory directions. The central hypothesis is that effective exploration emerges at intermediate levels of incoherence, while performance degrades under both overly rigid and overly disordered dynamics. To test this idea, we implement the framework on top of a Liquid State Machine (LSM) substrate and evaluate it on two standard benchmarks: the discrete-action LunarLanderv2 and the continuous-control BipedalWalkerv3. The proposed method achieves competitive performance on both tasks relative to established deep RL algorithms, including Proximal Policy Optimization (PPO) and Intrinsic Curiosity Module (ICM). We further show that the curiosity window is not recovered in Active Inference agents under the same analysis, suggesting that the proposed dynamics capture a distinct exploration regime --- ### The convergent laboratory: when AI reasoning, autonomous experiments, high performance and quantum computing reshape chemistry Authors: Eliu Huerta, Xiaoyun Wang, Geetika Gupta, Edward H. Sargent, Cameron J. Owen, Victor Fung, Abhishek Mitra, Austin Cheng, Emma Bouchard, Shams Mehdi Published: 2026-09-04 Categories: cs.AI, cond-mat.mtrl-sci arXiv: https://arxiv.org/abs/2609.05643v1 This Comment emerges from TPC26 (https://tpc26.org), a conference convening leaders from academia, national laboratories, and industry who are reshaping materials science discovery. The meeting explored how AI, autonomous agents, self-driving labs, higher performance and quantum computing converge to amplify their individual impact on materials science discovery. The perspectives here reflect the firsthand experiences of researchers at these frontiers and capture the essence of this global endeavor. As AI-driven reasoning, autonomous agentic frameworks, self-driving laboratories, and fault-tolerant quantum processors mature simultaneously, we offer this Comment as a reference at what we believe is a tipping point of transformative advances and productive disruption in the chemical sciences. --- ### A Nuclear-Norm Lower Bound for Dithered Scalar Quantization of Matrix Products Authors: Piyush Sao, Narasinga Miniskar, Pedro Valero-Lara, Keita Teranishi, Sudip Seal Published: 2026-09-04 Categories: cs.IT, cs.LG, math.NA arXiv: https://arxiv.org/abs/2609.05641v1 We consider the problem of minimizing error in quantized matrix multiplication $C=AB$. Scalar quantization of the factors introduces rounding errors whose scale depends on the maximum absolute entries -- the ranges -- of their rows and columns. These ranges determine the quantization grid steps. To reduce the error, we optimize over product-preserving transformations that alter the factor ranges and grid steps without changing $C$. Specifically, we seek the smallest leading expected squared error over invertible inner changes of basis and orthogonal outer rotations. Under independent, zero-mean subtractive dither noise on an unbounded lattice, we prove the output-only bound $E_{\rm lead} \ge (c_A+c_B)/K \Vert AB\Vert_*^2$, where $K$ is the inner dimension, $c_A$ and $c_B$ are normalized noise variances, and $\Vert AB\Vert_*$ is the nuclear norm. The bound is tight: an SVD-aligned Hadamard construction attains the infimum whenever a Hadamard matrix of order $K$ exists, including every power of two, while an SVD-aligned DCT construction is within a factor of two for every $K$. Without outer rotations, Gram-matrix balancing minimizes factorization energy, and finite-set flattening achieves the bound within $C\log(K(m+n))$. For power-of-two $K$, conditional expectations deterministically select the Hadamard signs in $O((m+n)K^2)$ exact-real operations. Synthetic experiments verify both constructions and illustrate the tradeoff between regularization and conditioning. These results characterize the full-gauge optimum and quantify the cost of preserving row and column indices. --- ### Facial Age Estimation for Age Fraud Detection in National ID Systems Authors: Sharib Athar, Arka Koner, Chetan Naik, Barada P. Sabut, Tanusree Deb Barma, Anoop M. Namboodiri, Anil K. Jain Published: 2026-09-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.05638v1 Identity fraud during biometric enrollment and updates remains a major challenge for large-scale national identity systems. A common fraud vector is misrepresenting one's age to access age-restricted services or welfare schemes. In this work, we present SwinAge, a facial age estimation system designed for use within the Aadhaar biometric enrollment pipeline, to assist quality-check (QC) operators to flag potential age-related fraud. This is critical for a system like Aadhaar (the world's largest national identity programme), that holds about 1.5 billion