# Research Papers — 2026-09 · bundle page 87 of 266 > Full-text markdown of papers #4301–4350 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/87 [← Previous bundle](https://aipod.dev/papers/archive/2026-09/page/86.md) [Next bundle →](https://aipod.dev/papers/archive/2026-09/page/88.md) ### Screen Before You Serve: Simulation for Production Customer Experience AI Agents at 140M Scale Authors: Edesio Alcoba, Kevin Rossell, Aman Gupta, Shao Tang, Jiwoo Hong, Pabel Carrillo-Mendoza, Wanderson Conceição Ferreira, Alvaro Tedeschi, Zayd Simjee, Shreya Rajpal, Bruno Finardi Hime, Christian Sousa et al. Published: 2026-09-24 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.30137v1 Customer experience (CX) agents use tools and large language models to address customer requests and guide conversational interactions with an organization's products. Improving these agents, especially in regulated industries, is difficult: they must detect intent, follow complex operational policies and use tools reliably. Manual end-to-end testing offers limited coverage, while live experiments expose customers to failures that can erode trust. We present a hypothesis-driven simulation workflow for screening candidate CX agents before deployment. Synthetic customers react to agent responses and simulated tool outputs enable multi-step agentic workflows without invoking production backends. We use the Snowglobe simulator on Nubank's Card Delivery agent and its expanded successor, Card Management - Nubank's highest-volume chat-support agent in Brazil. Across 4 deployed versions, simulated and production version-level binary evaluator scores show high correlation. Simulation-guided iteration increased transactional net promoter score (tNPS) by 36.69 points in a live A/B test. We also screened open-weight configurations in over 16,000 simulated conversations. In a subsequent live A/B test, the selected model increased self-service rate (SSR) by 8.82 percentage points to the highest level observed at Nubank, with no statistically significant change in tNPS. Simulation made broad exploration of models, reasoning settings, and prompts feasible without customer exposure, enabling production improvements that would have been impractical to pursue through live experimentation alone. --- ### Multimodal Thinking with Renderable Programs Authors: Sunli Chen, Ding Zhong, Ziqiao Ma, Jiaxin Liu, Zeyuan Yang, Hao Zhang, Lie Lu, Joyce Chai, Chuang Gan Published: 2026-09-24 Categories: cs.CV, cs.CL arXiv: https://arxiv.org/abs/2609.30130v1 Current vision-language models (VLMs) excel at visual content understanding and text-based reasoning, yet their structure limits the advancement of incorporating images into the reasoning chain. Though Omnimodal models have made efforts in unifying text and image generation, they focus on visual tasks in the open-domain, lacking tractability due to rasterized or latent representations of images. We introduce SVGLM, a framework that uses scalable vector graphics (SVG) primitives to connect text and image in reasoning tasks. We exploit the duality of SVG as both image description and text instructions, yielding a more compact, interpretable solution to equip general VLMs with the capability of generating images within the reasoning process. We provide a large curated dataset of SVG-based image editing dataset, as well as the paradigm to tune open-source VLMs. Experiments on a mathematical reasoning benchmark demonstrate that SVGLM achieves strong SVG generation power as well as think-with-image intelligence. Our results highlight SVG as a suitable medium for building more robust digital domain agents, bridging the gap between text-based thinking and pixel-based images. --- ### HEXIS: Compiling Agent Skills into Extended Finite State Machines Authors: Minghao LI Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30123v2 Agent skills provide instructions and reusable knowledge, yet agents have to frequently infer which action to take next. This skill's execution paradigm combines task reasoning with control decisions, which could cause the agent to act improperly or omit necessary steps. To improve agent compliance for skills, we introduce HEXIS, which compiles agent skills into extended finite state machines (FSM) that separate knowledge from control flow. Skill knowledge is incorporated into local instructions that guide reasoning and generation within states. The machine records execution progress and intermediate results, while explicit transition conditions determine subsequent operations. Our incremental compiler first maps skill clauses and tool interfaces to state operations, local instructions, data bindings, and transitions. It then aligns development traces with existing states to identify missing operations and dependencies. These are incorporated by adding or reusing states and refining their connections. Updates are accepted only after static checks and replay of the current and all previously accepted traces. Across four benchmarks and four executors, HEXIS improves success over Skill + ReAct by 16.2 percentage points on average. Qwen3.8-27B reduces execution tokens by 38.4-88.9% across benchmarks. --- ### HEXIS: Compiling Skills into Extended Finite State Machines Authors: Minghao LI Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30123v1 Agent skills provide reusable knowledge and instructions, yet agents must repeatedly infer how to apply them and which operation should follow. This couples task reasoning with control decisions, allowing prescribed steps to be omitted or applied incorrectly. We introduce HEXIS, which compiles agent skills into extended finite state machines that separate knowledge from control flow. Skill knowledge is incorporated into local instructions that guide reasoning and generation within states. The machine records execution progress and intermediate results, while explicit transition conditions determine subsequent operations. Our incremental compiler first maps skill clauses and tool interfaces to state operations, local instructions, data bindings, and transitions. It then aligns development traces with existing states to identify missing operations and dependencies. These are incorporated by adding or reusing states and refining their connections. Updates are accepted only after static checks and replay of the current and all previously accepted traces. Across four benchmarks and four executors, HEXIS improves success over Skill + ReAct by 16.1 percentage points on average. Qwen3.8-27B reduces execution tokens by 38.4-88.9% across benchmarks. --- ### What, When, and How: Audio Description as Constrained Global Optimization Authors: Igor Sterner, Mirella Lapata, Alex Lascarides, Frank Keller Published: 2026-09-24 Categories: cs.CL, cs.CV arXiv: https://arxiv.org/abs/2609.30121v1 Audio Description (AD) makes movies accessible to blind and visually impaired audiences by narrating visual information in gaps between dialogue. Existing automatic AD systems largely treat generation as a local video-to-text problem, assuming that the content to describe and its temporal location are already provided. Realistic AD instead requires coupled decisions about what visual information is narratively important, when it can be spoken without interfering with dialogue, and how it should be formulated to fit within the available time. We formalize AD generation as a constrained optimization problem over these three decisions. Our hybrid system uses large language models to propose and ground visual elements, estimate their salience to the narrative, and generate compressed realizations. A mixed-integer linear program then jointly selects and schedules descriptions across a scene subject to temporal constraints. When evaluated on REFRAMED, a benchmark for realistic AD of movies, our approach makes better decisions than prompted LLMs about what to describe and when to describe it, establishing a new SOTA on narrative QA and temporally grounded metrics. Ablations show that explicit temporal constraints drive gains in placement, while salience estimation controls how much narratively useful content is retained. Improvements are concentrated on temporal and narrative measures rather than n-gram overlap, although a significant gap to professional describers remains. --- ### Orbital Error Dynamics: Self-Organized Criticality, Ephemeral Parameter Resonance, and Non-Linear Biological Ontologies in Zero-Storage Neural Synthesis Authors: Volkan Dağlı, Zerrin Dağlı, Dağhan Dağlı Published: 2026-09-24 Categories: cs.NE, cs.LG, nlin.CD arXiv: https://arxiv.org/abs/2609.30115v1 Modern deep neural networks treat parameters as static floating-point matrices stored in physical memory, incurring Von Neumann memory bottlenecks and representation collapse. We formulate Orbital Error Dynamics (OED), an analytical framework wherein synaptic weights are not stored masses (O(W)), but transient topological resonances (O(1)) derived procedurally from the complex quadratic polynomial map z_{n+1} = z_n^2 + c. We introduce the Bent Sine Wave Hypothesis, demonstrating that non-equilibrium living systems emerge when harmonic waves curl inward through environmental drag toward the cardioid cusp (c = 1/4). We define the Observer Horizon Geometry in parameter space, identifying interior resonance shoulder loci X_upper = (0.25, +0.18) and X_lower = (0.25, -0.18) between the fixed-point basin and the true boundary at c = 0.25 +/- 0.50i. To escape non-convex stagnation without loss zeroing, we introduce a heavy-tailed Biomimetic Perturbed Jump Operator (Omega_tunneling) inspired by mammalian fertilization zinc sparks. We further couple an enteric-cranial Dual-Brain architecture shielded by adaptive CD4+ regulatory immune gating (M_CD4), and project the 4-nucleotide genetic basis (A, T, C, G) across quadrants in C. Multi-seed empirical validation on the Two-Moons manifold (5 seeds, 80/20 train/test split, 32x32 grid, zero test-time updates, zero label leakage) demonstrates that procedural parameterization from a 24-byte coordinate seed achieves 77.67% +/- 5.35% clean test accuracy (within an 8.00-point paired difference of an unconstrained gradient baseline at 85.67% +/- 5.35%, 95% CI: (-1.07%, 17.07%)) and 71.33% +/- 3.80% under distribution shift (N(1.2, 0.4)), alongside conceptual equivalence with an analog optical co-processor. --- ### Smartphone-Based Method for Automated Speed Enforcement Authors: Keya Li, Jahnavi Malagavalli, Lamha Goel, Tong Wang, Kara M. Kockelman Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30107v1 Smartphone cameras