# Research Papers — 2026-10 · bundle page 116 of 156 > Full-text markdown of papers #5751–5800 of 7799 published in 2026-10, newest first. AI-readable mirrors of the full 7799-paper 2026-10 archive (50 papers per file). Canonical page: https://aipod.dev/papers/archive/2026-10/page/116 [← Previous bundle](https://aipod.dev/papers/archive/2026-10/page/115.md) [Next bundle →](https://aipod.dev/papers/archive/2026-10/page/117.md) ### Seeing the Invisible: Physics-Guided Visual Prompting for Temperature- and Radiation-Aware VLA Navigation Authors: Hojoon Son, Fan Zhang Published: 2026-10-06 Categories: cs.RO, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.07558v1 Vision-Language-Action (VLA) models have become a major paradigm for Vision-and-Language Navigation (VLN). However, in safety-critical facilities, invisible risks such as radiation or temperature spikes cannot be detected by an RGB camera, and handling each risk is expensive, requiring a new encoder, new data, and model retraining. We propose Physics-Guided Visual Prompting (PG-VP), a plug-and-play multimodal perception module that instead reuses what a frozen VLA model already does well: avoiding visible obstacles. Given a proximal radiation or thermal source, PG-VP performs a physics-guided risk assessment to determine the avoidance direction and overlays a corresponding virtual obstacle that moves across consecutive frames (Dynamic Visual Prompting). The navigation policy then naturally detours around this invisible hazard. The identical virtual obstacle is used regardless of hazard type, so the visual prompting pattern remains fixed as sensors are added. When no hazard is detected, nothing is rendered, and the policy behaves exactly as it would without PG-VP. We evaluate PG-VP on OmniNav using the val-unseen splits of R2R-CE and RxR-CE, where it guides the policy toward intended low-risk actions in 84.9% and 83.2% of cases, at a cost of 6.8 and 7.9 percentage points in navigation success rate. We further test it with distinct scenarios on a real robot in the presence of actual thermal and radiation sources, all without any retraining. The real test shows that PG-VP effectively avoids these invisible hazards, improving worst-10% average trajectory safety by 63.45% and 32.59% against thermal and radiation sources, respectively. --- ### CheckerBench: Can Long-Horizon Agents Synthesize Static-Analysis Checkers? Authors: Hang He, Li Wang, Hao Chen, Yuchen Shao, Yuling Shi, Lisheng Wang, Peiyang Liu, Goose Lin, Zaiyuan Wang, Haiying Sun, Ting Su, Chengcheng Wan Published: 2026-10-06 Categories: cs.SE, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2610.07557v1 Static-analysis checker synthesis requires agents to interpret a defect specification, inspect a repository, implement analyzer-specific logic, and refine the checker through repeated compilation and analysis feedback. Existing coding-agent benchmarks focus on tasks such as patch generation or vulnerability detection and rarely assess whether an agent can develop a working checker in a repository from start to finish. We introduce CheckerBench, an executable benchmark of 300 tasks derived from 297 CVEs across 167 repositories, 85 CWEs, and five language ecosystems. Each task includes vulnerable and fixed revisions, a pinned analysis environment, and a checker scaffold. We further introduce CheckerLab, a common evaluation framework that independently rebuilds submitted checkers and measures vulnerable-fixed diagnostic contrast, patch localization, false positives, and tool use. Across 21 model-harness configurations and three independent repeats per configuration, mean Pass@1 is 32.30%, while the best reaches 45.33%. These results show that reliable, reusable checker development remains challenging for current coding agents. --- ### Decoupled Multi-Agent Orchestration Authors: Xinle Wu, Yao Lu Published: 2026-10-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07556v2 Learned orchestration can automatically construct effective language-model multi-agent systems, but existing approaches couple planning to fixed worker pools and train decomposition and collaboration from the same terminal outcome, limiting transfer and obscuring credit assignment. We introduce DeOrch, which separates worker-agnostic planning from concrete worker selection. Its two-stage planner first decomposes the task without worker information, then chooses collaboration operations using compact, worker-identity-free matchability feedback from the pool, enabling conditional credit assignment to decomposition and collaboration decisions. A lightweight matcher estimates worker suitability from behavior on a fixed probe set and adapts online with a contextual bandit, allowing new workers to be incorporated without retraining the planner or matcher. Across diverse in- and out-of-distribution tasks, DeOrch outperforms prior automatic MAS orchestration methods with fewer worker calls than competing learned orchestrators, remains effective when transferred to an entirely unseen worker pool without retraining, and shows consistent gains from both components. --- ### Global Transport Couplings for Classifier-Free Guided Flows Authors: Katarina Petrović, Zander W. Blasingame, Danyal Rehman, İsmail İlkan Ceylan, Michael Bronstein, Stephen Y. Zhang, Lazar Atanackovic, Alexander Tong Published: 2026-10-06 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07555v1 Optimal-transport couplings have been shown to reduce training variance in unconditional flow models, but their role in conditional generation remains unclear. A natural approach constructs separate couplings for each condition, but this is impractical for large or continuous conditioning spaces found in modern image foundation models. We introduce Global Transport (GT), a global class-agnostic optimal-transport coupling, computed without class labels. GT can associate different conditions with different regions of the source noise, and consequently worsens performance without guidance. However, when combined with classifier-free guidance (CFG), GT consistently improves generation across domains, model scales, and sampling budgets. This reversal suggests that couplings for conditional flows should be evaluated both empirically and theoretically under the guided flow used at inference, rather than on unguided generation. We evaluate GT over both discrete class and continuous text conditioned image generation across model scales, and investigate how coupling choice alters guided trajectories. These results identify coupling design in the guided flow setting as a simple training time axis to improve performance without modifying existing architectures, samplers, or guidance mechanisms. --- ### Which and When to Admit: Gradient Admission for Data-Centric Small Language Model Finetuning Authors: Hongyu Cao, Yanchi Liu, Kunpeng Liu, Xujiang Zhao, Wei Cheng, Zhengzhang Chen, Yanjie Fu, Haifeng Chen Published: 2026-10-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.07553v1 LoRA fine-tuning adapts small language models (SLMs) to heterogeneous instruction data within a low-rank update subspace, making it vulnerable to three structural problems: conflicting gradients that cancel, static data selection that cannot track evolving learning dynamics, and subspace saturation that causes later updates to overwrite useful directions. We argue that effective adaptation therefore requires controlling which data-induced gradients enter the LoRA subspace and when. We propose GRADE (GRadient-Aligned Data-centric rEcipe), a data-centric framework combining two mechanisms: a state-aware selector that continually admits samples aligned with the evolving multi-task gradient field, and a self-calibrating step-level gate that rejects updates likely to cause destructive overwrite near saturation. Across three current-generation backbones and a heterogeneous seven-dataset instruction pool, GRADE outperforms strong data-selection and PEFT-stabilization baselines in accuracy and robustness. It is the only method to improve consistently over standard LoRA on every architecture, while producing more coherent gradient trajectories and less destructive overwrite. These results show that successful SLM adaptation depends not only on which data are selected, but also on which gradients are allowed to enter and persist in the constrained update subspace. --- ### Is $\sqrt{d}$ Separation Necessary for Gradient EM to Learn Gaussian Mixtures in High Dimensions? Authors: Yiran Zhang, Mo Zhou, Weihang Xu, Maryam Fazel, Simon S. Du Published: 2026-10-06 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2610.07551v1 Learning Gaussian mixture models (GMMs) using the Expectation-Maximization (EM) algorithm and its gradient-based variants is a fundamental problem in machine learning. It is known that randomly initialized (gradient) EM fails to learn multi-component GMMs in the exact-parameterized setting, where the number of components matches that of the ground-truth GMM. Recently, global convergence of gradient EM has been established in the over-parameterized setting, where more components are used, provided that the ground-truth components are well separated. In particular, the minimum separation between ground-truth components is required to scale as $Ω(\sqrt{d})$, where $d$ is the dimension. In this paper, we show that this dimensional dependence is unavoidable in high-dimensional settings. Specifically, we consider a hybrid EM algorithm that uses standard EM updates for the mixing weights and gradient EM updates for the component means. For any $ε> 0$, we prove that when the dimension is sufficiently large, in the worst case a separation of order $Ω(d^{0.5-ε})$ is insufficient to guarantee global convergence of population gradient EM in sub-exponential time under random initialization, even in the over-parameterized regime. Our result establishes an almost optimal worst-case lower bound on the ground-truth separation required for learning Gaussian mixtures via gradient EM in high dimensions. --- ### Foundation Model-Aided Multi-Agent Reinforcement Learning for Wireless Random Access Network Optimization Authors: Myeung Suk Oh, Zhiyao Zhang, Alvaro Velasquez, Nathaniel D. Bastian, Jia Liu Published: 2026-10-06 Categories: cs.LG, cs.AI arXiv: https://arxiv.org/abs/2610.07550v1 Random access (RA) is one of the most foundational medium access control (MAC) layer scheduling schemes for handling unpredictable data traffic from multiple