unique identities, with 22.4 million new enrollments and 283 million updates in the last year. Building upon the SwinFace architecture with landmark-based similarity (warp affine) alignment, we train on a large in-house dataset of 1.45 million face images and evaluate on an independent, age-stratified test set of 283K images, both drawn from an ethnically diverse population of 716K unique subjects. We investigate three Aadhaar-specific operational thresholds (5, 18, and 60 years) and propose a deployment triage framework that flags suspected cases for manual review. Following NIST FATE, we report false acceptance/rejection rates (FAR/FRR) at each threshold rather than aggregate accuracy: at 1% FAR the model achieves an FRR of 3% (<5yrs), 0.4% (>18yrs) and 11.0% (>60yrs). SwinAge achieves a mean absolute error (MAE) of 2.94 years on the same test set, outperforming three zero-shot vision language models on all benchmarks, and improving the state-of-the-art on 5 out of 7 public benchmark datasets. We further report per-gender errors and distill lessons for national identity programs. --- ### Better Together: Complementary Query Rewriting Under a Strong RAG Baseline Authors: Sara Shanian, Xiaoqin Yi, Pavlo Ruban, Kurt MacDonald Published: 2026-09-04 Categories: cs.CL, cs.IR arXiv: https://arxiv.org/abs/2609.05637v2 A popular way to improve Retrieval-Augmented Generation (RAG) is to rewrite the user's question into several variants and search with all of them. We test whether this actually helps once the underlying search is already strong. Under one fixed, competitive pipeline (BGE dense retrieval, cross-encoder reranking, and MMR diversification), we compare four query-rewriting strategies (S1-S4) against two strong LLM baselines (HyDE, Query2Doc) on three datasets (HotpotQA, AmbigNQ, and the 512K-document EnterpriseRAG-Bench) over three seeds with paired-bootstrap significance tests. Our headline result is that rewriting alone is at best competitive with a strong baseline, but combining methods yields outsized gains because different strategies fail on different questions. A post-hoc union of four methods (S1+S3+S4+HyDE) improves HIT@10 over the baseline by +12.5 points on enterprise data (51.70 vs 39.22), and a five-method union reaches 52.98 (+13.8). Budget-matched controls capture only ~40% of this gain, confirming that complementarity, not retrieval budget, is the primary driver. On HotpotQA the union adds +1.6 to +1.8 points (p<0.001), saturating the all-method oracle; on AmbigNQ the same fusion hurts (-2.4 below the best solo, p<0.001), and we analyze when and why. Because rewriting is expensive, we evaluate in simulation a confidence-gated router that runs rewriting only when the baseline's own top-1 score is low. It captures about half of the enterprise full-merge gain (+4.3 HIT@10) while paying rewriting cost on <40% of queries, and automatically declines to rewrite on AmbigNQ. A downstream answer-quality evaluation confirms the router improves F1 by +1.92 (p<0.01) at roughly 40% of the expansion cost. In short: treat query rewriting as a complementary coverage source, applied through cost-aware routing, not as a standalone replacement for a strong baseline. --- ### Unsupervised Transfer Clustering for Mitigating Cold Start in Active Prompt Learning Authors: André Camargo Portella, Samuel Felipe dos Santos, Jurandy Almeida Published: 2026-09-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.05636v1 Vision-Language Models (VLMs) are able to achieve impressive zero-shot classification performance by aligning visual and textual representations, but each new task still demands handcrafted prompts. Active Prompt Learning (APL) combines Active Learning (AL) and Prompt Learning (PL) into a single framework, allowing for the usage of the VLM prior knowledge for iteratively querying the most informative images to be labeled. However, the cold-start problem is still relevant for APL methods, where the performance of the initial query can be worse than random sampling. While recent state-of-the-art APL methods mitigate with balanced sampling and multimodal features, they rely on rigid, distance-based clustering to group these features. This simplistic approach can struggle to capture the complex, high-dimensional semantic distributions inherent to VLMs, leading to suboptimal query representativeness. Unsupervised transfer can be applied to these features as a possible alternative, since it is capable of inferring the underlying human labeling of a task without any form of supervision. This way, samples can be grouped in semantically coherent clusters. This paper proposes Unsupervised Transfer Clustering with Selective