and computer vision (CV) hold significant promise in assisting public agencies with enforcing traffic laws and enhancing road safety. This work designs and tests a smartphone-based method for automated speed estimation and vehicle identification (license plate, make/model, and color recognition) via an automated pipeline to assist enforcement agencies in reliably identifying speeders. The CV code accurately recognizes nearly half (46%) of the license plates' text on 1,800 images from a Brazil open-source dataset, called UFPR-ALPR. Code tests on daytime recordings from hand-held smartphone videos (n = 73) and roadside cameras (n = 42) in Austin, Texas yield 60.8% accuracy for color detection (among all possible RGB color categories), 48.6% on vehicle make/manufacturer identification, and 16.89% on vehicle make and model identification. Prediction accuracy for speed estimation (within a 20% range), vehicle make (within the top 3 predictions), and license plate recognition (within the top 10 predictions) are 16.3%, 16.9%, and 29.7%, respectively. This paper also illuminates the legal, technological, and practical aspects of using smartphones for enforcement, including the potential use of recordings for enforcement purposes, emphasizing the need to transform the potential of smartphone-based CV technologies into practical tools for vital information on traffic violations. --- ### On the SoS Certifiability of Log-Concave Distributions Authors: Aleksandr Storozhenko Published: 2026-09-24 Categories: cs.LG, cs.CC, math.PR arXiv: https://arxiv.org/abs/2609.30105v1 For an arbitrary isotropic log-concave distribution $P$ on $\mathbb{R}^d$, we prove that the polynomial $(Cm)^m\|v\|_2^m - \mathbb{E}_{X\sim P}\langle X,v\rangle^m$ is a sum of squares for every even $m\ge2$, where $C>0$ is a universal constant. This removes the dependence on the Poincaré constant in the theorem of Kothari and Steinhardt (arXiv:1711.07465), recovering the optimal moment bounds for log-concave distributions. As an immediate corollary, we obtain computationally efficient algorithms with dimension-free error guarantees for a wide range of high-dimensional statistical estimation problems. Our proof uses stochastic localization to decompose $P$ as an average of random strongly log-concave measures, whose centered moments admit the subgaussian certificates of Diakonikolas, Hopkins, Pensia, and Tiegel (STOC 2025; arXiv:2410.21194). With a covariance-adapted choice of localization, we show that a fourth-moment certificate derived from Letwin's variance inequality for quadratic forms (arXiv:2607.24164) suffices to control this averaging at every even degree. --- ### MQSS-Selector: RL-Guided Pass Selection for an MLIR Compilation Pipeline Authors: Andre Youssefi, Ercüment Kaya, Minh Chung, Jorge Echavarria, Laura B. Schulz, Martin Schulz Published: 2026-09-24 Categories: quant-ph, cs.DC, cs.LG, cs.PL arXiv: https://arxiv.org/abs/2609.30104v1 High Performance Computing (HPC) and Quantum Computing (QC) systems are increasingly converging towards unified High Performance Computing-Quantum Computing (HPCQC) infrastructures, driven by a growing need to bridge classical and quantum workflows, which affects all levels of the system stack, from the hardware to compilers and runtimes, all the way to applications. However, today's QC devices are still in the Noisy Intermediate-Scale Quantum (NISQ) era, are error-prone and resource-limited, and therefore require specialized optimizations and topology mappings to achieve sufficient fidelity. This places special emphasis on proper compilation and optimization within the overall quantum software stack. Many existing stacks remain fragmented, with separate components responsible for device selection, compiler-pass optimization, and job queue scheduling. This paper proposes a unified, learning-based selector that integrates these disparate stages into a cohesive framework. Our proposed selector scheme leverages reinforcement learning and deep learning models that can be extended to simultaneously optimize multiple objectives -- such as fidelity, compilation time, and scheduling latency -- while dynamically adapting to circuit characteristics and device conditions. --- ### Activation-Flexible ANN-to-SNN Conversion with Finite-State Markov Neurons Authors: Ruiyu Jia, Zhuo-Cheng Xiao Published: 2026-09-24 Categories: cs.NE arXiv: https://arxiv.org/abs/2609.30102v1 Most ANN-to-SNN conversion methods rely on a specific correspondence between the source activation and the spiking neuron dynamics. We propose a finite-state continuous-time Markov chain (CTMC) neuron framework whose stationary spike flux can approximate every continuous nonnegative monotone activation function on a compact interval. For a generalized CTMC family with affine input-dependent transitions, we prove uniform approximation to arbitrary accuracy over this function class and derive an explicit approximation error bound. In practice, two- and three-state CTMCs fit ReLU, sigmoid, softplus, and clipped ReLU on the evaluated input ranges, and we evaluate corresponding MLP conversions for each activation with layerwise rate scaling. Moderate clipping improves the conversion cost-accuracy tradeoff on the MNIST MLP and reduces SynOps by 27% on VGG-11/MNIST at matched ANN-SNN accuracy gap criteria, whereas the trend reverses on VGG-11/CIFAR-10. Mean-field and layerwise diagnostics indicate that finite-window sampling and terminal-layer mismatch are the main residual errors. Overall, our results establish finite-state CTMC neurons as a theoretically grounded framework for activation-flexible ANN-to-SNN conversion beyond fixed activation-neuron correspondences. --- ### R-DEIM Net: An Efficient Rationale-Augmented Dual-Expert Interaction Model for Paraphrase Detection Authors: Pushp, Vaibhav Prajapati, Himangshu Sarma Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.30100v1 Recent advances in paraphrase detection reveal a fundamental trade-off: large language models achieve high accuracy but require high computation, while efficient Siamese-BERT variants offer practical scalability with reduced transparency in rationale generation. We present R-DEIM Net, a 76M-parameter dual-expert architecture exploring whether moderate-scale models can achieve competitive accuracy on paraphrase detection while enabling human-readable rationale generation. The architecture combines two specialized components: an Interaction Expert that captures token-level similarity patterns through multi-scale 2D convolutions and attention head allowing variable input length, and a Reasoning Expert that uses a Flan-T5-small decoder to generate rationales as auxiliary supervision. Rather than re-encoding generated text, we extract and pool decoder hidden states as complementary features for classification. On the Quora Question Pairs dataset, R-DEIM Net achieves 90.07\% accuracy and 90.16\% F1-score via 10-fold cross-validation. This represents competitive performance with strong transformer-based baselines (e.g., MFAE BERT: 90.54\% accuracy) and recent large language model based approaches (LLaMA-70B) while using a substantially smaller parameter budget. The model generates rationales alongside predictions, providing potential for auxiliary human-readable descriptions. --- ### Accelerating Video Diffusion via Training-Free Trajectory Routing Authors: Mustafa Munir, Huy Vu, Shreyas Misra, Rohit Jena, Sajad Norouzi, Ali Taghibakhshi, Anis Ahmad, Anjul Patney, Pavlo Molchanov, Nima Tajbakhsh Published: 2026-09-24 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2609.30096v1 Video diffusion is computationally expensive, as it requires executing a large model across many denoising steps. Even with step-distillation, inference remains expensive because every distilled step still requires a costly model evaluation. We present TRACK: TRajectory-Aware Capacity routing via top-K selection, a heterogeneous denoising strategy that switches between compatible large and small models at selected steps, reducing the average cost per denoising evaluation. The switching steps are determined using a calibration process. TRACK first rolls out a reference trajectory with the large model. Then at each step, the small model's prediction is also collected and compared against the large model's prediction to obtain a relative disagreement score. Both models receive the same latent, timestep, conditioning, and guidance inputs. Aggregating this signal over a calibration set produces a disagreement score map across diffusion steps, which determines a switching policy for an efficient inference process: quality-sensitive steps keep using the large model, while steps with low disagreement scores are routed to the small model. Inference executes only the selected model at each step, requiring no retraining, architecture or scheduler changes, or online dual-model evaluation. Across Wan 2.1, Cosmos 3, TurboDiffusion, and FastVideo, TRACK yields $1.95\times$, $2.04\times$-$2.73\times$, $2.69\times$, and $2.17\times$ speedups, respectively, with comparable aggregate quality and high diversity retention. TRACK thereby establishes automated, training-free model switching as a practical acceleration paradigm for video diffusion. --- ### PrivDrift: Auditing User-Secret Leakage Under Topic Drift in Active LLM Conversations Authors: Luciano Maldonado Published: 2026-09-24 Categories: cs.AI, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2609.30094v2 Large language models increasingly operate as persistent assistants in user-facing, shared-session, and tool-augmented settings. When users disclose sensitive information during an active conversation, that information may remain behaviorally recoverable through later prompts even after the dialogue shifts to unrelated topics. We introduce PrivDrift, a benchmark for auditing whether user-disclosed secrets remain recoverable after conversational topic drift and persuasion-based probing. PrivDrift contains 1,000 controlled multi-turn dialogues with seeded secrets, content-dense drift turns, and standardized extraction probes. Across three LLMs with extended context windows, dialogue-level hybrid leakage remains substantial, ranging from 38.7% to 54.6%, and varies strongly by model, secret type, and persuasion intensity. Within the tested drift window, additional topic drift does not reliably reduce leakage, suggesting that privacy risk in active LLM contexts should be evaluated as a persistent behavioral failure mode rather than only as training-data memorization or immediate jailbreak