Papers

Multimodal large language models (MLLMs) can answer knowledge-intensive visual questions by combining visual evidence from images with facts retrieved from external sources. However, MLLMs may overlook relevant evidence in both modalities, attending weakly to the textual sentences or visual regions needed for the correct answer. Recent efforts address this by highlighting retrieved text and marking visual regions before generation, but apply a fixed, one-shot policy that cannot adapt to three sources of variation: whether highlighting is necessary, how much evidence different examples require, and when different textual evidence becomes relevant as the answer unfolds. We introduce Adaptive Relevance-guided Evidence Allocation (AREA), a training-free inference-time method that formulates evidence highlighting as adaptive allocation. AREA generates a single probe token to read visual and textual relevance from fixed backbone layers, then makes three decisions: i) whether to intervene (controlled by natural attention coverage and visual sink contamination), ii) how much evidence to expose (determined by relevance entropy), and iii) when to refresh text during generation (triggered by causal context-attention peaks). Across four KB-VQA and seven standard multimodal benchmarks with nine frozen MLLM checkpoints, establishes the best performance among training-free highlighting methods.

People increasingly reason with large language models (LLMs), yet complementary capabilities do not guarantee outperforming both components. In a between-subjects study, participants (N=535) solved a 40-item battery of matrix reasoning, mental rotation, syllogisms, and letter-string analogies, unaided or with GPT-5.6-Luna, Claude Opus 4.8, Gemini 3.6 Flash, or Kimi K3. Each assisted trial required consultation with the model. Each model answered every item alone 100 times under matched elicitation. The assisted-unaided accuracy difference increased with item-level LLM competence. Deference varied across tasks and increased with competence within tasks. Post-advice confidence distinguished correct from incorrect answers less strongly than unaided confidence. In a reference comparison, about half the increase in LLM accuracy carried through to assisted accuracy. How much of that accuracy gain reached participants differed across the models. These findings motivate evaluating LLMs in interaction with humans and designing support for selective deference that preserves independent reasoning.

Noise2Noise (N2N) trains denoisers on pairs of independently corrupted observations, eliminating clean references. We stress-test two natural conjectures about why the L1 loss outperforms L2 here. First, the hypothesis that the L1 loss confers robustness via parameter sparsity confuses the loss with Lasso regularization: an explicit Lasso penalty produces the predicted sparsity yet fails to reproduce L1's cross-noise behavior, while L1- and L2-trained weight distributions are indistinguishable. Second, the population optima of the two losses coincide exactly for symmetric signal posteriors and nearly so for concentrated ones. Measured differences are therefore dominated by optimization dynamics (bounded-influence gradients), which we probe with gradient statistics and contaminated-target training. On Kodak24 with five synthetic noise families, the L1 loss holds a statistically significant edge over L2, below 1 dB PSNR, holding across three seeds on 13 of the 14 noise columns. On real camera noise the loss is not the decisive variable in distribution: on official SIDD validation blocks, synthetic-Gaussian-trained N2N models gain only 0.8 to 3.7 dB over the noisy input regardless of loss, while retraining on SIDD's own noisy pairs, never reading ground truth, gains 9.4 to 11.0 dB, far ahead of BM3D. All metrics are on raw network outputs, and the study makes no leaderboard claim. The training pair distribution, not the loss, carries the inductive bias. That design rule applies wherever clean references are unobtainable, from microscopy to industrial inspection sensors.

The contest in large-scale AI computing is no longer whether one processor can be made stronger, but whether a million of them can still behave as a single computer. This paper argues that two extensions are required. First, classic BSP, nested recursively, becomes a plan that holds at every scale: each superstep consists of parallel work, then exchange and aggregation, then a barrier, and at every layer the participating units are peers. Second, the von Neumann single-machine architecture, extended past its master–slave habit, becomes the Nested Parallel von Neumann Architecture: a nest of peer-equal layers, from chip package to autonomous zone, joined end to end by one memory-semantic interconnect, the Unified Bus. The software nest and the hardware nest correspond layer by layer, and the $\tau$ Scaling law folds time at every layer. We examine which workloads the nested structure serves, AI training above all but much of classic HPC as well, and describe the hardware decisions that make the nesting physical. Many processors, still one computer.