Querying (UTC+SQ), a framework that enhances a recent APL approach by leveraging state-of-the-art unsupervised transfer model. These models generate high-fidelity pseudo-labels that establish semantically meaningful clusters, allowing for the selection of more relevant samples. Experimental evaluations demonstrate that shifting from distance-based to projection-based clustering improves the representativeness of the queried subset, achieving accuracy gains in 6 of the 8 datasets tested. --- ### TamilEOT: A Dataset and Model for Semantic End-of-Turn Detection in Tamil Telephone Speech Authors: Santhoshkumar V Published: 2026-09-04 Categories: cs.CL, cs.AI, cs.SD arXiv: https://arxiv.org/abs/2609.05631v1 A voice agent has to decide, at every pause, whether the user has finished speaking. Without a model of the language that decision falls back to a fixed silence timeout: set it short and the agent interrupts, set it long and every turn pays the full wait. Open semantic end-of-turn detectors exist, but to our knowledge none covers a South Indian language. We release TamilEOT: 18,485 labelled turn boundaries cut from 116 real Tamil telephone conversations, and two audio-only detectors fine-tuned from Smart Turn v3. On a held-out split of 4,168 clips from 30 unseen calls, accuracy rises from 70.30% zero-shot to 83.71% (8.7 MB) and 86.13% (21 MB); ROC-AUC rises from 0.751 to 0.921. Both models run in under 150 ms single-threaded on a laptop CPU. We also report what building it cost. Rule-derived labels, checked against a blind human listening pass, were right 95.9% of the time on the positive class and 44.4% on the negative class, which is below chance, because the rule answered a different question than the model is asked. Replacing them with an audio-LLM labeller measured at 97.5% human agreement cost US$5.69. Of every training lever we measured, only encoder capacity moved the result; three runs at identical config and seed span 0.87 accuracy points, which is the floor below which none of our other deltas mean anything. Replaying the same labelled boundaries through the production VAD and streaming adapter costs a further 2.60 points, and 7.8% of boundaries are never surfaced to the model at all. Data, weights, code and every negative result are public. --- ### Bias-Corrected Subspace Intersection: Minimax-Optimal Shared Subspace Estimation in Multi-View Data Authors: Xianwen Song, Yuepeng Yang, Cong Ma Published: 2026-09-04 Categories: stat.ME, cs.IT, math.ST, stat.ML arXiv: https://arxiv.org/abs/2609.05617v1 Estimating a low-dimensional subspace shared across noisy data matrices is a fundamental problem in multi-view matrix estimation. We study this problem under the two-view JIVE model, where each data matrix contains shared and view-specific low-rank components. We demonstrate that standard plug-in subspace intersection, including AJIVE, suffers from a second-order bias caused by direction-dependent leakage of the empirical singular vectors. We propose bias-corrected subspace intersection (BCSI), which removes this bias before estimating the shared subspace. We establish finite-sample risk bounds for BCSI that accommodate unequal view dimensions, signal strengths, and view-specific ranks and require no condition-number assumptions on the signal matrices. When the shared and view-specific ranks are comparable, these bounds match our minimax lower bounds up to universal constants. The resulting minimax rate contains a new second-order term, arising from quadratic leakage perturbations relative to the shrinking spectral gap when the view-specific subspaces are nearly aligned. This term is absent from previous JIVE minimax lower bounds. Numerical experiments demonstrate the advantage of BCSI over AJIVE when the leakage bias is pronounced. Along the way, we establish a nonasymptotic concentration result for the bias-corrected leakage Gram matrix of a rectangular spiked matrix, which may be of independent interest. --- ### Time-Aware Assistive Navigation Authors: Masaki Kuribayashi, Zhongkai Shangguan, Eshed Ohn-Bar Published: 2026-09-04 Categories: cs.CV, cs.RO arXiv: https://arxiv.org/abs/2609.05596v1 Can interactive vision-and-language agents learn not just what to say but also \textbf{\textit{when}} to say it? Current language models rarely plan over whether and when to realize a real-time response to a user. However, providing accurate and timely support for human decision-making, such as when guiding visually impaired individuals through urban environments, requires careful real-time responsiveness--poorly timed responses can distract users or add unnecessary cognitive load. As a machine intelligence challenge