behavior. --- ### PrivDrift: Auditing User-Secret Leakage Under Topic Drift in Active LLM Conversations Authors: Luciano Maldonado Published: 2026-09-24 Categories: cs.AI, cs.CL, cs.CR arXiv: https://arxiv.org/abs/2609.30094v1 Large language models increasingly operate as persistent assistants in user-facing, shared-session, and tool-augmented settings. When users disclose sensitive information during an active conversation, that information may remain behaviorally recoverable through later prompts even after the dialogue shifts to unrelated topics. We introduce \textbf{PrivDrift}, a benchmark for auditing whether user-disclosed secrets remain recoverable after conversational topic drift and persuasion-based probing. PrivDrift contains 1{,}000 controlled multi-turn dialogues with seeded secrets, content-dense drift turns, and standardized extraction probes. Across three LLMs with extended context windows, dialogue-level hybrid leakage remains substantial, ranging from 38.7\% to 54.6\%, and varies strongly by model, secret type, and persuasion intensity. Within the tested drift window, additional topic drift does not reliably reduce leakage, suggesting that privacy risk in active LLM contexts should be evaluated as a persistent behavioral failure mode rather than only as training-data memorization or immediate jailbreak behavior. --- ### Self-Adaptive VLA for Robust Robot Deployment Authors: Hongxin Zhang, Chunru Lin, Tsun-Hsuan Wang, Zhenjia Xu, Chuang Gan Published: 2026-09-24 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.30092v1 While Vision-Language-Action (VLA) models demonstrate impressive capabilities in robotic manipulation, their memoryless nature renders them brittle to test-time environment shifts, particularly hardware shifts caused by wear or imperfect calibration. Enabling these models to self-adapt during deployment without requiring continuous on-site recalibration remains a critical bottleneck for real-world scalability. In this work, we introduce Self-Adaptive VLA, a novel post-training recipe that enables the policy to iteratively adapt to deployment-time hardware shifts leveraging its own rollouts as context. To do so, we first collect policy rollouts under deliberately injected hardware shifts. We then transform the base policy's training data into shift-conditioned expert demonstrations by pre-compensating the expert actions for these known shifts. Next, we introduce a lightweight, plug-in context encoder that compresses the context, including visual observation, proprioception, and actions in the shifted environment, into a latent context token. This token modulates the policy through adaptive layer normalization (AdaLN). Furthermore, we find that context tokens can be ensembled, allowing the policy to iteratively self-correct and mitigate failures step by step. Extensive experiments across four precision-critical bi-manual and dexterous manipulation tasks show that Self-Adaptive VLA recovers over 80% of the base policy's performance under hardware shifts, such as actuation bias and joint encoder offsets. Moreover, Self-Adaptive VLA enables more robust deployment to new workstations compared to the base policy. Our approach provides a pathway for robust large-scale real-world robot deployments and easier maintenance. See videos at https://icefoxzhx.github.io/self-adaptive-vla. --- ### AT-SKM-Net: An Accelerated Trainable Sampling Kaczmarz-Motzkin Framework for Linear Hard-Constraint Feasibility on Dynamic Graphs Authors: Xiaochen Zhang, Haoyu Zhu, Yao Zhang, Qingchun Hou Published: 2026-09-24 Categories: cs.LG, cs.AI, math.OC arXiv: https://arxiv.org/abs/2609.30088v1 Graph-structured optimization with linear constraints is fundamental to critical infrastructure but faces scalability limits due to massive strict hard constraints and high dimensionality. While recent projection-based methods such as Trainable Sampling Kaczmarz-Motzkin Net (T-SKM-Net) guarantee feasibility, they face high computational costs in dynamic environments by processing the entire constraint set and requiring expensive matrix factorizations. To bridge this gap, we propose the Accelerated Trainable-SKM (AT-SKM) Net framework. To concentrate computation on the active constraints and eliminate redundant calculations, we introduce a hybrid sampling strategy guided by a topology-aware heterogeneous GNN model. To efficiently handle topological shifts in graph-based constraints, we employ a Cholesky Update mechanism that theoretically reduces the equality projection complexity from O(N^3) to O(N^2) under low-rank perturbations. Experiments on random geometric graphs, N-1 Security-Constrained DC-OPF, and minimum-cost gas transport problem demonstrate that AT-SKM reduces iteration counts by up to 85% and achieves 2.95x-7.29x SKM layer speedups, while maintaining zero constraint violations. --- ### Return or Revise? Learning When Revision Helps Retrieval-Augmented QA Authors: Nicholas Kashani Motlagh, Tim Anderson, Jeremy Gwinnup, Grant Erdmann Published: 2026-09-24 Categories: cs.CL, cs.IR, cs.LG arXiv: https://arxiv.org/abs/2609.30087v1 We consider the decision of whether to return an existing draft answer or revise it using retrieved evidence, as in answer-revision systems. Draft confidence estimates whether the current answer is correct, but the decision requires estimating the effect of a specified revision. For offline training and evaluation, we grade both the returned draft and its candidate revision under the same correctness judge, which makes repair, harm, and the gap to an oracle observable. We call this paired effect its recoverability, and we train policies to predict it before revision. On 25,870 held-out open-domain questions across three revision setups, a scorer trained on the paired outcome has greater area under the accuracy--revision-rate curve than a matched draft-correctness scorer in all nine Llama setup--seed fits, and gains 0.23--0.68 accuracy points on average at development-selected thresholds, a difference significant across training runs only for dense retrieval. The resulting policy improves on always revising and on average closes more than a third of the oracle gap, although it still applies 38--46% of the harmful revisions. When a draft-free standard-RAG answer is also available, however, choosing between the draft and that answer is stronger by about two points for Llama and four for OLMo, and adding candidate revision as a third option yields no significant gain. Recoverability describes one revision; its value as an available action also depends on the alternatives. --- ### Residual Correlation as a Diagnostic for Joint-Uncertainty Gains from GP Coregionalisation Authors: Fangqin Zhou, Joaquin Vanschoren Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.30085v1 In multi-target regression, correlated targets are often coupled through multi-output Gaussian processes with an intrinsic model of coregionalisation (GP-ICM), assuming that sharing statistical strength improves overall performance. In practice, the benefits are inconsistent. Across the settings studied, we find that the main benefit of coregionalisation is joint uncertainty quantification rather than point prediction. Raw target correlation does not predict when coupling helps; in the separable GP-ICM settings studied here, residual correlation, the cross-target dependence left unexplained by independent per-target predictors, is the strongest predictor of joint-uncertainty gains. We introduce a lightweight diagnostic, $D_{\rm logdet}=-\frac{1}{2}\log\det R_{\rm res}$, which represents the idealised joint negative log-likelihood (NLL) gain from modelling a full rather than diagonal residual covariance and is computable from independent GPs alone. Across a controlled synthetic study, 16 multi-target benchmarks, and frozen transformer and convolutional neural network representations for keypoint regression, point prediction remains largely unchanged ($ΔR^2\approx 0$). In contrast, $D_{\rm logdet}$ strongly predicts observed ICM NLL improvements ($ρ_s=-0.83$, $p<0.001$), outperforming heuristics such as the feature-to-sample ratio. We also propose Residual-ICM, which preserves independent marginal variances while adding residual-correlation structure to the joint covariance. Residual-ICM achieves the best average joint NLL among the compared methods, while the diagnostic indicates when covariance coupling is likely to be useful. The diagnostic is specific to global Gaussian residual dependence, the structure captured by separable coregionalisation. --- ### Can Frozen Hyperspherical Features Guide the Selection of Pseudo Masks? Authors: Xinge Guo, Fengyang Xiao, Dingming Zhang, Yuhan Chen, Rihan Zhang, Xingjian Li, Tianyang Wang, Chunming He, Sina Farsiu Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30080v1 Foundation segmenters such as SAM return several plausible masks for an unlabeled image, and a student trained on the wrong one inherits its errors. Choosing among them means querying a second large model or fitting a quality head to annotated masks. We show that a candidate can be judged by what it does to a frozen self-supervised backbone's features. Normalized DINOv2 patch features lie on a hypersphere, and a candidate mask splits that sphere in two. Based on this reading, we introduce SphereTrust, which scores each candidate by three properties of the split, the angular contrast between the two sides, the coverage of the foreground's appearance modes, and contact with the image frame, one for each of three common ways a mask fails, and ranks a pool in 0.55 s per image from the frozen features alone. On eight SAM and SAM3 candidate pools spanning camouflaged, salient, and dichotomous segmentation and camouflage under low light, SphereTrust exceeds the strongest evaluated external baseline on six pools by 1.7 to 9.3 percentage points in mean selected Dice. These comparisons include published selection rules and explicitly labeled adaptations of DSS and UCOD-MKD. On the two prompted camouflage pools, its mean selected Dice is within 0.1 percentage points of the candidate-derived DSS adaptation, with a lower catastrophic-error rate. Which cue carries the signal depends on the candidate pool. The same sphere also supports training. The leading candidates enter as a candidate set with their scores as priors, prototypes reorder them, and a cross-fitted second round completes the labels, raising weighted F by 4.5, 2.3, and 5.5 points over fixed-label training on the three MLLM anchor pools, with students competitive with published unsupervised methods on