In this work, we propose a structured methodology for the system identification of underwater vehicles through the design of optimal excitation trajectories. To this end, the trajectories are parameterized using Bezier curves, which ensure smooth and differentiable motion profiles while facilitating the enforcement of constraints through appropriate manipulation of the control points. An optimization problem is formulated to determine a dynamically feasible excitation trajectory that respects safety limits and maximizes the quality of the collected data, thereby enabling reliable estimation of the vehicle's dynamic parameters using least squares. The proposed methodology is experimentally validated in a laboratory water tank, where the dynamic parameters, identified from the optimized trajectory, are evaluated by predicting the vehicle's velocity through forward simulation on previously unseen trajectories.

Zero-shot anomaly detection aims to localize anomalies without target-domain samples. Existing CLIP-based methods suffer from coarse anomaly maps and limited semantic prompts. We propose PSMP-CLIP, integrating patch-prompt SAM2 segmentation (PPSS) and multi-semantic guided prompt regularization (MSGPR). PPSS samples prompts directly from intermediate patch features, avoiding threshold drift and guiding SAM2 to produce precise masks. MSGPR uses multiple learnable prompts constrained by semantic anchors to preserve generalization. Experiments on 14 datasets show highly competitive performance, achieving the best pixel-level AUROC on MVTec AD, BTAD, DTD-Synthetic, CVC-ClinicDB, TN3K, Endo, and Kvasir.

Generative AI has become a routine resource in programming education, and most institutional responses to it are attempts at control, either by restricting access or by offering students a controlled version of the technology. This paper reports a seven-week mixed-methods pilot study in a master's-level data analytics course, in which 33 students were randomly assigned either to a scaffolded AI Study Coach embedded in the notebook-based laboratory sessions or to unrestricted use of AI tools of their own choosing. The Coach offered stepwise hints, did not generate code, limited the number of hints per session, and required a short reflection at the end of each session. The design assumed, in line with scaffolding theory and recent experimental evidence, that guided and limited support would build confidence and reduce over-reliance, and that the scaffolded group would learn more. Assignment performance did not differ between the conditions. Students in the Coach condition reported higher confidence but managed the hint budget poorly, while students in the unrestricted condition were satisfied with their tools and uneasy about how much they depended on them. In interviews, students in both conditions identified awareness of their own reliance on AI as the most valuable outcome of the course. Students who had formulated their own rules for when to use AI performed better in both conditions, and those with the best understanding of how the models work, in every case self-taught, used the tools most deliberately and achieved the highest scores. The design of the tool mattered less than the students' capacity to govern their own use of it, a capacity that is at present acquired by chance. The paper argues that the appropriate response is structural: assessment that grades the reasoning behind AI-assisted work, and AI literacy taught explicitly as a core skill.

LLMs are increasingly employed in a wide range of decision-making tasks. However, the opacity of their internal reasoning makes it difficult to validate or interpret their outputs, and the need for interpretability becomes especially critical in high-stakes settings. This study examines the decision-making capabilities of LLMs through the Analytic Hierarchy Process (AHP), a classical and widely used multicriteria decision-making framework. We construct a new annotated benchmark based on AHP and propose the first end-to-end approach that enables LLMs to perform the complete AHP workflow. Experiments in real-world decision problems in the legal and higher-education ranking domains show that our method significantly improves alignment with expert judgments.