for Multimodal Large Language Model (MLLM)-based agents, we introduce a large-scale multimodal benchmark for an egocentric, assistive navigation task in complex outdoor environments. Using this benchmark, we uncover a fundamental limitation of off-the-shelf MLLMs in delivering safe and time-sensitive navigation instructions, even with model fine-tuning on substantial amounts of data. We then demonstrate that a simple yet effective modification of the model, including direct supervision to predict the underlying reason for each instruction, yields significant performance gains across open-loop, closed-loop, and sim-to-real generalization settings. However, our analysis highlights persistent challenges in temporal reasoning, safety-critical object awareness, and relational and distance understanding. To advance the development of scalable assistive agents, we will release our simulation, benchmark, and code (available at the project website: https://timeli-icra.github.io/). --- ### SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolution Authors: Xingjian Ran, Xiaoye Mo, Sihao Liu, Jianyu Zhang, Li Luo, Bo Dai Published: 2026-09-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.05594v1 Diverse and simulation-ready indoor scenes are essential for interactive entertainment and embodied AI, yet their scalable generation remains challenging. Recent agentic text-to-3D scene pipelines that rely on vision-language models (VLMs) can generate scenes of high fidelity but require costly iterative object placement and refinement. Another mainstream paradigm, parametric image-to-3D scene models, produces scenes efficiently from strong priors learned from 2D images but often leads to imprecise and physically invalid scenes. More importantly, both paradigms struggle to output diverse scenes for a single input, making it hard for them to reflect the dynamically changing nature of real scenes. In this paper we propose \textbf{SceneMosaic}, a framework that combines the merits of both paradigms. It obtains the initial candidate from the learned image-based prior, and subsequently evolves the result through VLM agents, ensuring both efficiency and physical validity. Within the evolution process, SceneMosaic exploits the locality of natural scenes and decomposes a scene into independent local units, allowing separate evolution within each unit before composing the global scene via Cartesian product. On SceneEval-100, SceneMosaic matches the strongest agentic baseline in semantic layout quality with a 24x speedup, substantially reduces physical violations, and receives the highest human ratings. Our code is publicly available at https://github.com/rxjfighting/SceneMosaic. --- ### Adaptive Cost-Sensitive Machine Learning for Autonomous Robot Navigation Failure Prediction: When Not All Errors Are Equal Authors: Rifa Ferzana Published: 2026-09-04 Categories: cs.RO, cs.AI arXiv: https://arxiv.org/abs/2609.05593v1 Autonomous robot navigation failures differ not only in categorical severity but also in the physical context in which they occur. A near-miss at low speed under reliable sensing is not equivalent to the same event during rapid motion, close obstacle approach or degraded perception. This paper reframes navigation failure prediction as consequence-sensitive forecasting. We first establish a fixed baseline in which training weights are modulated by categorical severity, then introduce an adaptive extension defining a state-dependent consequence function combining severity with normalised velocity, obstacle proximity and sensing uncertainty, together with a risk-sensitivity term that rises as conditions deteriorate. We evaluate on 2,000 simulated differential-drive episodes (~1,000,000 timesteps) using episode-level GroupKFold, with external validation on the UCI SCITOS G5 dataset. Fixed weighting raises Logistic Regression high-severity recall from 0.851 to 0.985 and reduces missed consequence cost from 1,940 to 313; the adaptive extension reaches 0.998 and 82. Under matched false-positive conditions, however, the discriminative advantage is modest (0.986 versus 0.984), so most of the gain reflects a more conservative operating point rather than better ranking. The effect is consistent across all five folds and stable across a threefold span of context coefficients. Because the primary simulation produced no collisions, we add a controlled extension in which 108 of 600 episodes terminate in contact: collision recall rises from 0.850 to 0.966 (fixed) and 0.984 (adaptive), with missed collision cost falling from 1,000 to 105, at false-positive rates of 0.413 and 0.799, respectively. Context-dependent consequence modelling