terminals. While multi-agent reinforcement learning (MARL) has been explored to optimize RA-based wireless networks, its reliance on experience-driven, distributed policy learning incurs significant training overhead for each optimization task, limiting its feasibility in real-world applications. In this work, we propose to leverage a foundation model (FM) to improve MARL efficiency across diverse RA network optimization tasks. Specifically, we design an FM-aided actor-critic algorithm within a consensus-based decentralized MARL architecture and provide its convergence analysis under local reward exchanges and nonlinear value function approximations to show that our algorithm achieves the same convergence order as the conventional MARL with critic model exchanges and linear approximations. Our numerical results show that our FM-based approach significantly enhances MARL speed for RA network optimization. --- ### Quality-Aware Self-Correcting Speech Translation on an Edge Device Authors: Zubair Ajmal Farooq, Diptesh Kanojia Published: 2026-10-06 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.07545v2 We present a fully offline speech-to-speech translation pipeline that runs on a Jetson Nano (4 GB) and corrects its own weak translations without retraining. A Whisper-tiny ASR feeds an Opus-MT translator; multilingual BERT cosine similarity acts as a Quality Estimation (QE) gate, triggering a secondary-pass when confidence falls below a pre-defined threshold $τ$. We compare three correction methods: QE reranking (M1), Minimum Bayes-Risk decoding (M2), and constrained beam search (M3). On 1,012 FLORES-200 sentences (English-Spanish), M2 at $τ=0.90$ produces statistically significant improvements over greedy decoding on BLEU (+0.67, p<0.001), ChrF (+0.51, p<0.001), and COMET (+0.0020 at N=3, p=0.002); M1 yields no significant gains, and M3 is significantly worse than baseline (p>0.99). Our central finding is that QE functions effectively as a gate but poorly as a ranker: removing the QE model from candidate selection (M1$\to$M2) does not hurt quality and frees 680 MB from the critical path. Using a gain-to-edit ratio adapted from the post-editing-effort literature, we further show that smaller candidate pools (N=3) yield more surgical corrections with better semantic adequacy, while larger pools (N=10) maximise lexical reward. We release the system and demonstrate live translation across six language pairs. --- ### A Systematic Investigation of Bias in Large Language Models for Advertising Relevance Authors: Weiwei Wang, Yinchuan Xu, Jialu Gao, Youkow Homma, Jian Jiao Published: 2026-10-06 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07544v1 Large language models (LLMs) are increasingly used to judge how well an advertisement matches a query, but the fairness of these judgments has received limited attention. We conduct a systematic study of fairness in relevance judgments made by LLMs for queries and advertisements. Our counterfactual framework examines the effects of advertiser identity and possible popularity, input language, and demographic wording. We study GPT-4o as a categorical relevance judge and a Qwen-7B model trained specifically for relevance prediction. The advertiser and language experiments use query and advertisement pairs sampled from real advertising logs. Controlled synthetic queries are used to study demographic associations in employment, housing, and credit. For both models, changing the advertiser identity or input language can alter the relevance assessment. Selected demographic comparisons also show patterns consistent with common stereotypes, particularly those involving gender and occupation. We further study mitigation during model inference and training. The results indicate that its effectiveness depends on whether advertiser information is relevant to the query and how advertiser labels are distributed in the training data. These findings can help advertising practitioners identify fairness risks and develop suitable mitigation methods for LLM relevance systems. --- ### Preserving Unstable Modes Through Inverse Dynamics in JEPA World Models Authors: Leonardo F. Toso, Yann LeCun, James Anderson, Oumayma Bounou Published: 2026-10-06 Categories: cs.LG, cs.RO, eess.SY, math.OC arXiv: https://arxiv.org/abs/2610.07540v1 Robotic systems often exhibit unstable modes, along which small perturbations and disturbances can cause unbounded growth unless corrected through feedback. Controlling such systems from high-dimensional visual observations requires representations that preserve these modes. Joint-embedding predictive architectures (JEPAs) provide a natural framework for learning such representations and their dynamics from visual data. However, we demonstrate that next step prediction combined with anti-collapse regularization does not guarantee that controllable unstable modes are preserved: the training loss can be minimized while these modes are collapsed, making stabilization from the learned representation impossible. To address this, we augment world-model training with an action reconstruction objective (i.e., an inverse dynamics loss) that encourages control-aware representations, namely, visual representations that preserve crucial features for control. We prove that exact action reconstruction makes the encoder injective on the finite-horizon reachable subspace. Thus, the encoder cannot discard any state direction reachable by an action sequence within $H$ steps. Moreover, we show that, as $H$ grows, the dominant eigenspace of the finite-horizon controllability Gramian converges to the controllable unstable subspace. We establish our theoretical results for linear systems and demonstrate empirically that our findings extend to nonlinear visual control tasks (CartPole, Walker2D, and PointMaze), highlighting the benefits of control-aware representation learning. --- ### Disentangling Models from Personas in Heterogeneous LLM Simulations Authors: Dani Roytburg, Daphne Ippolito Published: 2026-10-06 Categories: cs.MA, cs.AI, cs.CL, cs.SI arXiv: https://arxiv.org/abs/2610.07535v1 Multi-agent simulations with large language models (LLMs) often operate networks of agents with a single base model. This overlooks the inter-model effects which may dominate engagement dynamics in real-world deployments. To show this, we simulate a heterogeneous social network powered by several different base models and show that the amount of engagement an agent receives depends more on its base model than on its assigned persona. The attraction or repulsion effects of a base model strengthen dramatically when more models are added in the mix, suggesting that networks dynamics may converge to base model effects at scale. To help explain this effect, we conduct a series of content-mediating analyses, showing the predictability of base models across contexts as well as the relationship between a model's lexical patterns and an engagement-maximizing style. In light of recent developments in mass multi-agent interaction, this work underscores the relevance of heterogeneous compositions in driving the outcomes of those networks --- ### SkillFormer: Skill-Decomposed Adaptation for Audio Language Models Authors: Lee Seung-woo, Bowen Qi, Kim Min-jun, Jang Won-young Published: 2026-10-06 Categories: cs.SD, cs.LG arXiv: https://arxiv.org/abs/2610.07533v1 Audio language models must handle dozens of distinct skills, from pitch comparison and speaker counting to musical tempo estimation and emotion recognition. Joint training on all skills at once causes interference: gains on one skill often come at the cost of another. We propose \textbf{SkillFormer}, which decomposes audio understanding into skill-specific low-rank adapters and composes them at inference time through a learned router. The router examines the question to decide which adapters to activate and how much weight each should carry, so that a pitch query engages different parameters than a genre classification query. An alternating training schedule updates each adapter on its own skill cluster before jointly calibrating the router, preventing the gradient conflicts that arise in standard multi-task optimization. SkillFormer adds fewer than 4\% of the base model's parameters and requires no changes to the audio encoder or language backbone. Evaluated on three architecturally distinct models across MMSU, MMAU-Pro, and MMAR, it raises the average accuracy by 2.5 to 4.1 points, with balanced gains across perception, reasoning, and semantic subcategories. --- ### Geometry-Aware Diffusion Approximate Posterior Sampling for Sparse-View and Limited-Angle CT Authors: Honglei Brinkmann, Carola-Bibiane Schönlieb, Ander Biguri Published: 2026-10-05 Categories: eess.IV, cs.LG arXiv: https://arxiv.org/abs/2610.08866v1 Sparse-view computed tomography (CT) reduces radiation dose and acquisition time and may mitigate motion artifacts. However, angular undersampling provides insufficient information to determine the image uniquely and stably. Limited-angle CT, arising from restricted angular coverage, produces strongly directional information loss associated with the missing angular range. In both settings, image directions may be strongly observed, weakly constrained, or unobservable, leading to severe ill-posedness and reconstruction ambiguity. Existing diffusion-based approaches incorporate measurement information through likelihood guidance, data-consistency operations, or range-null-space corrections. However, they do not generally use the continuously varying measurement sensitivity of the acquisition to jointly shape both reconstruction updates and stochastic exploration. We propose a geometry-aware diffusion-guided stochastic reconstruction framework for sparse-view and limited-angle CT. Its central component is a regularized noise-weighted pullback metric constructed from the CT forward operator and measurement-noise covariance. This metric continuously adapts both the measurement-aware update and stochastic exploration according to directional measurement sensitivity, suppressing changes along strongly constrained directions while permitting greater exploration along weakly constrained and unobservable directions. We complement this geometry-aware update with a regularized data-consistency correction and approximately null-space-restricted stochastic perturbations, implemented matrix-free using forward and backprojection operations together with conjugate-gradient solves. Experiments