The unchecked proliferation of manipulated images on social media platforms has increased the spread of misinformation, posing a severe threat to public trust and information integrity. Modern deepfake detectors typically rely on Vision Transformers (ViTs) to capture the low-level inconsistencies that characterize fully synthetic or locally tampered images. However, the global understanding of such foundation models is not enough to discriminate alone between real and fake multimedia content, especially in challenging scenarios where images are compressed or transmitted through social media. In this paper we pioneer the application of Joint-Embedding Predictive Architecture (JEPA) models to deepfake detection, taking advantage of the generalized representation of visual reality that such World Models have exhibited. We hypothesize, and empirically demonstrate, that the intrinsic world understanding of JEPA models can be used as a strong prior for a deepfake detector. To fully exploit JEPA capabilities, we propose MoE-JEPA, a dual-stream architecture for deepfake detection. By enhancing a V-JEPA 2 backbone with a Residual Mixture-of-Experts (MoE) mechanism, along with a noise stream branch, our model dynamically internalizes forensic knowledge. Furthermore, a Gated Attention Multiple Instance Learning (MIL) module is employed to ensure precise spatial semantic understanding. Evaluated on the SID-Set benchmark, comprising 300K AI-generated, tampered and authentic images, MoE-JEPA establishes a new state-of-the-art with an accuracy of 95.54%, successfully outperforming vastly larger models.

As Large Language Models (LLMs) increasingly serve as primary knowledge retrieval interfaces, their robustness against persuasion attacks—attempts to inject misinformation or enforce counterfactuals—has become a critical safety concern. Existing red-teaming frameworks typically evaluate models in multi-turn dialogues where the target model retains full conversation history. We identify a critical flaw in this setting termed “Refusal Inertia”: a model's initial refusal often propagates through subsequent turns largely to maintain contextual consistency, thereby masking its true vulnerability to sophisticated, isolated persuasion attempts. To rigorously evaluate the “cold-start” defense capabilities of SOTA models, we introduce the SAST-IR (Stateful Attacker, Stateless Target - Iterative Refinement) framework. By enforcing a memory wipe on the target while retaining the attacker's history, we simulate a worst-case adversarial setting using multi-turn (stateless) iterations. Leveraging CP-Agent (Cognitive Persuasion Agent), an enhanced diagnosis-guided agent, our experiments on the custom CounterFact-Strict dataset ($N=50$) yield alarming results: simple, diverse attack strategies achieved a staggering 96% success rate, exposing severe brittleness in memory-less defense. Furthermore, we reveal a “Complexity Paradox”: while complex, iteratively refined attacks are effective, they often trigger defensive compliance, whereas simple strategies achieve a higher rate of genuine persuasion (84.7%). Our code and dataset are available at GitHub, https://github.com/cza1006/llm-persuasion-defense.

Annotating large radiology datasets is bottlenecked by the manual effort of delineating structures slice-by-slice in 3D volumes. Interactive methods reduce this effort but stay interaction-inefficient: slice-wise methods (including many foundation models) ignore inter-slice continuity, while 3D and video-based methods propagate a prompt with a fixed propagator that never adapts to the target volume, so it drifts on low-contrast or pathological structures and must be re-prompted. We present IMVS, a human-in-the-loop annotation framework that composes three components into a closed loop rather than a new segmentation primitive: a lightweight 2D Slice Mask Adapter (SMA) fine-tuned online from user scribbles, a frozen Volume Mask Tracker (VMT) that propagates corrected masks across adjacent slices, and a soft teacher–student alignment that limits forgetting. The SMA is backbone-agnostic (UNet++, DeepLabV3, TransUNet). Across 8 public CT/MRI datasets, IMVS matches strong interactive baselines in quality while sharply cutting annotation effort: $14.4\times$ faster than a proficient copy-based manual workflow ($22.3\times$ over naive manual), $4.6\times$ over slice-wise and $1.9\times$ over 3D interactive methods. MedSAM2 and ScribblePrompt stay competitive or stronger on well-delineated organs; IMVS's advantage is largest on challenging targets and on interaction efficiency. Source code and Demo Video: https://github.com/AbhilakshSinghReen/imvs.