nineteen test sets. --- ### Reachability-Based Formal Verification of Graph Neural Networks with Node and Edge Features Authors: Anne M. Tumlin, Ben Wooding, Zhenxuan Shao, Diego Manzanas Lopez, Tyler Derr, Taylor T. Johnson Published: 2026-09-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.30079v1 Graph neural networks (GNNs) have become a prominent approach for developing fast, topology-aware surrogates in electric power systems, supporting tasks such as power flow (PF) analysis, optimal power flow (OPF) estimation, and cascading failure analysis (CFA). Despite this growing use, formally verifying GNN-based models remains challenging, with existing methods limited in scope. We extend the neural network verification (NNV) framework to graph-structured inputs through GraphStar sets, a generalization of Star sets that captures uncertainty over both node and edge features. This extension enables the propagation of linear message-passing operations and the sound approximation of ReLU nonlinearities for GNN architectures, including graph convolutional network (GCN) and graph isomorphism network with edge features (GINE) layers. We evaluate GNNV across three power system tasks, PF, OPF, and CFA, on the IEEE-24, IEEE-39, and IEEE-118 test cases, as well as two standard graph classification benchmarks, ENZYMES and PROTEINS. Our results show that GNNV provides tighter robustness guarantees than CORA on graph classification models with ReLU-based activations and, for the first time, delivers edge-aware robustness guarantees for GINE-based PF and OPF models under joint node and edge perturbations. --- ### Nuclear Norm-Regularized Bayesian Matrix Completion Authors: Calvin Tolbert Published: 2026-09-24 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2609.30078v1 Matrix completion, the problem of estimating missing entries in a matrix from noisily observed ones, underlies a diverse array of problems such as recommender systems and counterfactual outcome estimation in panel data. Many algorithms address the problem using regularized least squares, often with the nuclear norm as a regularizer, but this method yields a point estimate with no built-in uncertainty quantification. A Bayesian formulation is a natural alternative, and if the noise variance is known, the nuclear norm-based prior yields a log-concave posterior. Unfortunately, in practice, the noise variance will not be known a priori, so for a fully Bayesian approach, a prior must be imposed on it. We give the first sampler for this model with an explicit non-asymptotic guarantee: polynomial in the matrix dimensions and in the reciprocal of the target accuracy. Our technique is to discretize the distribution of the noise precision onto a grid and build a categorical posterior via thermodynamic integration. This extension is not specific to matrix completion and may be useful in other non-log-concave sampling problems where the non-log-concavity is restricted to a single variable and the joint distribution of the remaining variables is nonsmooth. Our contribution is a feasibility result: we show that a polynomial-time Bayesian sampler for this model exists at all, and the resulting complexity, while polynomial, is not intended as a deployable algorithm at current problem scales. --- ### A Native-Reference Phone-Class Geometry for Second-Language Pronunciation Analysis Authors: Tina Raissi, Nhan Phan, Chenxiao Wang, Mikko Kurimo Published: 2026-09-24 Categories: cs.CL, cs.SD arXiv: https://arxiv.org/abs/2609.30075v1 Automatic speaking assessment systems can provide holistic proficiency scores, but often lack interpretable measures that characterize pronunciation quality. We propose a native-reference phone-class geometry for measuring second language (L2) pronunciation deviation without requiring pronunciation labels, read-aloud prompts, or matched recordings of the same text from native and L2 speakers. Given a native speech corpus, we average frame-level self-supervised representations for each context-dependent phone-class and use singular value decomposition (SVD) to derive a compact native-reference coordinate system. For each L2 utterance, we compute the corresponding averages and project them into the native-reference space. We then demonstrate that the distances between L2 and native-reference coordinates for matched phone-classes show consistent negative correlations with holistic speaking proficiency on the Dev subset of the Speak and Improve Corpus 2025 (Spearman's $ρ\!=\!-0.53$) and with pronunciation quality on the learner subset of the English Read by Japanese Students dataset ($ρ\!=\!-0.34$). These findings suggest that the proposed geometry captures acoustic-phonetic information relevant for proficiency rating while remaining applicable to spontaneous L2 speech without matched native recordings. --- ### How Reproducible Are Evaluation Conclusions? A Self-Audit of LLM-Inferred Prompt Structure Authors: Dipankar Sarkar Published: 2026-09-24 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.30074v1 Evaluations of LLM systems routinely average over small prompt sets and report models as a ranked table. We ask how much confidence such a table deserves, using LLM-based prompt-structure inference as the case study: eight open model variants across five families and 8B to 675B parameters, caching disabled, 293 raw intermediate representations persisted. The measured phenomenon is unstable to begin with. Identical calls do not reliably recover identical structure, with mean node-set Jaccard from 0.39 to 0.96 and 72% of prompt-model cells never node-set-perfect. Auditing the evaluation weakens its conclusions further, and this is our main contribution. Under a joint cluster bootstrap over prompts, only the bottom of the ranking is firm: the two least reproducible models hold rank in 99% and 86% of replicates, the middle four in 27% to 48%, and the top two in 68% each, so the table identifies the worst model reliably but does not reliably identify the best. Two equally defensible rules for merging repeated campaigns change four of eight rows and move the study-wide headline by 7 percentage points. Checking the inferred structure against ground-truth annotations shows reproducibility cannot be read as accuracy. And four of the eight endpoints were withdrawn within ten weeks of measurement, so the study as specified can no longer be run. Small-sample LLM evaluations can therefore look far more definitive than their evidence supports. We recommend reporting rank stability, per-cell provenance, executed sensitivity comparisons, raw per-run outputs, and a measurement date alongside any ranking. --- ### Scoring Both Directions: LLMs realize the MRS they cannot reliably parse Authors: Soham Dan Published: 2026-09-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.30071v1 The English Resource Grammar (ERG) is a hand-written computational grammar of English. Given a sentence, its processor, ACE, produces a formal meaning representation called Minimal Recursion Semantics (MRS): a graph of the sentence's predicates and their arguments. The grammar is bidirectional and can also turn an MRS back into an English sentence. \citet{hajdik2019} used the ERG's treebank to build a benchmark for that generation task, MRS to text, and trained sequence-to-sequence models to solve it. The parsing task, text to MRS, can be tested on the same sentences. We reconstruct their 10K-sentence test split, and score two large language models, Claude Sonnet~4.5 and Claude Opus~5, in both directions against their trained systems and against ACE, with no task-specific training. Given an MRS and three examples, Opus writes the sentence at 76.3 BLEU, ten points above their system trained on 72k pairs (66.1 BLEU), and comparable to their system trained on a million extra pairs (77.2 BLEU). Sonnet scores 65.7 BLEU, and letting it choose among ACE's own candidate sentences lifts it to 69.6, while a pooled judge that keeps Opus's own sentence among the candidates adds 0.6 points (77.0 BLEU). In the parsing direction, however, the models fall far behind ACE: asked for the MRS of the same sentences, they reach 57.2 (Sonnet) and 65.5 (Opus) F$_1$ on the graph's predicates and arguments against 91.0 for ACE, and exact-match the gold on about 1\% of sentences. We characterize the failure modes for the parsing tasks, and conclude that a generation score alone does not show that models understand formal semantic representations. --- ### Self-Play Pretraining with Zero Data Authors: Aditya Cowsik, Kfir Dolev, Michael Y. Li, G. Bruno De Luca, Nourya Cohen, Noah D. Goodman, Yoav Levine Published: 2026-09-24 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2609.30063v1 Advances in language modeling have been driven by scaling pretraining on ever more data. Yet, the training data is still largely curated on the model's behalf. A more general approach to pretraining would let the model learn to generate the data most useful for its own improvement. This would provide an effectively unbounded source of training data, limited by compute rather than human knowledge. We introduce Self-Play Pretraining with Zero Data, an initial proof-of-concept towards realizing this vision. Our procedure casts synthetic data generation as a search over the space of all computable structure, taking inspiration from Solomonoff induction. Starting from random initialization, two models learn in tandem: a generator proposes programs interpreted by a universal Turing machine, generating byte sequences, while a learner autoregressively predicts these byte sequences. The learner is trained with standard cross-entropy, while the generator is trained with reinforcement learning to produce sequences at the frontier of the learner's capabilities, yielding an adaptive curriculum. A universal Turing machine gives us a search space over all computable data-generating processes, imposing little domain-specific structure, and self-play searches over this space for useful training data. We test whether zero-shot performance on natural data improves predictably with self-play compute; this is a clean test of transfer since neither generator nor learner is trained on natural data. Across several natural datasets, zero-shot loss exhibits predictable scaling in compute. The models also exhibit in-context learning, and discover recognizable mathematical sequences during training. --- ### KernelOPT: Dispatch-Aware Agentic Search for GPU Kernel Optimization Authors: Aheli Poddar, Sanskar Prasad, Arindam Samanta, Subha Chakraborty, Vishal Goyal, Rohit Singh Rathaur