on sparse-view, noisy, and limited-angle CT demonstrate competitive reconstruction quality, strong measurement consistency, and spatially resolved empirical uncertainty estimates. --- ### Adversarial RL for Port-Scan Evasion: Attacker Feature Visibility in Edge-Deployed IDS Authors: Logan Andrew North, Priya Sanjay Kaluskar, Shasi Kumar Ramachandran Prabhu, Peilong Li, Suman Saha Published: 2026-10-05 Categories: cs.CR, cs.LG arXiv: https://arxiv.org/abs/2610.08864v1 Machine learning-based intrusion detection systems (IDS) are increasingly used in resource-constrained Internet of Things (IoT) environments, yet their robustness is often evaluated against static attacks rather than adversaries that adapt to detection feedback. This paper investigates adaptive port-scan evasion against ML-based IDS models deployed on a Raspberry Pi 3B+. We implement a live Zeek-based IDS pipeline with XGBoost, a multi-layer perceptron, and a 1D convolutional neural network trained on TON_IoT telemetry, and use a Deep Q-Network (DQN) adversary to learn evasive combinations of probe timing, TCP flags, and payload size under black-box, gray-box, and white-box feature-visibility settings. Although the deployed IDS models detect conventional port scans at 91.1--99.8%, DQN final-50-episode evasion rates range from 61.9% to 98.3% across feature-visibility settings. Greater feature visibility does not monotonically improve evasion, and its effect is model-dependent: against XGBoost, the black-box agent achieves 92.9% evasion, compared with 61.9% and 76.9% for gray-box and white-box agents, respectively, whereas 1D-CNN is most vulnerable under white-box access at 98.1%. Because standard DQN can overestimate action values, we additionally spot-check representative conditions using Double DQN. The gray-box condition remains unstable in this check, providing no evidence that overestimation bias alone explains the observed instability. These results show that limited feature knowledge can still enable effective adaptive evasion against static edge-deployed IDS models, motivating more robust defenses for IoT edge environments. --- ### CIRRA: Dual-Level Continual Instruction Reconciliation with Ongoing Execution for Embodied Robot Agents in Interactive Household Tasks Authors: Ci Zhang, Enfu Nan, Arman Akbari, Lin Zhao, Li Wang, Chen Wang, Weiwei Chen, Yanzhi Wang, Geng Yuan Published: 2026-10-05 Categories: cs.RO, cs.CV arXiv: https://arxiv.org/abs/2610.08862v2 Household robots must accommodate new user instructions while executing ongoing tasks. Existing agents often regenerate or extensively revise the remaining task sequence, introducing plan ambiguity, logical inconsistency, and redundant execution. We formulate continual instruction reconciliation and propose CIRRA (Continual Instruction Reconciliation for Robot Agents), a dual-level framework combining LLM-based semantic reasoning with rule-constrained structural integration. CIRRA first grounds incoming instructions to unique executable skills and resolves underspecified actions and execution locations. It then preserves the ongoing subtask sequence as an execution backbone and generates integration candidates by inserting incoming subtasks into location-matched segments. The semantic reasoner evaluates only modified segments to identify dependencies and conflicts and select the most logically coherent local integration. This structure-preserving process maintains alignment with ongoing execution, mitigates ambiguity and inconsistency, and reuses shared subtasks to reduce redundant execution. We also introduce CHIRP (Continual Household Instruction Reconciliation and Planning), a text-based benchmark of 120 episodes across eight household environments and six categories of everyday activities. On CHIRP, CIRRA achieves 74.2% decision agreement, exceeding the strongest replanning baseline by 30 percentage points; every correct fusion decision yields a correctly placed, conflict-free schedule. On a Unitree G1 humanoid, CIRRA interrupts ongoing skills at the correct moment in every trial and significantly outperforms all baselines on every metric. --- ### LRCC: Generalizing Low-Rank Compression with Conditional Computation Authors: Thomas Vaitses Fontanari, Maximo Eduardo Rulli, Federico Alvetreti, Donatella Genovese, Simone Scardapane Published: 2026-10-05 Categories: cs.CL, cs.AI, cs.LG arXiv: https://arxiv.org/abs/2610.08858v1 Low-rank compression reduces the cost of pretrained language models by replacing linear transformations with low-rank factorizations. However, conventional methods use a fixed rank allocation during inference, assigning the same amount of compute regardless of the input token. We introduce Low-Rank Conditional Computation (LRCC), which adds token-dependent computation to pretrained models by training one lightweight router per Transformer block to select among a small set of nested low-rank paths. During training, the low-rank factors remain frozen, and only the routers are optimized. We evaluate LRCC on Llama and Qwen models for language modeling and zero-shot downstream tasks. Within the same average active-parameter budget, LRCC improves the predictive performance over static low-rank compression, including a 7.6 percentage-point gain in average downstream accuracy on Llama-2-7B over static methods. At matched batch-size-1 decoding latency, LRCC improves both perplexity and downstream accuracy on Llama-3.2-1B and remains competitive on Llama-2-7B, without specialized kernels. Finally, we assess the usefulness of assigning a token-wise path by analyzing the routers' path choices. --- ### Climbing the Design Ladder: Sequential Knowledge Distillation for Early-Stage Circuit Timing Prediction Authors: Reza Moravej, Fahad Rahman Amik, Zhanguang Zhang, Didier Chételat, Yingxue Zhang Published: 2026-10-05 Categories: cs.LG, cs.AR arXiv: https://arxiv.org/abs/2610.08457v1 Integrated circuit design involves multiple design stages: logic synthesis, floorplanning, placement, and routing, with each stage taking hours to weeks to complete. Discovering timing violations late in this flow forces costly iterations back to earlier stages, wasting computational resources and delaying product launches. While predicting post-routing timing from early-stage data could prevent these failures, existing machine learning approaches struggle with the massive abstraction gap between post-synthesis logical descriptions and post-routing physical layouts. We propose STEP-KD (Sequential Timing Evaluation via Progressive Knowledge Distillation), which leverages intermediate design stages as ``stepping stones'' for progressive knowledge transfer rather than attempting direct prediction. STEP-KD trains teacher models at the post-routing, post-placement, and post-floorplan stages, then sequentially distills their knowledge to a post-synthesis student model through representation alignment. Experiments on diverse circuits demonstrate that STEP-KD reduces timing prediction error compared to direct distillation and supervised baselines, and in most settings compared to the industry-standard Static Timing Analysis (STA) tool. STEP-KD reduces the weighted mean absolute percentage error of Total Negative Slack prediction to 19.78\%, compared with 74.84\% for STA. Our proposed method is step forward to identify timing problems earlier, avoiding expensive late-stage redesigns. --- ### Safeguarding LLMs via Model-Agnostic Latent Safety Signals from Dark Knowledge Authors: Wonjun Lee, Kyungsik Yang, Gaeun Ji, Vaidehi Patil, Haon Park, Bumsub Ham, Mohit Bansal, Suhyun Kim Published: 2026-10-05 Categories: cs.CR, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.07532v1 LLMs have advanced rapidly, raising growing concerns about their safety. Recent work has proposed approaches to detect and defend against attacks including defenses at decoding stage that leverage models' hidden states. However, existing decoding-stage defenses suffer from two limitations. First, they introduce a trade-off between safety and over-refusal, where strengthening safety degrades the model's helpfulness on benign queries. Second, many of these methods rely on internal hidden states and are thus restricted to specific architectures, incurring substantial overhead and limited generalization across models. To address these limitations, we introduce LADE (Latent Safety Signals for Defense), which leverages latent safety signals extracted by contrasting harmful and benign queries from dark knowledge (i.e., information carried by the output probability distribution beyond its argmax) in the first-token output probability distribution. Our key insight is that, beyond surface-level refusal tokens, the dark knowledge in the first-token distribution contains latent safety signals, defined as tokens whose probabilities differ sharply between harmful and benign queries. We show that these signals consistently align across LLMs, forming a model-agnostic direction that emerges from safety alignment. LADE consists of three components: (1) Extracting Latent Safety Signals from Dark Knowledge, which selects top-k safety-discriminative tokens from the first-token probability distribution; (2) Tokenizer Mapping, which maps these tokens across different tokenizers to enable model-agnostic application; and (3) kNN-based Discrimination, which classifies queries via a k-Nearest Neighbors search over the mapped tokens. Across diverse LLMs and benchmarks, LADE is robust against a wide range of jailbreak attacks and lowers attack success rates while maintaining a competitive safety-utility trade-off. --- ### Targeted search shows that random-device testing underestimates worst-case error in a simulated wave-based neural operator Authors: Samrendra Roy, Jason Yoo, Souvik Chakraborty, Syed Bahauddin Alam Published: 2026-10-05 Categories: cs.LG, cs.ET, physics.optics arXiv: https://arxiv.org/abs/2610.07529v1 Wave-based processors promise fast, energy-efficient Fourier layers for neural operators. They are usually validated on randomly sampled devices, but using them requires knowing how large their error can become under fabrication and alignment variation. In a stylised numerical case study, a hybrid Fourier neural operator runs its four spectral layers on simulated coherent 4f processors with 32 toleranced knobs, whose half-widths are representative rather than calibrated. For 120 models (four tasks, six training methods, five seeds), we compared the worst of N random in-spec