thus provides a principled mechanism for allocating conservatism by physical risk. --- ### WolfSociety: Understanding Collective Risk from Harmful-Agent Scaling in Financial Agent Societies Authors: Lejun Zhang, Sarah Lu-Liang, Xin Jiang, Muning Wen, Weinan Zhang, Shangding Gu Published: 2026-09-04 Categories: physics.soc-ph, cs.AI, cs.SI arXiv: https://arxiv.org/abs/2609.05591v1 Safety evaluations typically focus on individual agents, but interacting agents can spread harmful information and influence the environment in which later decisions are made. We study how collective failure changes with harmful-agent fraction and society size in a controlled financial agent society, where agents communicate over a social network and trade in a shared market. In the primary financial scenario, collective failure requires broad harmful diffusion together with severe price dislocation or liquidity stress. Across all tested society sizes, failure remains rare at low harmful fractions but rises sharply over a narrow range. As society size grows from N=100 to N=2000, the harmful fraction associated with a 50% failure probability decreases from 4.7% to 2.2%, while the corresponding number of harmful agents increases from approximately 5 to 44. In contrast, when the number of harmful agents is held fixed, their impact becomes weaker as the society grows. Controlled interventions further show that broader network reach shifts the collapse boundary toward lower harmful fractions, whereas stronger conformity alone has little effect. To characterize these effects, we introduce Agent Society Dynamics, a finite-size framework for relating harmful-agent fraction, society size, and interaction structure to collective failure. Overall, our results reveal a nonlinear, size-dependent collapse transition in financial agent societies, showing that collective failure depends not only on the prevalence of harmful agents but also on the size and interaction structure of the surrounding society. Code is available at https://github.com/SAIL-Research-Lab/WolfSociety. --- ### GE-Act 2.0: Pretraining and Scaling a World-Action Model for Robotic Manipulation Authors: AgiBot Research Team, Renhang Liu, Wenzhi Zhao, Zhuo Yang, Liliang Chen, Pengfei Zhou, Shengcong Chen, Guanghui Ren, Youlun Peng, Rongjun Jin, Nan Wang, Sukai Wang et al. Published: 2026-09-04 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.05588v1 World-action models (WAM) predict future states to guide robot actions, enabling learning from both action-free video and action-labeled interaction. Most inherit pretrained video generators, leaving WAM pretraining and scaling underexplored. We introduce Genie Envisioner Act 2.0 (GE-Act 2.0), a world-action model whose trainable generative and action components are all initialized from scratch on manipulation data. It combines a control-oriented autoencoder (CoAE), a single-step visual planner (SVP), and an inverse dynamics model (IDM). CoAE retains action- and instruction-relevant information under aggressive compression, while SVP produces a complete future state in one differentiable pass, so visual planning and inverse dynamics can be pretrained separately on complementary data. The components are then jointly trained with knowledge-aligned selective optimization (KASO), which reduces mismatched supervision by selecting only predicted futures judged behaviorally compatible with the recorded action. We evaluate pretrained checkpoints directly, without per-task fine-tuning, on 100 tasks across 20 manipulation skill groups with held-out scenes, backgrounds, lighting, and object instances. Scaling co-training data from 300 to 30,000 hours raises success from 17.1% to 44.1% on G1-OP and from 13.4% to 31.1% on G2-90D; despite comprising less than 2% of the co-training data, G2-90D improves by 17.7 points, suggesting cross-embodiment transfer. Gains span 19/20 and 18/20 skill groups, and skill-specific coverage strongly correlates with zero-shot out-of-distribution (OOD) success (Pearson r=0.80; Spearman rho=0.85). Under the same protocol, the model grounds object, color, shape, and position references in at least 90% of trials and follows explicit instructions even when they conflict with an already-committed behavior or a conventional scene association. --- ### Agents' Overreliance on Unreliable Tools Authors: Hoyeol Yang, Woojung Song, Taewon Kim, Jonghyun Song, Seoyeon Park, Yohan Jo Published: 2026-09-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.05587v2 LLM agents use tools to access information and perform computations beyond their parametric knowledge. Existing tool-use benchmarks evaluate whether agents select and call the