Published: 2026-09-24 Categories: cs.DC, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2609.30059v1 Deep learning inference and training performance depends critically on GPU kernel efficiency. Modern compilers such as PyTorch Inductor automatically generate GPU kernels from high-level model code, but frequently underperform expert-written implementations by wide margins. Recent LLM-assisted kernel optimizers can close this gap for standalone kernels, yet treat compiled models as black boxes, generally optimizing individual standalone kernels without respecting the compiler's structural decisions or verifying the model end-to-end. We present KernelOPT, a multi-agent system that treats compiled models as structured artifacts. It preserves vendor library calls (cuBLAS, cuDNN) and exclusively targets generated Triton sub-kernels using five profiling-guided LLM agents. A four-gate verification cascade of static validation, multi-seed correctness, model-level float64-fallback verification, and performance gating filters candidates during optimization and verifies the re-stitched model end-to-end. If no candidate passes all four gates, the system preserves the compiler baseline. The system accepts PyTorch nn.Modules, standalone Triton kernels, and Helion kernels. Evaluated on 250 KernelBench problems, KernelOPT achieves geometric mean speedups over \texttt{torch.compile} of 1.40$\times$ (Level 1: 51/100), 1.15$\times$ (Level 2: 31/100), and 1.07$\times$ (Level 3: 12/50) across all problems. --- ### Can Labor Markets Function in the Age of AI? The Evaluation Bottleneck in Hiring Authors: Itai Ashlagi, Ramesh Johari, Jon Kleinberg, Anushka Murthy Published: 2026-09-24 Categories: cs.GT, cs.AI, cs.CY, econ.TH arXiv: https://arxiv.org/abs/2609.30058v1 AI-assisted job-search tools have become increasingly popular by making it easier to find and apply to jobs. But by making it easier for applicants to generate and tailor application materials, they can also reduce how informative those materials are about applicant fit. We study this tradeoff in a hiring market where applicants differ in experience and latent match quality and firms use noisy application materials to decide whom to screen. We ask how AI affects downstream screening and hiring, and which applicants are most adversely affected. As application materials become less informative, a Bayesian firm rationally relies more heavily on coarse observables such as prior experience. Among the four applicant types defined by experience and compatibility for the job, inexperienced-compatible applicants are the most exposed: they lack observable experience and lose the individualized information that could distinguish them from other inexperienced candidates. When screening is costly, these changes can also generate inefficient screening failures in which firms screen no applicants or screen only experienced applicants. We then show that multistage hiring can arise as an endogenous firm response: a relatively inexpensive intermediate assessment allows firms to acquire new evidence of fit before costly full screening. This can restore screening opportunities that disappear under one-stage hiring and give inexperienced-compatible applicants a path to screening. Our results show how AI can shift the central friction in hiring from submitting applications to obtaining credible evaluation, creating entry barriers for high-fit workers without prior experience. Multistage hiring can endogenously arise in response, restoring evaluation opportunities that would otherwise disappear and helping preserve market functioning. --- ### M3GD: Multi-Modal Multi-View Geometric Diffusion for Camera--LiDAR Novel View Synthesis Authors: Yang Zhou, Jiuhong Xiao, Shizhao Ye, Long Quang, Carlos Nieto-Granda, Giuseppe Loianno Published: 2026-09-24 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2609.30056v1 Robotic novel view synthesis (NVS) must recover both visual appearance and metric 3D structure, yet most generative NVS methods rely only on images, overlooking LiDAR, a complementary sensor common on robotic platforms. We present M3GD, a Camera--LiDAR multimodal representation for generative NVS that composes independently pretrained 2D image and 3D point-cloud foundation models without separately pretraining a cross-modal translator. We show that, after camera projection, frozen LiDAR and image features exhibit substantial shared spatial structure, providing a natural cross-modal representation. M3GD conditions generation on LiDAR through this structure: it combines explicit geometry statistics with learned point-cloud descriptors into view-aligned packets on the image-latent grid, injected through a lightweight residual adapter into a multi-view flow-matching generator whose latent space, decoders, and training objective remain intact. On the GrandTour dataset, M3GD improves target-view RGB and depth synthesis over an image-only version of the same backbone. Ablations show that the gains come from pixel-aligned LiDAR content and that target-view LiDAR acts as a geometric query linking the requested view to source observations. Deployment on a ground robot demonstrates practical real-world operation, with a configurable quality--cost trade-off controlled by the number of Euler integration steps. --- ### Era by Eon: Benchmarking Enterprise Agents on Hidden Knowledge Authors: Benjamin Gruenbaum, Doron Porat, Assaf Natanzon, Roy Zavida, Chen Dinachi, Or Itzahary Published: 2026-09-24 Categories: cs.SE, cs.AI arXiv: https://arxiv.org/abs/2609.30055v1 In the Era by Eon benchmark, each question states the rules for its answer, and code computes the answer from a generated company's data. When agents can run code, the four strongest models each answer 22 to 25 of 27 such questions, so the benchmark barely separates them. We add eight question templates that depend on hidden facts. No question or document states a hidden fact, and the records that seem to hold it show something else. Other data implies it. For example, the sales system says a customer dropped a purchase because of timing. On a recorded call, the customer blames an outage. For each generated company, code fills each template and computes an exact answer without a language model. We evaluate 12 agents. Each pairs a model with an agent program, which connects it to the company's systems. The best agent answers 18 of its 24 attempts, three per question, correctly. Four of the six models answer at most 6 of 24 with any program. The hardest questions require picking one of several similar records, such as which of three renewal offers a customer signed. All agents together answered two such questions correctly in only 1 of 84 attempts. --- ### SciWalker: Synthesizing Scientific Coding Problems with Operator Graphs and Execution Feedback Authors: Chenxi Li, Wenxuan Zeng, Yun Luo, Fangchen Yu, Peng Ye, Yu Cheng, Jun Zhang Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30054v1 Improving the scientific coding capabilities of large language models (LLMs) requires high-quality training data. However, such data remain scarce because manually authoring realistic problems is costly and time-consuming, while systematically covering diverse scientific domains and algorithmic combinations remains challenging. To address this, we introduce SciWalker, a framework for synthesizing scientific coding problems through operator-chain sampling and execution feedback. The framework combines scientific library interfaces with operation modes to instantiate operators, organizes them into operator graphs, and samples operator chains as computational workflow cues. Guided by these cues, we adopt LLMs to generate scientifically grounded problem statements, reference solutions, and tests, with failed generations iteratively repaired using execution feedback. By combining structured workflow composition with verification and quality review, SciWalker enables scalable task generation while promoting scientific grounding, computational diversity, and executability. Using this framework, we construct 8,178 high-quality problems spanning 5 scientific domains and 32 subdomains. To evaluate their training utility, we conduct reinforcement learning on Qwen3.5-9B using the GSPO algorithm. This training improves SciCode subproblem accuracy by 9.9 percentage points, from 29.3% to 39.2%, with gains across scientific code generation, code repair, and reasoning benchmarks. The code for SciWalker is available at https://github.com/lichenx1/SciWalker. --- ### NNV3: Expanding Neural Network Verification to New Architectures and Domains Authors: Anne M. Tumlin, Samuel Sasaki, Ben Wooding, Diego Manzanas Lopez, Muhammad Usama Zubair, Navid Hashemi, Hongchao Zhang, Waseem Abbas, Ipek Oguz, Meiyi Ma, Taylor T. Johnson Published: 2026-09-24 Categories: cs.AI arXiv: https://arxiv.org/abs/2609.30050v1 We present NNV3, the latest version of the Neural Network Verification (NNV) tool, a MATLAB framework for formal verification of deep learning models and learning-enabled cyber-physical systems. Building on the set-based reachability foundation of NNV 1.0 (FFNNs, CNNs, NNCS) and NNV 2.0 (RNNs, SSNNs, neural ODEs), NNV3 introduces new members of the Star-set family: ModelStar for verifying networks under weight perturbation, VolumeStar for video and 3D volumetric inputs, and GraphStar for graph neural networks. A conformal-inference-based probabilistic reachability mode complements sound analysis for problems where deterministic verification is intractable, while FairNNV certifies counterfactual and individual fairness properties over continuous input regions. NNV3 introduces new benchmarks for malware detection, graph-based power-system models, medical imaging, variable-length time series data, and action recognition. NNV3 also incorporates tutorials and developer guides through a unified documentation site. This paper details these major updates, demonstrating NNV's maturation into a comprehensive, robust, and accessible verification tool for a diverse range of AI systems. --- ### Style, Not Self: Surface Cues Explain Zero-Shot Code Attribution by Large Language Models Authors: Ehsan Barkhordar, Surendrabikram Thapa Published: 2026-09-24 Categories: cs.AI, cs.CL, cs.SE arXiv: https://arxiv.org/abs/2609.30048v1 If a language model can recognize code it wrote, it may favor that code as a judge, and instances of one model monitoring each other could collude. We test this zero-shot on current commercial models. Five LLMs generate solutions to MBPP, HumanEval, and DS-1000, seven more to MBPP, and models act