devices with a searched one. On a deterministic simulator with one frozen draw of the random static errors, the searched device's held-out error was 1.08-3.10 times the maximum over 200 Monte Carlo devices and 1.06-2.71 times that over 1000. With 20 fresh static draws, it still exceeded the maximum over 200 random devices in 116 of 120 models. Under uniform sampling, the probability of drawing such a device is at most 0.37% per model (two-sided 95% Clopper-Pearson), which says nothing about how large its error is. The gap persisted with uniform or Sobol' sampling at the search's budget, shared knobs, a second crosstalk model, box scales of 0.25-2 and a pixel-level device model. Models trained only with random static errors reached 3.7-39.9 times their nominal error on searched devices, and fine-tuning on random and gradient-searched devices gave the lowest searched error of the six in all 20 task-seed pairs. For two heat-exchanger quantities, a search targeted at each exceeded the worst of 1000 random devices in all 39 models, and hence the Wilks 95/95 limit (worst of 59). For the mean pressure of 11 models, no random device exceeded a 1% error threshold, but the searched device did. Random testing estimates how often errors exceed a threshold; worst-device search gives a lower bound on how large they can be. --- ### Activation Denoising: A Robustness View on Parallel vs Sequential LLM Quantization Authors: Yan Scholten, Rachel Lawrence, James Hensman, Stephan Günnemann, Alicia Curth, Riccardo Grazzi Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07522v1 Post-training quantization is a powerful tool for compressing large language models. The most scalable methods quantize every layer in parallel, but quantization errors then compound through the residual stream, as no layer corrects for the errors of the layers before it. Sequential quantization accounts for this error compounding by re-calibrating each layer on the already-quantized outputs of its predecessors, yielding stronger results but at the cost of a serial schedule that becomes a bottleneck at scale. As a solution, we propose parallel quantization with activation denoising, which recovers much of the sequential benefit while keeping quantization fully parallel. Rather than re-calibrating layer-by-layer, we take a robustness perspective and model the upstream error as noise, regularizing to be robust to it through a preprocessing step followed by metric-weighted rounding. Applied at every layer, this regularization forms a depth-compounding smoothness penalty that dampens how strongly quantization errors amplify through the model. Unlike orthogonal rotations commonly used in quantization, which must preserve the model's function, we multiply the weights by a more general linear transformation. We find that the two are complementary and their effects compound. Empirically, our robustness regularization recovers a significant part of sequential quantization's benefit in a single parallel pass, at a fraction of its time. Overall, by treating compounding quantization errors as a robustness problem, we offer a principled foundation for more efficient and accurate LLM quantization at scale. --- ### Grounding What Shapes the Plan: Rethinking Groundedness for Physical Intelligence in Autonomous Driving Authors: Minkyoung Cho, Zewei Zhou, Wenhao Ding, Shuhan Tan, Boyi Li, Yuxiao Chen, Yan Wang, Zheng Lian, Min-Hung Chen, Chaowei Xiao, Zhuoqing Mao, Boris Ivanovic et al. Published: 2026-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07521v1 Driving models increasingly ground reasoning in causal relations, spatial structure, perceptual evidence, and predicted futures. These advances make reasoning more faithful to the driving scene, but leave a fundamental question unresolved: what should groundedness mean when the model ultimately outputs an action? Correctly grounded reasoning does not, by itself, ensure desirable driving outcomes. We introduce GroundAct, which starts from a simple premise: driving unfolds through physical entities and their interactions. Entities therefore become the unit of grounding; a lightweight reference token keeps each selected entity's continuous state addressable through symbolic reasoning; and only the referenced entities' interactions with the evolving proposal correct the plan. The result is an explicit path from what reasoning grounds to what the plan does, which we call grounded planning. To assess its practical value, we evaluate GroundAct in both open- and closed-loop settings. GroundAct shows strong open-loop planning across normal, out-of-distribution, and safety-critical scenarios, with closed-loop results extending this evidence to driving in simulation. --- ### Not What a Child Expressed: Auditing the Sign-to-Text Safety Interface in Child-Facing AI Authors: Muhammad Rafiullah Memon, Viet Vo, Wanlun Ma, Yang Xiang Published: 2026-10-05 Categories: cs.CL, cs.CR, cs.CY arXiv: https://arxiv.org/abs/2610.07519v1 Automatic sign language translation (SLT) has entered consumer products, turning American Sign Language into English text for dictation, messaging, and queries put to a conversational assistant. Child-facing AI and platform trust-and-safety tooling decide on text, using filters on minor accounts and grooming classifiers that score chat messages. A signing child who uses SLT therefore reaches these safeguards through a translation. We found no publicly documented system in which the two have been jointly evaluated, and the leading deployed SLT model was neither trained nor formally evaluated on signers under 18. Errors that alter negation, participant roles, secrecy, urgency or help-seeking could change a safety decision without disturbing fluency. This paper proposes a Deaf-informed pre-deployment audit of that boundary, with a failure taxonomy, a sanitised scenario schema, four comparison conditions, and four outcome measures. Auslan is the planned first case study. --- ### Harmful SFT Leaves a Continuous Trace in LLM Checkpoint Updates Authors: Ziqun Bao, Xinyu Zhang, Yuchen Shao, Chengcheng Wan Published: 2026-10-05 Categories: cs.LG, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2610.07518v1 Safety auditing of post-trained large language models typically relies on model behavior, requiring model execution and depending on the coverage of available evaluations. This work asks a different question: Do the target behaviors optimized during supervised fine-tuning (SFT) leave readable evidence directly in checkpoint updates? We find that harmful-compliance SFT induces a continuous, objective-dependent ordering in checkpoint-update space. Using a reference geometry defined by pure harmful-compliance, safety-targeted, and benign-utility SFT, we find that a checkpoint-level coordinate s_H tracks controlled harmful-objective composition with Spearman correlations of 0.986-0.992 across four 7-8B backbones, with the same ordering persisting at larger model scales. Matched compliance-versus-refusal controls show that this checkpoint trace reflects the SFT objective rather than harmful-input exposure, while additional controls rule out simple explanations based on harmful-example count or generic training intensity. Building on this structure, we introduce TRACE, a weights-only auditing method that localizes an unknown checkpoint update relative to frozen harmful and non-harmful reference prototypes and converts this geometry into a continuous harmful-objective score. TRACE requires neither model queries nor access to the unknown SFT data, and can be evaluated directly from checkpoint updates. Across distribution shifts, unseen data, different SFT configurations, partial checkpoint access, and LoRA/full-parameter fine-tuning, the trace remains stable and is positively associated with independently measured attack success rates. TRACE remains informative even at low harmful-objective proportions, providing a complementary auditing signal when behavioral evaluation is unavailable or incomplete. Code is available at https://anonymous.4open.science/r/Code4TRACE-54D3. --- ### From Local Evidence to Safety Verdicts: Causal Tracing in Vision-Language Models Authors: Faezeh Dehghan Tarzjani, Mevan Wijewardena, Alexander Romanus, Sampad Mohanty Published: 2026-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07514v1 A vision-language model may need to combine an image with a prompt to recognize a safety risk that neither reveals alone. Where does this joint safety judgment become accessible inside the model? We introduce SSU-Bench, a dataset of matched safe and unsafe image-text combinations constructed using single-item prompt edits or image edits with annotated intended regions. Using three vision-language models, we transfer internal states between paired inputs and measure the resulting change in the safety verdict. Across models and both types of counterfactual, interventions at the changed input positions are effective in earlier decoder layers, while interventions at the final input token become effective later. Directions estimated from other examples produce similar late-layer effects. A linear readout of the final-token state also predicts the model's own verdict, including incorrect judgments, and cross-model comparisons reveal similarities in the patterns of counterfactual change. These findings identify a recurring transition in where interventions can influence a joint safety verdict and distinguish a readable model decision from a correct safety judgment. --- ### MobileVISTA: Generative Data Augmentation for Pose Generalization in Mobile Manipulation Authors: Suzannah Wistreich, Stephen Tian, Isabella Huang, Vitor Campagnolo Guizilini, Sergey Zakharov, Katherine Liu, Jiajun Wu Published: 2026-10-05 Categories: cs.RO, cs.CV, cs.LG arXiv: https://arxiv.org/abs/2610.07511v1 Mobile manipulators such as humanoid robots are increasingly deployed in dynamic, unstructured environments to perform dexterous manipulation tasks. However, end-to-end manipulation policies trained to imitate demonstration data collected from a single robot pose are brittle: even centimeter-scale deviations in robot pose at deployment can drive ego-centric observations and end-effector trajectories out of the training distribution, leading to sharp drops in performance. We introduce MobileVISTA, a data generation framework that transforms demonstrations captured at canonical poses into diverse, pose-perturbed training data by jointly (1) augmenting egocentric visual