right tools, assuming that tool returns are reliable. However, tools can return plausible but incorrect outputs. We evaluate 14 models with three tools (web search, an LLM sub-agent, and a code executor), corrupting their returns to examine whether agents overrely on unreliable tools. Agents adopt corrupted returns at high rates, with mean adoption exceeding one third for every tool and reaching 68.0% for web search. Corrupted returns also frequently override correct answers that agents give without tools, and this reliance persists in more complex tasks and in tasks that combine multiple tools. Agents often make more tool calls under corrupted returns, and reasoning traces show them questioning a return and even stating the correct answer. Nevertheless, they still pass the corrupted content to users, rarely warning them of the conflict. We further test interventions spanning prompts, tool metadata, post-training, and activation steering. Verification prompts reduce adoption across all three tools while largely preserving accuracy with correct returns, and reliability labels provide partial mitigation in web search. The tested post-training methods offer limited gains, and the effect of activation steering depends on the model and intervention strength. Overall, our findings provide a basis for diagnosing tool overreliance and for developing agents that assess tool returns rather than assuming their reliability. --- ### An overview of 3D Vision-Language Models Authors: Márcus Lobo, Vitor Matias, Afonso Paiva, Jeová Farias, Tiago Novello, Moacir Ponti Published: 2026-09-04 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.05583v1 Vision-Language Models (VLMs) are reshaping computer vision by aligning visual and textual embeddings, allowing models to recognize visual concepts and reason about them using natural language. Traditional 3D deep-learning models, however, are typically trained for specific tasks, such as classification, segmentation, or detection, and do not naturally support cross-modal retrieval from their embedding spaces using text or images as queries. To address this issue, Contrastive Language-Image Pretraining (CLIP)-based methods align 3D embeddings with pretrained image and text representations, giving rise to 3D Vision-Language Models (3D VLMs) that support zero-shot classification, cross-modal retrieval, and open-vocabulary recognition of 3D shapes. This tutorial provides an overview of 3D VLMs, ranging from basic definitions of 3D representations and their encoding into embeddings to cross-modal contrastive alignment, modern multimodal frameworks, and 3D Vision-Large Language Models (3D VLLMs). We present the main definitions of contrastive learning for multimodal embedding alignment and highlight recent advances in language-guided 3D Gaussian splatting, 3D shape generation, and embodied AI for robotics. --- ### HB-PVI: A Hierarchical Bayesian Personalization and Value-of-Information Framework for Complex Activity Recognition Authors: Hammed A. Olayinka Published: 2026-09-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.05582v1 Personalization can improve activity-recognition performance, but participant-specific gains are heterogeneous, and every additional calibration label has an acquisition cost. This study presents HB-PVI, a hierarchical Bayesian personalization and value-of-information framework jointly modeling participant heterogeneity, the benefit and harm of four personalization mechanisms, and the economic value of an additional label, for the 47-participant MUSIC-CAR complex-activity cohort. A leakage-safe, leave-one-participant-out evaluation combines a sequential-Monte-Carlo participant-effect updater with a Student-$t$ hierarchical gain model and a one-step expected-value-of-sample-information (EVSI) stopping rule. Adapter personalization produced small positive mean F1 gains, growing from 0.00099 at one label to 0.00198 at ten, while adapter-plus-head and prototype-residual personalization were negative on average. Under the primary practical-benefit threshold ($Δ_{\min}=0.01$) and cost setting, one-step EVSI was zero at every decision state, so the policy purchased no labels and retained population inference for all 47 participants, matching always-stop exactly (region-of-practical-equivalence probability $=1$). Relative to fixed ten-shot adapter personalization, this reduced labeling by 100\% while keeping the posterior mean F1 loss at 0.00217 (95\% credible interval, 0.00048 to 0.00389), with posterior probability 0.9992 of remaining below the 0.005 tolerance. HB-PVI was utility-optimal in 199 of 216 cost-threshold settings and in every setting at or above the primary