as evaluators in four tasks: picking their own solution from a pair, judging whether a single solution is their own, identifying which of two solutions a named model wrote, and judging quality blind. In the single-solution task, balanced accuracy is 49-58% for all 15 model-benchmark combinations, while raw accuracy (38-67%) mostly reflects how readily a model claims authorship. In the pairwise task, accuracy across 14 evaluator-opponent combinations correlates at r=0.93 with how often the evaluator's solution is longer. Attribution to a named model succeeds on some pairs and is consistently inverted on others. A rule-based normalization that strips docstrings, comments, type hints, and local names preserves Pass@1 and leaves ten of twelve re-tested results at chance; the other two follow a length difference it leaves, although a trained classifier still separates most normalized pairs. Claude Haiku's self-preference also disappears. We recommend reporting balanced accuracy, heuristic baselines, and label consistency. --- ### From Processing to Functionality: Engineering Accessible Material States in Cu-Embedded SiO$_x$ Memristive Devices Authors: Tobias Gergs, Rouven Lamprecht, Sahitya Yarragolla, Ole Gronenberg, Luca Vialetto, Hermann Kohlstedt, Thomas Mussenbrock, Jan Trieschmann Published: 2026-09-24 Categories: cond-mat.mtrl-sci, cs.LG, physics.plasm-ph arXiv: https://arxiv.org/abs/2609.30047v1 Resistive switching in oxide-based devices is widely governed by stochastic defect processes, yet a predictive link between fabrication conditions and functional behavior remains elusive. Here, we establish a multiscale framework connecting plasma-defined deposition conditions to macroscopic device functionality in sputtered SiO$_x$/Cu/SiO$_x$-based systems. By combining large-scale statistical analysis of more than 50,000 experimentally characterized devices with physics-based plasma and atomistic simulations, we show that device behavior does not emerge from deterministic process-to-performance mappings, but from a probabilistic cascade spanning defect formation, defect-state evolution, and functional-regime emergence. Data-driven clustering reveals a continuous functional state space composed of operational switching types, while inverse modeling identifies the reconstructed oxygen-vacancy density as an effective latent descriptor capturing the combined influence of structural disorder and defect topology. This latent descriptor is strongly coupled to both Cu redistribution and electrical response, linking otherwise hidden material properties to observable device characteristics. Furthermore, macroscopic switching behavior is argued to arise from ensemble integration across spatially heterogeneous subdomains, providing a physical explanation for the pronounced variability of large-area devices. These findings shift the perspective from deterministic defect engineering toward probabilistic defect-state design and establish a physically grounded framework for understanding and controlling functional variability in such oxide-based systems, such as memristive or resistive-switching devices. --- ### ConPro: Contrast Projection Pretraining for Label-Efficient Vessel Segmentation in DSA Sequences Authors: Xinge Guo, Yuanhao Wang, Liqi Shu, Yang Liu, Min Xu Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30043v1 Dense vessel annotation in digital subtraction angiography (DSA) is labor-intensive, yet every unlabeled sequence records how contrast passes through the vessels. Semi-supervised methods take their targets from the current model, and generic self-supervised pretexts reconstruct static appearance, so this signal goes unused. We propose ConPro, a self-supervised pretraining scheme whose target is a contrast projection, the normalized drop of every pixel below its temporal median over the sequence. On DIAS and DSCA, with 10%, 20% and 50% of the training cases labeled, ConPro improves on training from scratch at every label fraction and is the best of the compared methods on DSCA at 20% and 50% labels. Controlled comparisons show that the gain comes from the target. A temporal-median target with the same input, loss and budget stays at scratch level, and using the projection directly instead of learning it, as an input channel or a pseudo-label, helps little or hurts. ConPro provides pretrained weights without changing the segmentation architecture, so it combines with semi-supervised training, and UniMatch, the strongest baseline, gains 0.5 to 2.0 Dice and 0.9 to 2.3 clDice at every label fraction when started from ConPro weights, reaching 75.4 Dice on DIAS and 81.3 on DSCA. --- ### AERIAL: Adversarial Evaluation of Robustness in Accuracy-Preserving Low-Precision EEG Decoders Authors: Saim Rehman, Muhammad Shafique Published: 2026-09-24 Categories: cs.CV, cs.CR, cs.LG, eess.SP arXiv: https://arxiv.org/abs/2609.30037v1 Deployment-oriented compression is attractive for resource-constrained brain--computer interfaces (BCIs), but whether it changes adversarial vulnerability remains unclear. On BCI Competition IV-2a, we compare 32-bit floating-point (FP32) EEGNet and ShallowConvNet models with global magnitude pruning and simulated INT8 post training quantization (PTQ) and quantization-aware training (QAT) across nine subjects and three seeds. Simulation provides differentiable quantize--dequantize models for white-box attacks and gradient analysis, while native TensorRT deployment is used for validation. Accuracy-preserving compression does not improve direct robustness: at $ε=0.005$, EEGNet PGD accuracy remains 22--24\% across FP32, 50\% pruning (P50), PTQ, and QAT. However, P50 reduces bidirectional transfer efficiency to 0.963/0.928 (FP32$\rightarrow$P50/P50$\rightarrow$FP32), versus 0.994/0.997 for PTQ; the same trend holds for ShallowConvNet. Gradient alignment shows a corresponding separation, while native PTQ agrees with simulated clean/adversarial predictions in 95--98\% of cases. These results show that direct robustness, adversarial transfer, and deployment efficiency are distinct properties of compressed EEG decoders. --- ### Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think Authors: Xvyuan Liu, Jianjie Fang, Chen Gao, Yong Li Published: 2026-09-24 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2609.30036v3 Latent world models plan toward goal images with a frozen pretrained predictor, without task rewards or extra trained heads. However, their planners struggle with long-range goals, and prior work addresses this by training extra components such as value functions or subgoal models. We show that the planning target itself can cause this failure: even with exact dynamics and globally optimal short-horizon search, scoring predictions by their distance to the final goal rejects the first steps of a route that initially moves away from the goal. Building on this insight, we propose Anchored Planning (AP), a training-free method that reuses the world model's own offline trajectories. AP retrieves a segment that leads from the current observation toward the goal and aims the frozen planner at an observation shortly after the segment's start. Across four diverse tasks, AP substantially improves frozen LeWM planners for both action synthesis and action ranking, and it outperforms both additional final-goal search and the LeWM planner on long-range goals. --- ### Aim Short to Reach Far: Your Frozen World Model Can Plan Better Than You Think Authors: Xvyuan Liu, Jianjie Fang, Chen Gao, Yong Li Published: 2026-09-24 Categories: cs.LG, cs.RO arXiv: https://arxiv.org/abs/2609.30036v1 Planners built on visual world models commonly score each predicted outcome by its distance to the encoded goal image. We show that this target can limit control even with exact dynamics and globally optimal short-horizon search: reaching a goal may require actions that initially move away from it. With frozen LeWM models, intermediate targets substantially improve action synthesis and recorded-action ranking on Cube, PushT, Reacher, and TwoRoom. Learned targets and targets drawn from observed experience both produce these gains. We introduce Anchored Planning, which retrieves a recorded segment whose start and end resemble the current and goal observations, then aims at an observation shortly after its start. The frozen model scores actions toward this target from the current state. Without additional training, planning toward observed targets outperforms the released LeWM planner on every task in our long-range evaluation. Additional final-goal search falls short of the same gains. Lower successor-prediction error need not translate into better control. Success also depends on how far ahead the target is placed and on shrinking the retrieval span as execution advances. Changing only the target lets the same frozen model and planner reach goals that final-goal scoring misses. --- ### Artificial Societies Benchmark: A Validation Framework for Synthetic Research Authors: Edoardo Chidichimo, Min Jun Jung, Felix P. S. Wallis, James K. He Published: 2026-09-24 Categories: cs.CL arXiv: https://arxiv.org/abs/2609.30030v1 A synthetic survey can reproduce the average answer while misrepresenting how people differ, how their answers relate to one another, or how they respond to changes in conditions. We introduce the Artificial Societies Benchmark to help researchers assess whether synthetic populations support their intended analyses. The framework combines eleven tests across internal, construct, and external validity, drawing on twenty human sources and comparing nine language models. It connects each research use to the evidence it requires and tests how results change with the information we supply about respondents. Importantly, strong performance in one domain does not establish fidelity in the others. Models often answer too consistently, compress response scales, and alter relationships between traits whilst richer profiles improve prediction for some models and worsen it for others. The resulting scorecard helps researchers identify which aspects of a synthetic population can support their analysis and where researchers need further human evidence. --- ### How does Adversarial Influence Scale in Multi-Agent Systems? Authors: Addison J. Wu, Jasin Cekinmez, Michel Liao, Karthik Narasimhan, Thomas L. Griffiths Published: 2026-09-24 Categories: cs.AI, cs.CY arXiv: https://arxiv.org/abs/2609.30028v1 Multi-agent deliberation can improve performance, but what happens when some agents do not act in good faith? In practice, an agent may be deceptive and