observations and (2) retargeting actions to compensate for base pose changes. Unlike prior methods, which assume a camera rigidly mounted off the actuated chain or non-trivial articulated robot geometry largely out of frame, MobileVISTA targets compatibility with egocentric platforms (e.g., humanoids) where the camera is both influenced by and must observe the robot's kinematic chain as it moves. We study MobileVISTA in simulated tasks spanning humanoid and bimanual embodiments, and on a real Galaxea R1 Pro. We find policies trained on MobileVISTA-augmented data demonstrate improved robustness to previously out-of-distribution poses encountered at test time, without additional demonstration collection or a trained generative model. Additionally, we find MobileVISTA's benefit is largest on tested humanoids, where the camera rides the actuated chain and the robot fills much of the frame. Additional videos and appendix can be found on our website: https://mobilevista.github.io --- ### On Open-Ended Information Seeking for Information Elicitation Agents Authors: Victor De Lima, Grace Hui Yang Published: 2026-10-05 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.07509v1 Information elicitation is an open-ended information-seeking problem in which an interaction can unfold in many potentially valuable directions, requiring an elicitor to continually determine which information to pursue as new information emerges. In agentic elicitation, these decisions may be delegated to a foundation model, yet how model choice shapes the resulting information-seeking behavior remains understudied. We study how judgments about information value vary across LLMs and how these differences shape sequential information seeking. We first examine these judgments across 11 LLMs spanning multiple model families and parameter scales, using a shared set of information and elicitation objectives. We then develop a controlled elicitation simulation in which different models encounter the same information space and use the same selection rule, isolating these judgments from question generation and respondent behavior. Using this setting, we characterize the breadth-depth behavior that emerges from model-specific information-seeking preferences over the course of elicitation. We further examine how interaction history changes the evaluation and subsequent selection of prospective information. We test the robustness and boundaries of these findings through sensitivity analyses and ablations over the opportunities available to the elicitor, the response labels used to operationalize information-seeking preferences, the presence of interaction history, and whether redundancy is explicitly relevant to the assessment. The project code, data, and trajectory files are available at https://github.com/infosenselab/open-elicitation. --- ### Jarvis: A Proactive Speech Agent for Multi-Party Conversations Authors: Seunghyun Oh, Hirotaka Hiraki, Shuyue Stella Li, Yulia Tsvetkov, Shyamnath Gollakota Published: 2026-10-05 Categories: cs.HC, cs.AI arXiv: https://arxiv.org/abs/2610.07506v1 Speech agents are reactive and dyadic: they speak when spoken to, and to one person at a time. We ask what it takes for a speech agent to instead take part in a conversation among several people and speak up only when it can help. We introduce Jarvis, a real-time proactive speech agent that audibly participates in multi-party human conversations. Grounded in a document shared beforehand, Jarvis follows the discussion and intervenes when the group misses or misstates a fact and does not correct itself within a few turns. We make three contributions: a problem setting based on epistemic breakdowns that makes proactive intervention measurable, realized as CHI-180-proactive, a synthetic multi-party dataset seeded with known gaps, errors, and self-corrections; a proactive backbone that harnesses a small, open-weight model with deterministic checks and grounds every claim in a source sentence; and interaction techniques for taking the floor in live speech and showing the cited evidence on screen. On CHI-180-proactive, Jarvis is correct on most events it addresses and stays silent 97% of the time when the group resolves an issue itself. A live study with 23 participants confirms these trends with real-time interventions. --- ### MARS: Multi-resolution Adaptive Routing for Sequential Recommendation Authors: Ming Yin, Sixun Dong, Yudong Liu, Wen-Yun Yang, Yunjiang Jiang, Yiran Chen Published: 2026-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07505v1 Long-history recommenders often compress each user's history into a compact, candidate-independent memory that is cached and reused to score large candidate pools. We show that real user histories exhibit multi-scale semantic structure, with short-lived intent, medium-term interests, and long-term preferences coexisting in one sequence, and that monolithic cached memories preserve these scales unevenly: linear probes recover recent and mid-range content far worse than long-range content. We call this failure mode \textit{temporal aliasing}. We propose \textbf{MARS}, a multi-resolution user memory that writes the full history into recurrent state tracks anchored to different half-lives, and a sparse routing reader that materializes compact seed memories by selecting the relevant temporal resolutions for each seed, preserving fixed-size candidate scoring. MARS outperforms strong baselines on three public datasets, with gains that grow with history length. Component-matched ablations with paired tests show that temporal diversity and selective routing each contribute beyond what hard-window memories or added capacity provide. The advantage of MARS over its interface-matched baseline also widens after within-user behavioral shifts, at about $1.02\times$ that baseline's warm-cache serving latency for $1{,}000$ candidates per user. --- ### Two-Sample Testing for Random Graphs without Vertex Correspondence Authors: Soham Dan Published: 2026-10-05 Categories: stat.ML, cs.LG arXiv: https://arxiv.org/abs/2610.07503v1 Two populations of graphs often have to be compared without any correspondence between their vertices, for instance when networks come from different communities, or when a graph generative model is evaluated against held-out graphs. We study how many graphs such an unaligned two-sample test needs, and which graph statistics can detect which differences. For an Erdős--Rényi null and a planted two-block difference that leaves every expected degree unchanged, we show that $m\asymp t^{-3}$ graphs per group are necessary and sufficient when the per-graph signal-to-noise ratio is $t<1$. Signed triangle counts attain this rate, and the lower bound holds for every graph size. With aligned vertices $m\asymp t^{-1}$ graphs suffice, so misalignment costs a factor of order $t^{-2}$. When the triangle signal cancels, the rate becomes $t^{-4}$ and $4$-cycles are needed. Statistics built from trees have exactly the same expectation under both hypotheses, and tests based on finitely many of them have asymptotically no power. In the graphon limit, this class includes degree distributions and message-passing graph neural network features. For a non-constant null, a generic difference is visible at first order, and a simple motif test attains the aligned order of sample size, suggesting that misalignment is costly mainly for differences that are invisible at low orders. We also give an exactly valid test for one or two graphs per group, at a cost in power. In our simulations, the fitted exponents are close to the predicted ones, and degree-based and random-GNN evaluation metrics stay at their level in a setting where signed triangles need about $65$ graphs. --- ### Closing Ambient Clinical Documentation Gaps with Automated Provider Queries Authors: Joseph Paul Cohen, Raj Shah, Han-Chin Shing, Fang Wang, Susan Nguyen, Chaitanya Shivade, Jack Moriarty Published: 2026-10-05 Categories: cs.CL arXiv: https://arxiv.org/abs/2610.07502v1 Provider queries are clarifying requests sent by clinical documentation specialists to physicians to close gaps in the clinical note and ensure accurate billing. Prior work automates note drafting, ICD-10 coding, and order extraction assuming a complete transcript, leaving these gaps unaddressed. We study whether an LLM can automate the query loop, termed DAU (Draft, Ask, Update), across those three tasks. An audit of 3,000 real visits identifies the sources of missing documentation, from which we build five transcript-degradation benchmarks on public data. Analyzing 21k clarification turns on real conversations, we find useful-question predictors are task-specific: oracle confidence dominates, but note completeness needs only simple recall questions while ICD-10 coding needs harder, multi-option ones. About 9% of turns hurt performance, driven by redundant questions and non-answers that still trigger a rewrite. Deployment depends on learning "when not" as much as "what to" ask. --- ### Source-Learned Reliance for Selective Test-Time Adaptation of Multimodal Time Series Authors: Payal Mohapatra, Yueyuan Sui, Haodong Yang, Benjamin Lundell, Stephen Xia, Qi Zhu Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07499v1 Multimodal wearable systems must remain reliable when sensor streams become noisy or unavailable. Existing multimodal test-time adaptation (TTA) methods often assess reliability online, but cross-modal agreement can be misleading when sensors measure different physical processes, and evaluating alternative modality configurations adds inference cost. We propose CARAT, which decouples model reliance from runtime corruption detection to guide omission or attenuation, amortizing reliance estimation through source training. An asymmetric modality-dropout curriculum prepares a missingness-resilient backbone for omission and derives a frozen, backbone-specific reliance proxy from windowed input-projection gradient norms. At deployment, a lightweight one-class detector flags suspect streams, and the proxy guides a joint choice between replacing the suspect set with the backbone's trained missingness symbol and attenuating its representations before fusion, without candidate-subset evaluation. Across four wearable datasets, five corruption types, three backbones, and eight TTA baselines, CARAT achieves the highest overall macro-F1 and best mean rank (2.42), exceeding EATA, the strongest baseline, by 1.58 F1 points across 12 equally