label cost. These results argue for a population-first deployment policy whenever personalization gains are small relative to labeling, computation, and harm costs, and show that value-of-information reasoning, not raw predictive accuracy, should drive personalization decisions in health-sensing applications. --- ### ViT3Flow: A Test-Time Training Transformer MeanFlow for Postoperative Radiograph Synthesis in Scoliosis Authors: Rui Tang, Sicheng Yang, Moxin Zhao, Hongqiu Wang, Guankun Wang, Lei Zhu, Hongliang Ren, Menglin Cong, Nan Meng Published: 2026-09-04 Categories: cs.CV, cs.AI, eess.IV arXiv: https://arxiv.org/abs/2609.05579v1 Predicting postoperative spinal morphology from preoperative radiographs could provide valuable support for scoliosis surgical planning, but remains challenging because surgical correction induces large spatial changes while anatomical structures must be faithfully retained. We formulate this problem as postoperative scoliosis radiograph synthesis and construct ScoliSurg, the first paired dataset for this task, comprising 632 preoperative--postoperative whole-spine radiograph pairs with structured morphology information. We further propose ViT$^{3}$Flow, a single-NFE conditional MeanFlow framework for efficient postoperative radiograph synthesis. ViT$^{3}$Flow models surgical correction as finite-interval generative transport and replaces conventional self-attention with test-time-training token mixers that perform sample-specific inner adaptation to the anatomy and deformity pattern of each case. In addition, a Spinal Morphology Extraction Agent extracts distributions of dominant-curve region and direction from the preoperative radiograph. These distributions guide Diagnosis-Routed Interval Cross-Attention (DRICA), which performs interval-dependent vertical, horizontal, joint, and global retrieval from a separate preoperative token stream. This design enables the evolving postoperative representation to incorporate spatially corresponding anatomical evidence throughout the transport process. Extensive experiments on ScoliSurg demonstrate that ViT$^{3}$Flow achieves the best performance among the compared methods in perceptual image quality, anatomical fidelity, and clinically relevant geometric accuracy, while requiring only a single network evaluation. These results highlight the potential of ViT$^{3}$Flow for efficient and anatomically faithful postoperative radiograph synthesis in scoliosis surgical planning. --- ### Planning and Scheduling Business Processes under Control-Flow Uncertainty: Extended Version Authors: Michel Kunkler, Stefanie Rinderle-Ma Published: 2026-09-04 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.05578v2 Scheduling activities in business processes can improve efficiency (e.g., reduce makespan), but is challenging because the exact sequence of activities required to complete a case is often uncertain due to decisions based on data that emerges during execution. Nevertheless, probabilistic information regarding such decisions can often be estimated or derived from historical execution logs, and can help anticipate which execution paths are likely to lead to successful completion. Planning with particular execution paths affects feasibility, i.e., the probability of successful completion, and the expected number of superfluous activities that are planned but never executed. We frame the problem as a chance-constrained optimization problem and present two formulations: A decomposed approach with two stages, a planning stage that minimizes the expected number of superfluous activities subject to a feasibility constraint, and a scheduling stage that minimizes the makespan over the planned activities; and an integrated approach that combines planning and scheduling into a single formulation. Evaluation on two real-world and one synthetic dataset shows that the integrated approach yields superior makespans but is intractable at scale, while the decomposed approach scales to large settings. --- ### EnvCraft: Synthesizing Executable Environments in Agentic RL for Claw-like Agent Authors: Yirong Zeng, Shen You, Jinhang Feng, Yufei Liu, Xiao Ding, Yutai Hou, Hao Cong, Yuxian Wang, Wu Ning, Wang Xu, Bibo Cai Published: 2026-09-04 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.05576v1 The paradigm of LLMs has rapidly shifted from passive language interfaces to autonomous Claw-like agents that execute long-horizon tasks across stateful workspaces. While Agentic Reinforcement Learning (Agentic RL) provides a promising path to optimize these agents, its scaling is heavily bottlenecked by the severe scarcity of interactive training environments. Existing