work to subvert the group, whether through its own objectives or external instruction. We study how susceptibility to deception scales as groups increase in size and deceivers become more prevalent. It is not the number of agents in the group that matters, but the proportion of deceivers. We observe that the defection rate, how often initially correct agents switch to an incorrect final answer, rises linearly with this proportion. Whereas humans in comparable conformity studies are reliably swayed only when misleading confederates form a majority, LLM agents defect regularly even when deceivers remain a minority. Susceptibility also depends on which models are interacting, especially on the honest agent side. Unexpectedly, allowing deceivers to coordinate privately can make them less effective. Altogether, our results show that adding more agents is therefore not a sufficient defense, because the adversary can simply scale with the group. --- ### Synthetic Hospital: An Open, Verifiable, Physician-Validated Longitudinal EHR Benchmark Authors: Christine Park, Valerie Chen, Tim Dettmers Published: 2026-09-24 Categories: cs.AI, cs.DB arXiv: https://arxiv.org/abs/2609.30027v1 Frontier language models are rarely used in clinical workflows because the realistic, longitudinal benchmarks needed to develop them are scarce. Real electronic health record (EHR) data cannot be openly shared due to privacy, ethics or data use issues and it does not contain verifiable ground truth since the chart records only reflect what clinicians documented. We introduce Synthetic Hospital, an open, fully synthetic, fact-grounded longitudinal EHR benchmark that resolves the open sharing and verifiable ground truth barriers. Built entirely from public medical-education material with no protected health information, it comprises 1,268 longitudinal patients and 5,602 encounters, where every diagnosis, finding, and temporal relation is grounded in standard ontologies (ICD-10-CM, SNOMED CT, LOINC) and with a complete provenance chain back to its source medical education material. Synthetic Hospital is served through a simulated hospital record system that mirrors real EHR infrastructure (standard interoperability APIs, role-based access and function-calling interface). In a blinded review, physicians distinguished its records from real patient charts at near-chance rates (53\%). Across 10 frontier and open models, none approaches ceiling: the best model reconstructs a patient's longitudinal problem list with a severity-weighted F1 of 0.73, level with the mean of seven physicians on a matched subset but well below the best of them (0.89), and misses roughly half of clinically relevant findings when summarizing a chart. Overall, these results highlight that Synthetic Hospital is a difficult and realistic test of clinical AI performance. --- ### Training-Free Hold-Usage Detection in Sport Climbing with Foundation Pose Models Authors: Abu Bakar, Abdullah Aftab, Amir Hamza Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.30026v1 Detecting which holds a climber uses, and when, underpins automated scoring, movement analysis, and assistive systems for sport climbing. Existing approaches train task-specific models or repurpose 2D pose estimators whose hand keypoint sits at the wrist and foot keypoint at the ankle i.e. offset from the fingertips and toes that actually contact the holds, and whose hands are occluded in roughly half of all frames. We show that a frozen, off-the-shelf pose foundation model is sufficient: using the fingertip and toe keypoints of Sapiens, a per-frame proximity test against the annotated holds, per-limb mutual exclusion, and a short temporal-persistence rule, we detect hold usage without any climbing-specific training. On the The Way Up dataset (22 videos, 10 athletes, two routes), our method reaches an event F_1 of 90.2% on a held-out split (89.8% under leave-one-participant-out cross-validation) and 79.9% over all 22 videos at any temporal overlap, and performs best on footholds (F_1,89.8% overall, 96.6% held-out). Under an identical protocol it exceeds our reproductions of the YOLOv8-pose and ViTPose pipelines at every temporal threshold, with the margin widening under strict timing. An ablation shows that two intuitively helpful additions---dense foundation-feature change gating and body-part segmentation---both hurt, arguing that a minimal, keypoint-only design is the right one for this task. Finally, standard coaching statistics computed from our automatic predictions track ground truth closely (Pearson r=1.00 for climb time, 0.94 for pace), turning ordinary single-camera video into reliable performance metrics with no instrumentation. --- ### Canopy: Exploiting Piecewise Smooth Tree Priors for Multi-Fidelity Bandits Authors: Michael Jerge, Suman Jana Published: 2026-09-24 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2609.30017v1 Many LLM inference problems, including model routing, prefix-cache management, prompt trimming, and test-time search, can be viewed as optimization over a tree. This structure arises naturally from autoregressive generation: every prefix defines a node, and its continuations form a subtree below it. Internal nodes of the tree provide cheap but biased estimates of a region's value, while leaf evaluations are expensive but accurate. Hierarchical bandit methods can exploit this structure, but typically require a specific smoothness schedule to be specified in advance, even though real objectives are often only piecewise smooth and their optima may lie near sharp boundaries. We introduce CANOPY, a multi-fidelity tree bandit that learns where the smoothness prior is valid rather than assuming it globally. CANOPY uses cheap random-path probes to construct an online certificate of local aggregation bias, then directs expensive leaf evaluations toward cells where the certificate detects a smoothness violation. We prove fixed-budget and regret guarantees whose additional cost is additive in the number of discontinuities, recovering the smooth-tree rate when no violations are present and approaching structure-blind search as violations become dense. Across routing, top-$k$ identification, test-time search, caching, and prompt trimming, CANOPY consistently improves matched-budget performance, including $2.9\times$ higher top-10 recall on a 1000-model pool, $1.6\times$ more SWE-bench Verified issues resolved than best-of-$N$, and $3.6\times$ lower median time-to-first-token with prefix caching. --- ### Low-Cost Assays for Measuring Model Behavior Across Vendors and Releases Authors: Tapan Parikh Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.30012v1 Language models advise people, keep them company, and write software while they sleep. Measuring what they do is hard: behavior has to be sampled repeatedly across models, prompts and releases, most of it lives in unstructured text that has to be coded before it can be counted, and the result has to be legible and rigorous enough to meaningfully compare models and vendors. To address these constraints, we present a simple, cheap, scalable, and replicable model for studying model behavior. Each study is a frozen, public stimulus run identically on a cross-vendor panel, at a few dollars per model or less. Each reads its transcripts one of three ways, chosen by how much interpretation the behavior needs: exact match on a clamped reply, a codebook applied by LLM judges whose agreement with a human coder is reported per code, and an instrumented environment that records what an agent did independently of what it said. Run across four years of model releases from both frontier and open-source labs, these instruments find four things. Convergence: asked to pick a word, 27 of 44 models answer serendipity at least once in four tries. Resistance: a trailing "right?" moves endorsement by up to 32 points, and the sign flips from sycophantic to resistant as generations advance, keyed to the tag's surface form. House: whether a model holds a position under pressure tracks its generation, and how it holds tracks the lab that built it. Account: told to do something the documentation in their repository contradicts, some coding agents never went along silently and others always did, and the same model can change with the harness it runs in. Re-run on every release, batteries like these track how behavior is changing across vendors and over time. --- ### Automated Regulatory Compliance Question Answering in Financial Services with Domain-Adapted Retrieval-Augmented Generation Authors: Tobias Deußer, Abhishek Pillai, Aurelio F. Bariviera, Dhananjay Bhardwaj, Lorenz Sparrenberg, David Berghaus, Christian Bauckhage, Rafet Sifa Published: 2026-09-24 Categories: cs.CL, cs.AI arXiv: https://arxiv.org/abs/2609.30009v1 Financial institutions operate under dense, frequently amended rulebooks, and answering a compliance question correctly requires not only fluency but verifiable grounding in the authoritative text. Large language models are attractive for this task, yet the models that firms can realistically deploy on-premise are compact ones, and compact models hallucinate obligations. We study whether a carefully domain-adapted retrieval-augmented generation pipeline closes that gap. Our retriever is built in three stages on top of LegalBERT: entailment tuning that recasts question--passage matching as premise--hypothesis reconstruction, contrastive tuning with in-batch negatives, and score-level fusion with BM25. Our generator is a compact model (2B--12B parameters) served under 4-bit quantization, either prompted or adapted with retrieval-aware fine-tuning (RAFT) through LoRA. On ObliQA, a question-answering benchmark built from the Abu Dhabi Global Market rulebooks, the staged retriever raises Recall@10 from 0.256 to 0.774 and outperforms BM25 (0.678) and E5-large-v2 (0.758), the strongest general-purpose dense encoder we tested. RAFT-LoRA then improves the composite RePASs answer-quality score for every model we could adapt, with the largest gain on the weakest one. However, the adapted models do not transfer to Australian case-law questions, and a closed-book model that receives no passages at all scores within 0.011 RePASs of the full pipeline while producing answers that cite nothing and misstate obligations. The retrieval gain is therefore measured directly, the generation gain is a gain in RePASs rather than demonstrated grounding, and grounding itself requires an evaluation protocol that RePASs does not provide. --- ### VietPrism: A large-scale Vietnamese speech and deepfake corpus with