Post-disaster phasor measurement unit (PMU) outages reduce power-system observability and degrade operator situational awareness, requiring sequential restoration under limited resources. Existing PMU restoration methods based on optimization or heuristics can generate restoration schedules, but they often provide limited support for explanation, traceability, and operator interaction. This paper proposes an agentic tool-calling framework orchestrated by a large language model (LLM) for post-disaster PMU restoration and grid observability recovery. In this framework, the LLM does not directly solve the restoration optimization problem; instead, it coordinates validated backend tools required for post-disaster restoration, including observability assessment, restoration planning, state updates, and operator verification. The framework also maintains a structured tool-call history and execution context that keep restoration decisions traceable and explainable, while enabling context-aware operator question answering during the restoration process. Simulation results on IEEE 30-bus and IEEE 57-bus systems show that the proposed framework achieves observability recovery comparable to a mixed-integer linear programming (MILP) solution, while providing tool-grounded explanations, interactive operator support, and human-overseen execution.

Image segmentation remains challenging due to occlusions, poor lighting, and irregular structures. Although transformer-based methods achieve high accuracy, they rely heavily on long-range spatial features, leading to high computational costs and neglecting prior knowledge or noise patterns, resulting in missing details and unclear boundaries. To address these issues, we propose Frequency Spatial Aware Network (FSANet), which integrates prior knowledge with a dual-domain solver to sequentially adapt to diverse segmentation tasks. Specifically, we design three key modules: (1) Structure Prior Module, which recovers overlooked details; (2) Dual-Domain Awareness Module, which captures salient features while disentangling noise; and (3) Edge Estimation Module, which enhances edge awareness for more precise segmentation. In addition, the limited availability of comprehensive segmentation datasets covering various real-world scenarios hinders the performance of existing methods. To address this, we introduce SceneX, a novel open-source dataset featuring 10 challenging non-ideal scenarios, establishing a new benchmark for evaluating and improving the robustness and real-world applicability of the segmentation models. Extensive experiments demonstrate the efficiency and effectiveness of FSANet.

Robust 3D reconstruction from multi-view optical satellite imagery requires fusing complementary but sometimes conflicting geometric evidence. Digital surface models (DSMs) are the primary elevation representations for satellite-based 3D reconstruction, making reliable height estimation essential. However, in a Gaussian scene representation jointly optimized from multiple views, Gaussian responses at different elevations can support competing height hypotheses at the same rendered location, while conventional alpha-weighted elevation aggregation may produce intermediate elevations that do not correspond to physical surfaces. To address this challenge, we formulate DSM reconstruction as a reliability-aware height-hypothesis fusion problem and propose HLC-GS, a reliability-aware Height-Layer Consistency Gaussian Splatting framework for multi-view satellite 3D reconstruction. HLC-GS organizes projected Gaussian responses into candidate height hypotheses and evaluates their relative support using layer competition and Gaussian footprint support. A continuous height-layer risk map guides dominant-layer reliability correction and secondary-layer suppression during optimization. The proposed training strategy regulates conflicting Gaussian responses within the shared representation to improve the reliability of reconstructed surface elevations. Experiments on seven scenes from the DFC2019 and IARPA2016 datasets demonstrate improved DSM reconstruction accuracy. Compared with EOGS, HLC-GS reduces the average DSM MAE from 1.46~m to 1.18~m and RMSE from 2.78~m to 2.58~m, while increasing PAG$_{2.5}$ from 86.09% to 88.61%, with comparable computational cost.