weighted dataset-backbone settings. Across five profiled configurations, CARAT uses 9.49% fewer GFLOPs and updates 47.82% fewer parameters than EATA. A pattern also emerges across sensing regimes: multimodal TTA methods such as PTA are competitive on IMU-dominated homogeneous datasets, whereas unimodal TTA methods like TENT and EATA match or exceed it on heterogeneous datasets. These results position CARAT as a practical default to wearable TTA, offering competitive robustness with modest computational requirements and benefits that vary across backbones and dataset regimes. --- ### Does Muon Need Fine-Grained Spectral Shaping? Authors: Meher Chaitanya, Tianyi Zhou, Aristides Gionis Published: 2026-10-05 Categories: cs.AI, cs.LG, stat.ML arXiv: https://arxiv.org/abs/2610.07497v1 Muon combines current and past gradients into matrix momentum. For $M=UΣV^\top$, the idealized polar update $Q=UV^\top$ gives every singular direction the same weight. We refer to this as the flat profile. Several recent optimizers replace this flat profile with fine-grained spectral maps that give each direction its own gain. We ask how much of this spectral detail a Muon update needs. Our spectral diagnostics show that approximately $94$--$97\%$ of measured singular modes lie below an estimated noise edge, yet collectively align positively with a reference gradient. We introduce BulkBoost, a two-band spectral reweighting framework with fixed-rank and noise-calibrated variants. The latter uses split-minibatch gradient differences to calibrate a Marchenko--Pastur reference edge for Muon's Nesterov input, separating the bulk below the edge from the spikes above it. Both variants increase the bulk's relative weight through one shared gain while preserving the Frobenius norm of each matrix's unreweighted direction. For a fixed partition, our theory gives the first-order condition under which moving weight toward the bulk lowers the loss. It also quantifies the fraction of the maximal first-order improvement rate, over all per-mode reallocations, that two bands can capture. Across 30 continued-pretraining settings spanning Pythia-14M to 410M and six corpora, two-band reweighting is competitive with the fine-grained power-law profile of Freon and outperforms Spectra. Measured against Muon's flat profile, Freon reduces final loss by $0.022\%$ of the pre-adaptation loss on average, whereas the two-band variants achieve reductions of $0.073$--$0.147\%$. These observations suggest that useful departures from the flat profile are surprisingly low-dimensional: a single bulk-to-spike gain captures at least as much benefit as the fine-grained spectral profiles. --- ### The interface of data assimilation and machine learning Authors: Eviatar Bach Published: 2026-10-05 Categories: physics.ao-ph, eess.SY, nlin.CD, stat.ML arXiv: https://arxiv.org/abs/2610.07496v1 Data assimilation (DA) is the process of combining forecasts from a model with observations in order to optimally estimate the state of a system. This is critical for chaotic systems, such as the atmosphere, since if observations are not continually assimilated the model will quickly lose skill. DA is routinely performed (usually every 6 hours) at operational forecasting centres around the world. In this article we discuss the interface of machine learning (ML) and DA. This is still an emerging and quickly developing field, and this article tries to give an overview of some of the main topics and methods. --- ### Who Bears the Burden? Learning Responsibility for Shared Constraints in Multi-Agent Reinforcement Learning Authors: Xiaoyang Cao, Jingqi Li, Zhe Fu, Alexandre M. Bayen Published: 2026-10-05 Categories: cs.LG, cs.AI, cs.GT, cs.MA arXiv: https://arxiv.org/abs/2610.07491v1 When multiple agents share a cost budget, a common Lagrange multiplier can enforce the aggregate constraint but does not determine how its penalty should be allocated across agents. Uniform penalties ignore heterogeneity in the rewards agents sacrifice, while agent-specific multipliers may still rely on the same aggregate cost signal. We introduce Lagrangian Responsibility Allocation (LiRA), which learns each agent's share of a common multiplier by optimizing social welfare over a finite training horizon. The multiplier enforces the aggregate budget, while responsibility shares redistribute its influence without modifying the original rewards or constraints. For convex games under standard regularity conditions, varying these shares induces a smooth family of normalized generalized Nash equilibria in which active constraints remain at their budgets while welfare varies. To optimize responsibility before convergence, we derive a welfare gradient that accounts for both learning updates and the induced change in data distribution. Across CityLearn, MABIM, Harvest, and MetaDrive, spanning 3 to 400 agents, LiRA improves average social welfare by up to 29% over uniform and agent-specific multiplier baselines. Grid and driving costs remain within budget, inventory violations decrease, and Harvest makes more effective use of available budget. --- ### Deep Defence on Wheels: A Dual Intrusion Detection System Architecture for Comprehensive In-Vehicle Network Security Authors: Shashwat Khandelwal, Shanker Shreejith Published: 2026-10-05 Categories: cs.CR, cs.AR, cs.LG arXiv: https://arxiv.org/abs/2610.07489v1 Increasing connectivity to the outside world and the lack of inbuilt security mechanisms have made legacy intra-vehicular networks vulnerable to cyberattacks. Initial research focused on maximising detection accuracy for known and unknown attacks, often using large, full-precision machine learning models. However, embedding IDSs into vehicular electronic systems also requires low detection latency, energy efficiency and minimal electronic control unit (ECU) resource overhead to process about 2,000 CAN frames/s. Lightweight models must balance accuracy with these deployment constraints. We propose a dual IDS framework comprising supervised and unsupervised learning-based solutions, each optimised for real-time, resource-constrained automotive platforms. A quantised LSTM-based IDS (QLSTM-IDS) achieves over 99.9% detection accuracy for DoS/Flooding, Fuzzing and Spoofing/Malfunction attacks using a single model architecture evaluated on two widely used datasets. The model is trained using the Brevitas quantisation-aware training library, transformed into a dataflow accelerator with custom blocks compatible with AMD's FINN toolchain, and synthesised using Vitis HLS. Complementing this, an 8-bit quantised convolutional autoencoder-based IDS (QCAE-IDS), quantised using AMD's Vitis-AI toolchain, detects previously unseen anomalies that alter CAN-ID sequence patterns with over 99% accuracy. An integration architecture enables both models to operate on a single FPGA, bridging the network interface IP and processing system to minimise software overhead. QLSTM-IDS achieves 0.25 ms inference latency and 0.8 mJ energy consumption per message, while QCAE-IDS achieves 0.42 ms and 1.1 mJ per block. Both solutions are deployed and evaluated on the ZCU104 SoC (XCZU7EV FPGA), demonstrating a flexible hardware/software co-design for real-time detection of known and unknown attacks on high-speed CAN buses. --- ### Robust Importance Sampling for Rare Events via Constrained Gaussian Mixtures Authors: Paweł Lorek, Rafał Nowak, Rafał Topolnicki, Tomasz Trzciński, Maciej Zięba Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07485v1 We study estimating rare-event probabilities $I = \mathbb{P}(g(\mathbf{X}) > γ)$ with $\mathbf{X} \sim \mathcal{N}(\boldsymbolμ, \boldsymbolΣ)$ and general $g : \mathbb{R}^d \to \mathbb{R}$. We address this problem through importance sampling, and propose a framework that substantially improves efficiency and robustness over baselines such as crude Monte Carlo, adaptive cross-entropy, variational-inference-based methods (including reverse- and forward-KL approaches), as well as Safe-ICE, Subset Simulation, and Sequential Monte Carlo, drawing on ideas from both rare-event estimation and cross-entropy optimization. The key contribution has two parts: first, we separate the problem into coverage, to overcome the cold-start barrier, and fitting, to refine proposals once a meaningful signal is available; second, we constrain the final GMM proposal so that it has finite importance-sampling variance (since coverage alone is not sufficient -- without safeguards, importance sampling may still suffer from infinite variance). Together, these ingredients yield expressive proposals; finite variance does not by itself guarantee practical stability at a fixed sampling budget. Extensive experiments demonstrate substantial variance reduction, strong robustness across diverse benchmarks, and favorable cost--efficiency trade-offs, with the proposed approach often outperforming these baselines, particularly in high-dimensional and multimodal settings where competing methods frequently become unstable or fail. Our code is available at https://github.com/lorek/robust-cfi-is. --- ### SpecBraM: What Should an EEG Foundation Model Predict? Masked Band-Power Prediction versus Waveform Reconstruction Authors: Peng Xie, Yequan Bie, Jianda Mao, Kani Chen Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07484v1 Self-supervised EEG models often reconstruct masked waveforms or predict discrete codes. We study a task-aligned alternative: masked band-power prediction (MBP), which predicts fixed narrow-band log spectral energy for masked channel-time patches. This target retains rhythm power relevant to sleep staging while avoiding phase-sensitive waveform reconstruction and a learned codebook. Across three pretraining seeds, we compare band-power and waveform targets with matched backbones, pretraining data (2,388 hours), and training steps, including a 2x2 tokenizer-by-target design. On ISRUC and HMC sleep staging, MBP exceeds raw- and band-waveform reconstruction by 1.6-2.8 balanced-accuracy points with all labels and 4.7-7.3 points with 1% of labels under a strict linear probe; the target effect exceeds the tokenizer effect. Its frozen features reach 0.7916/0.7425 balanced accuracy, versus 0.7636/0.7227 for a matched rich handcrafted spectral baseline, although the gap is about one point with 1% of labels. Full fine-tuning reaches 0.8107/0.7669. The gains do not extend to every task with spectral cues, including motor imagery, depression screening, and vigilance regression. These