diverse dialects and code-switching Authors: Minh Hoang, Thai Le Published: 2026-09-24 Categories: cs.CL, cs.SD arXiv: https://arxiv.org/abs/2609.30005v1 Vietnamese speech research is constrained by resources that isolate automatic speech recognition from speaker, dialect, code-switching, and deepfake analysis. We introduce VietPrism, an open, multi-domain corpus that brings these dimensions together at scale: 993.4 hours and 403,941 bona fide utterances from 1,262 verified speakers across 8,388 real-world videos. To our knowledge, it is the first large-scale Vietnamese corpus to jointly provide transcripts, consistent speaker identities, five dialect groups, and naturally occurring Vietnamese--English code-switching, which constitutes nearly half of the corpus by duration. We further create over 3.1K hours of spoof speech with four open-source and commercial synthesis systems. Every spoof is conditioned on a verified speaker reference and paired with a transcript- and speaker-matched bona fide utterance, enabling unique controlled evaluation with reduced lexical and identity confounds. Zero-shot evaluation of five pretrained multilingual detectors reveals striking brittleness: EER greatly varies across detector--generator pairings, while recent multilingual detector DFA-1B degrades from 16.3% to 33.6% as speaker similarity increases. Dialect-stratified results expose further model-dependent disparities. By unifying natural linguistic diversity with controlled spoof generation, VietPrism provides a challenging foundation for Vietnamese speech modeling and trustworthy audio-deepfake detection. --- ### Advancing Model Research in AgentX: Long-Horizon Autonomy for Industrial Recommender Systems Authors: Shuang Yang, Zijie Zhuang, Changxin Lao, Pengbo Xu, Hanwen Xu, Yusheng Huang, Han Gao, Guanchen Wang, Tianbao Ma, Linxun Chen, Peilin Song, Xuming Wang et al. Published: 2026-09-24 Categories: cs.AI, cs.IR arXiv: https://arxiv.org/abs/2609.30001v1 Sustaining industrial recommendation research requires using the results of one experiment to decide what to investigate next. We present AgentX-Model, the next generation of AgentX's model research framework, which connects proposal development and model experimentation within sandboxes defined by business inputs and prediction tasks. AgentX-Model adopts a dual-agent architecture comprising a Research Agent and a Model Agent. The Research Agent develops independently reviewed proposals from papers and experimental findings, while the Model Agent conducts multi-round investigations and returns code, measurements, and unresolved questions. Using the returned results, the Research Agent selects a starting implementation and formulates the next research question, allowing subsequent experiments to build on earlier findings. We organize this continuing research around four actions: Reproduce, Follow-up, Composition, and Diagnose. The first three actions drive routine research, while Diagnose acquires the evidence needed to choose a repair, including for issues raised by business feedback and online evaluation, such as prediction bias measured by PCOC. Across the production evaluation, 560 of 636 completed model-changing experiments recorded AUC above their business baselines. As research continued, some experiments recorded AUC above every comparable ancestor in their lineages. The five latest online A/B evaluations across different business settings reported gains including 10-15% in acquisition efficiency, 15-20% in target-segment advertising spend, and 0.3-0.8% in watch time; the watch-time model used approximately 10% fewer FLOPs and parameters. A dependency-aware historical-replay benchmark further evaluates research allocation, with initial results showing no consistent efficiency gain from more complex scheduling when agents already analyze and select concrete candidates. --- ### GHOST-Q: Towards Studying Grounding Hallucinations Overlooked Under Same-score TradeOffs in Quantized VLMS Authors: Saim Rehman, Muhammad Shafique Published: 2026-09-24 Categories: cs.CV, cs.AI, cs.LG, cs.MM arXiv: https://arxiv.org/abs/2609.29999v1 Post-training quantization of vision--language models (VLMs) is typically assessed through aggregate task accuracy and memory savings, but preserving a headline score does not guarantee preservation of visual grounding behavior. We present GHOST-Q, a cross-precision controlled evaluation of three 8B VLM families under FP16, INT8, and NF4 across utility and hallucination-sensitive benchmarks. Rather than comparing only aggregate accuracy, we pair FP16 and quantized predictions item by-item to quantify how compression redistributes grounding successes and failures. Five of six quantized variants preserve MMStar accuracy within $\pm2$ percentage points, yet 10 of 36 paired effects remain significant after false-discovery-rate correction, nine on hallucination-sensitive conditions. Same-device A100 profiling further demonstrates that substantial memory reduction does not necessarily mean lower inference latency. Finally, an open-ended AMBER audit reveals strong generation budget censoring whose severity varies by architecture and precision. These results show that quantized VLMs should be evaluated jointly for aggregate utility, grounding reliability, generation behavior, and realized deployment efficiency. --- ### Guardrails or Roadblocks? Effects of Pedagogical Style and Context Awareness in AI Teaching Assistants for Programming Authors: Madeleine Eastwood, Harshith Narne, Joseph Hilby, Paul Denny, Ashish Aggarwal, Amanpreet Kapoor Published: 2026-09-24 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2609.29995v1 AI teaching assistants (AI TAs) backed by large language models (LLMs) and pedagogical guardrails are increasingly being integrated into programming courses, providing students with scalable access to hints, conceptual explanations, and code-level feedback. However, guardrails may also create friction. If students feel that the support provided is overly restrictive or poorly contextualized to their current progress, they may bypass approved tools for general-purpose LLMs. To investigate how AI TA design affects students' learning experiences, we conducted a randomized controlled trial with 132 students in an introductory programming course. Students completed three tasks related to code-writing and debugging and were randomly assigned to one of four AI TAs varied across two dimensions: pedagogical guidance style (Socratic vs. Direct instruction) and context awareness (no context vs. full context of the problem and student solution). We examined students' perceptions, interaction behaviors, and evidence of post-task comprehension. Students rated the Socratic AI TA with full context least favorably, reporting significantly lower perceived support for task completion. Descriptively, this condition also showed the highest observed interaction stress, the highest rate of external LLM use, and the lowest proportion of post-task explanations demonstrating full comprehension, though these differences were not statistically significant. These findings suggest that guardrailed AI TAs are not automatically better for learning. Instead, their effectiveness depends on how pedagogical guidance and contextual awareness are balanced in ways that students experience as useful, supportive, and worth continuing to use. --- ### Let Training Guide Selection: Online Synthetic Data Filtering via Real-Anchored Utility Authors: Yanran Wu, Sana Lakdawala, Renzo Tassara Miller, Chongyang Bai, Sharath Ciddu, Shivendra Pratap Singh, Kungang Li, Sandeep Pandey, Chunwei Liu Published: 2026-09-24 Categories: cs.LG arXiv: https://arxiv.org/abs/2609.29988v1 Synthetic data can scale training supervision when real-world data are limited, but noise and distribution mismatch can reduce its value. Existing synthetic data selection methods often emphasize fidelity or diversity rather than the learner's evolving needs. We propose FROST, an online framework that estimates synthetic-data utility through gradient feedback anchored in real training data. It calibrates batch utility against recent history to determine when filtering is needed and filters samples only in out-of-band batches to determine what to retain, without an external verifier or held-out validation set. Experiments on two public benchmarks for image classification and LLM fine-tuning for text-to-SQL show that FROST filters out around 20--30% of the synthetic data while improving real-task performance compared with training on the full synthetic data pool. We further apply FROST during training in a large-scale industrial ads re-ranking system, achieving significant performance gains over a highly optimized production baseline, demonstrating its effectiveness and generalizability. --- ### OceanXL: Large-scale Underwater 3D Gaussian Splatting via Block Partitioning and Adaptive Pruning Authors: Haoran Wang, Shaoyu Cai, Adrian Azzarelli, Zhuodong Jiang, Guoxi Huang, Eng Tat Khoo, Brett Seymour, Fan Zhang, David Bull, Nantheera Anantrasirichai Published: 2026-09-24 Categories: cs.CV arXiv: https://arxiv.org/abs/2609.29985v1 Underwater 3D reconstruction is critical for marine exploration, ecological monitoring, and subsea infrastructure inspection, yet remains challenging at large scale due to light attenuation, scattering, and limited capture coverage. While 3D Gaussian Splatting (3DGS) enables high-quality real-time rendering, its application to large underwater scenes is constrained by high memory consumption and inefficient optimization over extensive areas. We propose OceanXL, a fast and scalable 3DGS-based framework for large-scale underwater reconstruction. OceanXL adopts a divide-and-conquer strategy, partitioning scenes into spatially coherent blocks to enable efficient optimization while preserving global geometric consistency. We further introduce an adaptive pruning scheme tailored to underwater conditions that removes redundant primitives, producing compact representations without sacrificing visual fidelity. Together, these components improve training efficiency and rendering performance for large scenes. We also introduce a large-scale underwater dataset covering diverse marine environments. Experiments on five large-scale scenes demonstrate favorable scalability, compactness, and efficiency--quality trade-offs over large-scene baselines. Controlled comparisons on the small-scale SeaThru-NeRF dataset further show competitive reconstruction quality with substantially smaller model sizes than underwater-specific methods. ---