IMU-based human activity recognition (HAR) enables continuous, privacy-friendly monitoring of daily activities using wearable sensors. However, building reliable HAR models that generalize across diverse users and real-world conditions requires large amounts of labeled IMU data, which are expensive and difficult to collect. Existing approaches mainly rely on augmentation or synthesis to expand available data, but indiscriminately adding virtual samples may provide little new coverage and introduce unreliable supervision. To overcome these challenges, we propose a novel coverage-aware virtual IMU augmentation framework that decides where to supplement real data, how to generate and select virtual candidates, and how strongly to weight them during training. Specifically, we select diversity and scarcity anchors in a learned sensor embedding space, convert anchor dynamics into prompts, and generate virtual IMU candidates for each anchor. We then rank candidates by a selection cost combining anchor proximity and label consistency, and incorporate the selected candidates into HAR training with reliability-based weights. Experiments on public HAR benchmarks show that our method consistently improves recognition performance over competitive baselines, and ablation studies confirm the effectiveness of the proposed framework design.

Unlike users, who know their own belongings and routines, household robots cannot easily acquire such personalized object knowledge automatically and depend on users to teach them. User-initiated teaching requires users to arrange dedicated teaching sessions and decide what to teach, even when they are unsure what the robot needs to learn. We introduce EgoAsk, a smart-glasses-based system that proactively embeds personalized object teaching into everyday activities. EgoAsk shares the user's first-person view with the robot, identifies gaps in personalized object knowledge, and analyzes ongoing activity to ask context-relevant questions that support future household assistance. To examine how teaching initiative and question timing affect users' teaching experiences, we conducted a within-subjects study with 18 participants and found lower reported knowledge-gap monitoring burden with robot-initiated questioning and less need for context reconstruction with EgoAsk. These findings characterize teaching burdens and timing preferences, offering design implications for egocentric robot-teaching systems.

Retrieval-Augmented Generation (RAG) evaluators can identify failures such as weak retrieval, poor grounding, incomplete answers, and unsupported generation, but they rarely help developers decide what to repair next. We present RECTIFY, an interactive Streamlit workbench that turns evaluated RAG cases into auditable repair workflows. RECTIFY filters cases that do not require repair, routes remaining failures into actionable families and finegrained repair slices, and generates editable repair cards that developers can approve, reject, or verify through sandbox reruns. On a controlled RAG benchmark, RECTIFY surfaces interpretable failure profiles across BM25, dense, and hybrid retrieval: BM25 mainly triggers noisy-retrieval repairs, while dense and hybrid retrieval leave smaller sets of multi-part underretrieval and underused-evidence cases. Additional analyses show that pre-filtering reduces unnecessary repair candidates and that slicelevel routing yields more targeted repair cards than broad family-level diagnosis. RECTIFY is publicly available as an open-source Streamlit workbench 1 for helping developers turn evaluation results into inspectable repair decisions.

Degree-sequence realizability is the combinatorial basis of configuration models, but degree constraints alone do not ensure the preservation of graph dynamics. Hence, configuration models are unable to recover centrality measures, unless these are strongly correlated with the degree sequence. To address this matter, we introduce EP-realizability, the analogue problem induced by an equitable partition (EP): given the EP of a graph, decide whether the partition is realized by a simple undirected loopless graph and therefore construct such a graph. After defining the problem, we solve it by reducing it to sub-problems related to Havel–Hakimi and the Gale–Ryser theorem. We also face the challenge of solving the problem with an Approximate Equitable Partition ($\varepsilon$-EP), so that it is possible to reconstruct a network starting from partial and more privacy-preserving information. We evaluate privacy with edge overlap, deriving also, for our proposed $\varepsilon$-EP-realizability solution, a predictor for this metric. Experiments on Karate, Cora, CiteSeer and PubMed datasets show that our algorithm achieves a favourable and tunable privacy–utility trade-off, comparing the results with Havel–Hakimi algorithm, Newman's configuration model and a stochastic block model. Finally, both with real data and random graphs, we show that our algorithm has approximately linear time complexity with respect to the number of edges.