results support choosing pretraining targets to match the physical quantities and spatial and temporal scales relevant to downstream labels. --- ### In With the Old: Enhancing 'Classical' Document Automation with Generative AI Authors: Marc Lauritsen, Hannes Westermann Published: 2026-10-05 Categories: cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.07480v1 Software-based legal assistance systems have leveraged many different forms of knowledge representation and reasoning. This article explores how document automation services rooted in expert system style and other symbolic approaches can usefully enhance and be enhanced by current generative AI approaches. We discuss the possible benefits and challenges, and report on preliminary experiments in using large language models to identify and fix issues in texts written by laypeople. --- ### Can Power Draw Constrain Covert Compute? Limits of Analogue Verification for AI Governance Authors: Tom Kimpson, Mauricio Baker, Emlyn Graham Published: 2026-10-05 Categories: cs.CY, cs.AI, cs.CR arXiv: https://arxiv.org/abs/2610.07476v1 Frontier AI treaties or agreements on limiting computation require external verification; an external auditor must be able to confirm how much computation actually ran and that parties are adhering to the agreement. Analogue, off-chip measurements such as power draw provide an information channel for verification. It is unknown how well these analogue channels can constrain computation against an adversary who actively tries to subvert the audit. We derive a closed form for $β$, the largest hidden computation a power trace cannot exclude, as a fraction of the declared machine capacity. Measurements on NVIDIA A100 GPUs constrain $β= 1.16$ in the worst case, while adversarial matched-energy strategies are shown to hide at least $β= 0.41$ of compute. Analogue power measurements alone therefore constrain compute weakly. Additional restrictions granted by the threat model, such as the ability of the verifier to re-execute the declared work at an observed operating point, let the verifier push $β$ down to $0.059$ in the maximally restricted case. This gives a quantitative estimate of what analogue measurements can contribute to compute verification. --- ### Adapting to Changes in Agent Behavior via Finite-Depth Policy Sensitivity Authors: Lan Shi, Daigo Shishika, Xuan Wang Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07475v1 Adapting a reinforcement learning policy to changes in another agent's behavior typically requires a large amount of new interaction data. Policy sensitivity provides a first-order prediction of how a locally optimal policy changes with a behavioral parameter, but its computation requires second-order derivatives whose effects propagate across future interactions. We develop a finite-depth framework to estimate this sensitivity by approximating the policy Hessian and mixed derivative using information from a reference environment. The method features an adjustable propagation depth which determines where derivative propagation along the trajectory is truncated. We characterize the derivative contributions omitted by finite-depth propagation and derive truncation-error bounds for the approximated derivatives and resulting policy sensitivity. The bounds are nonincreasing with propagation depth and vanish at full-horizon propagation. Using a belief-driven pursuit-evasion game as a validation scenario, the proposed method generally achieves lower derivative-estimation errors as the propagation depth increases and outperforms the baseline methods in both estimation accuracy and policy adaptation. The sensitivity-based initialization improves zero-shot return over direct transfer, and also shows advantages for the subsequent fine-tuning in the target environment. --- ### PsyCIDRA: A Dual-Agent Framework for Psychiatric Interviewing and Diagnostic Reasoning Authors: Milad Mohammadi, Fatemeh Akrami Shamsabadi, Zahra Mohseni, Amirhossein Safdarian, Malekfarhad Malek, Hadi Moradi, Hesham Faili Published: 2026-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07473v1 Large language models show promise in clinical reasoning, but psychiatric interviewing requires guiding an evolving conversation. Their ability to carry out this interactive assessment remains less studied. We present PsyCIDRA, a dual-agent framework linking free-form psychiatric interviewing with diagnostic reasoning for expert review. Its interviewer agent uses tools to maintain working notes, load expert-written skills, and retrieve ICD-11 references to guide inquiry. Its diagnostic reasoning agent then receives the completed interview transcript and reports hypotheses alongside supporting, conflicting, and missing evidence, withholding a final hypothesis when none is sufficiently supported. Using patient profiles generated with PsyCPG, we first evaluate PsyCIDRA in simulation. Across four models on 53 evaluation cases, it achieves higher diagnostic agreement than direct prompting. On 81 held-out simulated cases, rank-1 accuracy is 60.5% versus 51.9%. In a blinded study of 101 human participants in separate arms, PsyCIDRA agrees with psychologists on whether to propose a diagnostic hypothesis in 79.6% of cases, compared with 65.4% for direct prompting. Together, these findings support the potential of LLM agents to assist psychiatric assessment through free-form dialogue. By examining diagnostic reasoning, interview quality, and safety together, this study contributes to understanding the capabilities and limitations of psychiatric interview agents. --- ### Structure, Not Belief: Correlated Thompson Sampling from LLM-Derived Covariance in Combinatorial Semi-Bandits Authors: Vikram Kakaria, Anish Kataria, Anany Kotawala Published: 2026-10-05 Categories: cs.LG, cs.AI, stat.ML arXiv: https://arxiv.org/abs/2610.07470v1 Combinatorial Thompson sampling (CTS) draws independent posterior samples for every arm, so its exploration dynamics ignore any relation among arms. We study a minimal change to those dynamics: an LLM is queried once for a partition of the arms, the partition becomes a positive-definite correlation matrix $Σ$ through an RBF kernel on cluster ranks, and the per-round posterior sample is drawn with covariance $Σ$ while the Beta posteriors are updated from real rewards only, so the LLM shapes how the sampler moves, not what it believes. We give a self-contained Bayesian regret bound for the idealized Gaussian sampler whose information gain splits into a $K\log T$ term from the $K$-cluster structure and a ridge term that grows to $d\log T$: the $\sqrt{d/K}$ improvement over independent sampling is a finite-horizon transient, exact only as the within-cluster correlation tends to one. The correlated sampler reduces regret by 19% over CTS on 16 synthetic Bernoulli families at $T=2{,}500$ (6-7% at $T=25{,}000$ with data-adaptive kernels) and by 41% on the Microsoft MIND-small news benchmark ($d=200$ real articles), while pseudo-observation warm starts give nothing. An LLM-free ablation with a simulated oracle of controlled quality shows that on unstructured instances the gain is a property of the kernel shape (a random partition, or a plain tempering of the sampling noise, reproduces it), while belief injection at matched oracle quality never helps. --- ### COMPASS: Finding Where Reasoning Lives in Language Models Authors: Pratyay Dutta, Kowshik Thopalli, Vivek Narayanaswamy Published: 2026-10-05 Categories: cs.AI arXiv: https://arxiv.org/abs/2610.07469v2 Explicitly eliciting reasoning substantially improves LLM performance. Existing approaches require a predefined characterization of reasoning, whether through CoT prompt design, contrastive CoT directions, or via SAE derived reasoning features. For mathematical reasoning with verifiable answers, we show that a much simpler signal suffices, which is the correctness of the model's own direct answer attempts. This signal yields a latent direction that elicits reasoning. This direction is decodable within the activations of most attention heads, but only a small subset of them can be effectively intervened. We introduce COMPASS, an inference-time steering method that identifies these heads using a logit-space attribution score and steers their activations along the correctness direction, requiring only per-head activation statistics. Across three model families and multiple math benchmarks, COMPASS outperforms the activation-steering baselines we compare against, improves GSM8K accuracy by 16 percentage points on average, and approaches CoT accuracy with 20-70\% fewer generated tokens. Interventions transfer without re-fitting to unseen benchmarks, and ablations show that both the correctness direction and the small set of heads carrying it are necessary, with the effect concentrated in remarkably few heads. --- ### Efficient Multimodal Inference through Adaptive Acquisition and Sequential Fusion Authors: Payal Mohapatra, Haodong Yang, Yueyuan Sui, Stephen Xia, Benjamin Lundell, Qi Zhu Published: 2026-10-05 Categories: cs.LG arXiv: https://arxiv.org/abs/2610.07466v2 Multimodal systems often encode every available input, even when a subset suffices for prediction. Adaptive acquisition can reduce this cost by using predictions from incrementally fused evidence to decide which modality to encode next and when to stop. However, sequential fusion makes these predictions order-dependent, so decisions based on them may need to distinguish factorially many histories of the same acquired set. We introduce SemARC, which couples a Sequential Modality Aggregator (SeMA) with an Adaptive Runtime Controller (ARC) and uses acquired evidence to select each modality before its encoder runs. SeMA executes only selected encoder and fusion branches, updates a fixed-size state, and predicts after each acquisition without recomputing earlier branches. We supervise every acquisition prefix under randomized modality subsets and orders to encourage consistent predictions across acquisition orders. ARC combines a set-dependent marginal-utility prior with residual fitted-Q learning to select the next available modality or stop, without inspecting unacquired inputs or retaining acquisition order. Across six multimodal classification datasets and eleven baselines, SemARC achieves 3.2% higher macro-F1 and 61.4% lower total inference GFLOPs on average relative to each dataset's most accurate baseline. End-to-end latency falls by 44.0% across GPU and CPU and