synthetic environments are strictly limited to tool-calling endpoints, rendering them insufficient for accommodating the end-to-end real-world demands of claw-like agents. To bridge this gap, we introduce EnvCraft, an automated framework for synthesizing executable environments and scalable training data. Specifically, EnvCraft employs an environment synthesis engine to build sandbox-isolated workspaces, alongside a topology-aware data generation engine to produce coherent task trajectories. Overall, we synthesize 139 interactive environments comprising approximately 20K complex tasks for Agentic RL training. Experiments on Qwen3/3.5 models (8B-32B) show that our method yields gains of up to +11.9% on Claw-style benchmarks and +8.0% on general tool-use benchmarks, with concurrent reductions in inference token cost. The results confirm that synthesized executable environments provide robust and generalizable learning signals for training. --- ### Capsule Lens: Locating and Tracking Concept Geometry in Model Representations Authors: Yiming Tang, Harshvardhan Saini, Samyak Jha, Huaming Chen, Xufeng Duan, Dianbo Liu Published: 2026-09-04 Categories: cs.LG, cs.CV arXiv: https://arxiv.org/abs/2609.05575v1 Understanding how concepts are encoded in the internal representations of machine learning models is a central problem in mechanistic interpretability, essential both for the science of deep learning and for the trustworthy deployment of increasingly capable models. Existing approaches to interpret model representations mainly map representations onto more interpretable spaces and do not directly characterize how concepts occupy representation space; various hypotheses have been proposed, but often lack of rigorous validation and largely focus on static representations. In this work, we introduce Capsule Lens, a framework that matches the region a concept occupies with a simple, trackable geometric form, a capsule, defined by several interpretable parameters, fitted in closed form to each concept's geometry and validated on held-out samples. We apply Capsule Lens in two major settings: static and dynamic representations. On static representations, we demonstrate how to locate concept geometry across various models, and how the span and norm curves uncover important geometric characteristics. On dynamic representations, we present three case studies tracking representation drifts induced by distinct training settings, CLIP pretraining, RL post-training on visual question answering, and RL post-training on mathematical reasoning. These analyses reveal qualitatively different geometric dynamics, ranging from broad network-wide restructuring in CLIP pretraining to localized and concept-specific changes in RL post-training. Our results include findings aligned with existing literature as well as novel observations. We believe Capsule Lens stands as a promising tool for locating, analyzing, and tracking concept geometry in both static and dynamic representations. --- ### Multi-granularity Adaptive Hypergraph Representation Learning via Granular-ball Authors: Sen Zhao, Yifan Guan, Jinyuan Ni, Gaojie Xu, Zhang Xu, Xiaoyu Lian, Yi Liu, Yi Wang, Wei Wang Published: 2026-09-04 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.05574v1 Hypergraph representation learning aims to capture high-order information in graphs by constructing hyperedges that simultaneously connect multiple nodes. These hyperedges adapt to the graph's topological features, facilitating the extraction of high-order relationships at multiple granularities. Most prior work relies on predefined definitions to generate hyperedges, overlooking the diversity in graph topological structures and the multi-granularity characteristics of hyperedges. As a result, this limits their ability to effectively and adaptively discover high-order relationships and efficiently process complex structural information. To address this limitation, we propose a novel framework called \underline{M}ulti-\underline{G}ranularity \underline{H}ypergraph \underline{R}epresentation \underline{L}earning (MGHRL). MGHRL introduces an Adaptive Granular Hypergraph Generation strategy, which generates hyperedges at multiple levels of granularity through the adaptive splitting of granular-ball, effectively capturing high-order relationships based on the graph's topological structure. Additionally, we propose a Multi-Granularity Hypergraph Network with multiple sub-networks, capturing features from hyperedges at different granularities and integrating them via hierarchical reversible connections. Experimental results show that MGHRL significantly outperforms baseline models on benchmark datasets. ---