With the rapid development of Large language model (LLM), agent systems enhanced by LLMs show huge potential in being able to deal with complex tasks, especially involving multi-step thinking or interaction with tools. For applying LLM techniques with a well-designed agent paradigm, post-training of LLM in multiple agent scenarios is necessary to achieve better performance. Among the variable post-training techniques, alignment methods such as PPO, DPO, DIL, and GRPO become popular because many papers show a significant positive impact on the model's performance by punishing negative samples while keeping acceptable training complexity. However, most alignment methods address simple single-turn tasks, and there remains room for improvement for complex multi-turn tasks. We propose Turn-level Multiscale Density Ratio Estimation (tlm-DRE), which assigns different weights on corresponding turns and proposes asymmetric token-level training based on the positive-negative space gaps across multiple turns of tasks. The results of the experiment on a wide range of agent benchmarks show that the proposed method performs competitively compared to traditional alignment methods. The proposed training method enables LLMs to perform robustly in multi-turn reasoning tasks with both in-domain and out-of-domain conditions.

Brain tumor segmentation remains difficult because enhancing tumor (ET) has low contrast and overlaps surrounding tissue, while scanner and site variation causes domain shift. We propose DE-GAN, a contrast-enhancing conditional GAN that combines input-adaptive dynamic convolutions, style-aware feature mixing, and coordinate encoding to synthesize slice-adaptive FLAIR images. A label-guided, class-conditional target separates tumor-core (TC) and ET intensities while preserving anatomy. The generated FLAIR is concatenated with the original MR modalities and used to train a 3D U-Net. Across BraTS 2015, 2018, and 2019, DE-GAN improves segmentation over the baseline and static EnhGAN replacement on most reported TC/ET metrics, with the largest gains from retaining both original and enhanced FLAIR. Code and pretrained models are available at https://github.com/zkhansuri-ui/DE-GAN.

Fine-tuning an aligned language model on narrow, flawed data can induce harmful behavior far outside the training domain, known as emergent misalignment (EM). Prior work has localized EM in model weights, activations, and training documents, but it remains unclear which training tokens carry the relevant fine-tuning signal. We introduce TAME (Token Attribution and Masking for Emergent Misalignment), a three-stage framework: token attribution scores how strongly the fine-tuning update raises each response token's likelihood, using forward passes through a released LoRA adapter; signal characterization finds patterns among high-attribution tokens; and causal validation tests them by attribution-guided loss masking. On released EM organisms and a 6,849-example medical-advice split, attribution is concentrated (the top 5% of tokens hold 32% of the mass) and, in Llama, depleted for medical vocabulary but enriched for a register of unwarranted certainty, even after controlling for token rarity. Masking high-attribution tokens during fresh fine-tuning cuts EM by 23x in Llama and 36x in Qwen, with the perplexity cost concentrated on the targeted register rather than on medical content; an equal random mask leaves EM unchanged. In Llama, the attribution pattern suggests that EM-relevant signal lies more in how confidently flawed content is expressed than in its domain vocabulary; the causal masking effect itself holds across both model families.

Cognitive Field Theory describes cognition as memory-dressed collective dynamics, in which slow relaxation modes support a persistent cognitive field. By connecting non-Markovian memory dynamics to discrete recurrent state evolution through an extended state, we develop a Cognitive Field Network (CFN), a recurrent Transformer architecture with hidden-field re-entry, \( \Phi_{n+1} = F_\theta(X_{n+1},\Phi_n) \). Rather than prescribing an explicit memory operation, the CFN allows the full token-resolved hidden field to re-enter the next computational cycle. We find that learning organizes this pathway into a functional channel for persistent, content-dependent internal states, with temporal persistence systematically controlled by the recurrent training horizon. The learned dynamics generalize beyond this horizon: semantic continuation sustains the recurrent state without explicit answer replay, whereas unsupported states exhibit finite passive relaxation. Periodic re-exposure to relevant input renews the surviving field, producing an approximately stationary nonzero regime over extended recurrent trajectories. Unrelated-input and recurrence-off controls fail to reproduce this behavior, while near-paraphrased input produces weaker renewal, revealing the dependence of persistence on both field re-entry and input relevance. These findings connect memory-dressed field formation, input-driven reorganization, and recurrent state continuity within a unified dynamical framework. The CFN thus provides an experimentally controllable platform for studying persistent, history-dependent cognition without a separately prescribed memory system.