by 47.2% on Android INT8 relative to the fastest measured baseline, on average. Under varying runtime modality missingness, SemARC still skips available modalities, matching or exceeding the best baseline macro-F1 in 21 of 24 conditions with 14.8% lower total GFLOPs on average. SemARC thus offers a practical path toward efficient multimodal inference across heterogeneous devices. --- ### Protective Perturbations Must Survive the Resize: Scale-Robust Image Immunization against Malicious Editing Authors: Zhongliang Guo, Yan Lin, Yifei Qian, Daizong Liu, Ognjen Arandjelović, Jianfeng Dong, Chun Pong Lau Published: 2026-10-05 Categories: cs.CV arXiv: https://arxiv.org/abs/2610.07464v3 While instruction-guided editors built on diffusion transformers can turn a personal photo into a convincing forgery with a single sentence, protective perturbations aim to make such edits fail before the photo is ever shared. The threat is most acute for the photos people actually share, which have 10 megapixels or more and which current editors first downscale by a factor that the owner does not know. Existing protections are computed at one resolution or averaged over random scales, and they lose most of their effect once the photo is downscaled. They implicitly assume that the editor sees the scale for which they were optimized, or that protecting the average scale protects every scale. We propose SRIM, a protection that remains effective at every scale an editor may use, without knowing its preprocessing and at the same computational cost as standard expectation over transformation. SRIM covers the range of scales implied by the photo size with anchor scales and spends each optimization step on the scale where the protection is currently weakest. A spectral analysis shows that this worst-case protection degrades only logarithmically with the width of the range, a value that averaging over scales cannot reach. Extensive experiments on full-resolution photos and popular instruction-guided editors validate that SRIM gives the strongest protection, outperforming existing methods even with a quarter of their $\ell_\infty$ budget, remaining the strongest under input purification. --- ### ElasticFit: Fit-Aware 3D Object Insertion via VLM Reasoning and Generative Adaptation Authors: Tzu-Hsin Hsieh, Ricardo Marroquim Published: 2026-10-05 Categories: cs.CV, cs.AI arXiv: https://arxiv.org/abs/2610.07460v2 Inserting objects into existing 3D scenes requires more than selecting a plausible location: the inserted object must also fit local geometry while preserving semantic intent and physical plausibility. Although recent Vision-Language Models (VLMs) and generative models enable semantic reasoning and visual content creation, they offer limited 3D grounding and geometric control when an inserted object must fit into constrained local spaces. We introduce ElasticFit, a VLM-guided framework for fit-aware object insertion centered on a novel scene-grounded representation. Given a language instruction and rendered scene observations, ElasticFit infers structured fitting cues that specify where the object should be grounded, what volume it should occupy, how it should be oriented, and its adaptation mode (rigid placement, uniform scaling, or elastic fitting). These cues convert high-level VLM reasoning into explicit 3D constraints that condition object generation and guide downstream geometric fitting. ElasticFit then generates a scene-conditioned object prior, reconstructs it in 3D, and refines the mesh through mode-specific fitting while enforcing collision avoidance, contact consistency, and physical grounding. In fixed-asset baseline comparisons, ElasticFit improves spatial relation success from 50.8% to 69.7% and support success from 48.3% to 91.7% over the strongest baseline, while providing novel support for generative "make-it-fit" insertions in complex scenarios. --- ### Auditable Claims about AI Agents Authors: Yue Zhao, Jiate Li, Li Li, Yi Nian, Jinbo Liu, Xiaolin Zhou, Xiyang Hu Published: 2026-10-05 Categories: cs.AI, cs.CL, cs.CY, cs.MA arXiv: https://arxiv.org/abs/2610.07459v1 Organizations make claims about their AI agents: a person approves every external email, every action is logged, an evaluation shows the agent is safe to deploy. Article 12 of the EU AI Act requires high-risk systems to allow the automatic recording of events but does not say which records settle a given claim. The position is one sentence: to be checked, a claim about an agent must first name its policy, its scope, the records that would settle it, and who writes them. Adapting the preconditions of an assurance engagement, we call a claim auditable when these elements and a decision rule are fixed before any verdict and the records are obtainable. This extends the Policy Checkability dimension of our Auditable Agents framework from single actions to claims. Agents add three conditions: coverage by an independent record, authorization bound to each action's arguments, and completeness beyond integrity. Under an explicit model, we prove that support is impossible without each wherever its hypotheses hold. A claim-check table applies the method to six common claims, anchored in current NIST, IETF, and OWASP drafts. A worked case follows one claim through five evidence states. We close with a practice box and steps for operators, buyers, auditors, and standard setters. --- ### Interpretable Hypergraph Learning via Neural Additive Models Authors: Shihan Feng, Xin Zheng, Shiyi Yang, Ren Wang, Chudi Zhong, Can Chen Published: 2026-10-05 Categories: cs.LG, cs.SI arXiv: https://arxiv.org/abs/2610.07458v1 Hypergraphs offer a natural framework for modeling networked data, where dependencies among entities are governed by higher-order interactions. While hypergraph learning methods such as hypergraph neural networks have demonstrated remarkable predictive performance, most existing approaches rely on black-box message-passing architectures, making it difficult to disentangle the contributions of node attributes and higher-order structural information. To address this challenge, we introduce the hypergraph neural additive network (HGNAN), an inherently interpretable framework for learning on hypergraph-structured data. HGNAN extends classical neural additive models to higher-order relational data by integrating feature-wise nonlinear decomposition with hypergraph-aware structural aggregation, enabling transparent prediction for both node- and hyperedge-level tasks. Extensive experiments on benchmark datasets demonstrate that HGNAN achieves performance comparable with state-of-the-art hypergraph learning methods while providing intrinsic and meaningful interpretability. --- ### AlignQuant: Tile-Aligned Mixed-Precision Quantization for Efficient LLM Generation Authors: Hanzhi Zhang, Qiao Zhang, Qinglei Cao, Heng Fan, Yan Huang, Kewei Sha, Yunhe Feng Published: 2026-10-05 Categories: cs.LG, cs.AI, cs.CL arXiv: https://arxiv.org/abs/2610.07457v1 Fine-grained mixed-precision quantization promises efficient large language model inference, but local precision choices can conflict with regular GPU storage and computation units. This precision-boundary mismatch limits the translation of compression into practical acceleration. We introduce AlignQuant, a post-training quantization method that uses GPU-compatible two-dimensional weight tiles as the common unit of precision allocation, compact storage, and execution. This shared partition lets precision follow sensitivity within output channels. Joint prefill/decode calibration scores precision reductions using projection-output perturbations weighted by language-model loss gradients under quantized activations. Phase-normalized scores prioritize higher precision for tiles important to either phase under a model-wide weight-storage budget. Each tile stores one selected representation, while phase-specialized kernels reuse the packed model and expand lower-bit weights for INT8 computation with 8-bit activations. Across four LLMs spanning 3B to 14B parameters, AlignQuant achieves up to $2.50\times$ generation speedup over BF16 while preserving model quality. Evaluations further cover three GPUs and contexts up to 64K tokens. These results show that local precision flexibility and regular GPU execution can coexist through a shared tile unit. The implementation is available at https://github.com/HanzhiZhang-Ulrica/AlignQuant. --- ### Active Feature Acquisition for Cost-Efficient Temporal Prediction with Reduced Participant Burden Authors: Yunni Qu, Bing Cai Kok, Whitney Ringwald, Grant King, Aidan Wright, Kathleen Gates, Junier Oliva Published: 2026-10-05 Categories: cs.LG, cs.AI, stat.AP, stat.ME, stat.ML arXiv: https://arxiv.org/abs/2610.07452v1 Accurate forecasting of pathological outcomes is a central problem in psychology. To do so, psychologists often collect intensive longitudinal data. However, in such studies, the desire to acquire a large number of variables for the sake of accurate prediction is often counteracted by the need to minimize participant burden. Acquiring more variables per occasion can yield better predictions, but having too many acquisitions increase the risk of non-response and attrition. Longitudinal Active Feature Acquisition (LAFA) is a principled approach to resolve this conundrum. Instead of requiring responses to every item at every acquisition occasion, LAFA produces a policy that seeks to optimally select dynamic subsets of items to be acquired at each timepoint while preserving our ability to forecast a specific outcome. However, existing LAFA methods are mostly based on Neural Networks (NN) that are difficult to interpret in practice. In this work, we introduce a tree distillation method for learning an interpretable policy from NN-based LAFA networks. We validated our method through both a simulation and an empirical EMA dataset on forecasting daily alcohol consumption. In both cases, we find that we can meaningfully reduce the number of items acquired at each occasion with minimal loss in accuracy. Networks (NN) that are difficult to interpret in practice. In this work, we introduce a tree distillation method for learning an interpretable policy from NN-based LAFA networks. We validated our method through both a simulation and an empirical EMA dataset on forecasting daily alcohol consumption. In both cases, we find that we can meaningfully reduce the number of items acquired at each occasion with minimal loss in accuracy. ---