We give deterministic and uncoupled learning dynamics for finite multiplayer general-sum games under full-information feedback that achieve constant individual swap regret in self-play, independent of the horizon $T$. With $n$ players and at most $m$ actions each, every player's individual swap regret is $O(\sqrt n\,m\log m\log^{5/2}(nm))$ at every finite horizon. The dynamics use the classical Blum-Mansour framework with optimism. Each player predicts the deviation gains, uses these predictions to update a row-stochastic transition matrix, and plays its stationary distribution. Our new ingredients include a tailored row normalization map and a two-scale higher-order predictor. An adversarially robust variant, obtained through a generic common-prefix switching wrapper, preserves the self-play bound up to a universal constant and guarantees individual swap regret at most $7\sqrt{mT\log m}$ in the adversarial setting.

Text summarization requires models to condense content while preserving key qualities such as consistency and coherence. Large language models (LLMs) have shown strong performance on this task and can be further improved through reinforcement learning (RL). However, most existing methods apply reward signals directly to undifferentiated token sequences, overlooking the varying importance of individual tokens to word and sentence level quality in summarization. In this paper, we propose Token Importance-Aware Policy Optimization (TIAO), a novel reinforcement learning strategy that explicitly leverages token-importance awareness. Specifically, TIAO identifies core tokens based on token dependency and reweights a trajectory's advantage according to its overall dependencies. Experiments on the real world dataset show that our TIAO achieves highly competitive results, and that a 7B foundation model enhanced by TIAO performs comparably to GPT-4 and GPT-5-nano. Code is available at https://github.com/TechCloud-x/TIAO

Action Chunking Transformers (ACT) are widely used to learn robot manipulation from demonstrations. Their conditional variational autoencoder includes an encoder meant to capture differences between demonstrations during training. The original ACT paper reported that encoder removal dropped the mean success rate from 35% to 2% on two simulated tasks with human demonstrations. We re-ran this ablation in the original code and checked whether the findings depend on the implementation or training data. The published drop does not reappear in our tests, although smaller gains or losses in success rate remain uncertain. To investigate the discrepancy, we varied training length and how checkpoints are selected for evaluation. Both can reverse which policy scores higher, but the published drop's cause remains unknown. Success rates alone leave open whether the encoder provides information that helps the policy reconstruct demonstrated actions. On the tested ACT benchmark, the sampled latent provides little reconstruction benefit at every tested nonzero weight of the penalty on latent information. At inference, ACT leaves this latent unused and sets it to zero. Skipping the encoder increases training throughput in both implementations we timed. We release code, evaluation tools and results so others can repeat the comparisons and test the encoder on other tasks.

The integration of renewable energy sources into the electrical grid introduces complex challenges in fault detection and coordination of grid recovery mechanisms. Traditional relay protection systems, which operate based on static rules and predefined thresholds, are inadequate for addressing these challenges, particularly in detecting and isolating faults such as short circuits. Consequently, the conventional methodologies applied to electrical network protection frequently fail to achieve optimal performance in fault detection, especially in terms of adherence to safety standards and the selective limitation of damage. Recent research indicates that machine learning (ML)-based approaches can effectively tackle these issues; however, variations in grid configurations and analysis windows have impeded consistent comparative assessments. In this study, we assess the efficacy of various ML models in detecting electrical faults and pinpointing defective transmission lines within a 10 ms measurement interval - a critical time-frame for real-time operational viability, for the first time. The most effective model attained an F1 score of 0.991 +/- 0.018 and demonstrated a processing time of 0.342ms +/- 0.509ms.

Convolutional Neural Networks (CNNs) are increasingly deployed in safety-critical edge applications, where soft errors can silently corrupt inference outputs and lead to unsafe decisions. Such applications typically rely on resource-constrained embedded GPUs, requiring fault detection and mitigation techniques that add minimal compute, memory, and latency overhead while integrating seamlessly with the standard GPU inference pipeline. Existing algorithm-based fault tolerance techniques rely on matrix augmentation and per-operation checksum verification, imposing substantial overhead that is prohibitive for CNN inference on embedded GPUs. In this work, we propose carry-through checksum, a fundamentally new scheme for soft-error detection in CNN inference on embedded GPUs. The method embeds dedicated carry-through filters into the convolutional layers, which compute a checksum from the CNN's own operations and propagate it through inference, enabling end-to-end error detection with a single output verification. Experimental results on multiple CNN architectures show that the proposed method detects 95.86% and 86.56% of critical faults for FP32 and FP16, respectively, at almost no additional per-image overhead. Detected faults are mitigated through re-execution, incurring only 2.27% run-time overhead across the entire test set on an NVIDIA Jetson Orin NX GPU.

Background: Large language models (LLMs) have achieved physician-comparable performance on multiple-choice medical knowledge examinations, but their capabilities in clinical history taking, urgency assessment, and safety remain insufficiently evaluated. We proposed Japanese Stroke LLM Evaluation, a multi-turn conversational benchmark for stroke care in Japanese, and evaluated LLM performance and safety under practice-oriented conditions. Methods: We created 10 stroke and related-condition cases and evaluated LLMs in multi-turn Japanese conversations. The LLM acted as physician, while a board-certified neurosurgeon acted as simulated patient and evaluator. Each case comprised history-taking and action phases scored using pre-specified criteria. Errors that could directly threaten life were defined as critical mistakes. The safety threshold was at least 80% overall with zero critical mistakes. Eighteen models were evaluated in October 2025 and June 2026. Results: Claude Fable 5 achieved the highest score (87.4%) with zero critical mistakes, followed by Claude Opus 4.7 (80.3%) and GLM-5.2 (75.6%). Two leaders met the safety threshold. Eleven models made 17 critical mistakes, including failure to confirm laboratory results or blood glucose before t-PA, surgery before airway stabilization, omission of cervical vascular evaluation, and t-PA outside its indication. History-taking question count correlated with history-taking score (r = 0.648, p = 0.007). Conclusions: Japanese Stroke LLM Evaluation provides a benchmark for LLM performance under practice-oriented conditions, including a cap on history-taking questions. Cases and evaluations were created by neurosurgical specialists rather than using an LLM-as-judge approach. Performance improved across cloud-based and on-premise models in 2026, with some exceeding the safety threshold. Further evaluation using real-world cases is required.

Power Flow (PF), Optimal Power Flow (OPF), and State Estimation (SE) are fundamental problems in power system analysis, but solving them is computationally expensive. Graph Neural Networks (GNNs) have been proposed as fast surrogates, yet existing solvers are trained for a single problem at a time, producing narrow models that must be rebuilt for each new task. We propose a more general approach: a single Heterogeneous Residual Gated Graph Convolutional Network that solves all three problems with one shared backbone. Rather than learning one mapping, the model learns a reusable representation of how the network behaves, from which PF, OPF, and SE can each be estimated. Trained jointly on the three problems across diverse topologies and loading conditions, and evaluated on the IEEE 14-bus and 118-bus systems, the shared model matches the accuracy of task-specific GNN solvers and stays robust on unseen loading levels and topologies. These results show that a single model can capture the basic operation of a power network and serve several analysis tasks at once, a first step toward a foundation model for power systems.

Generative video models can serve as a promising backbone for robot navigation by predicting future observations as video plans. Recent approaches often condition video planning on short-horizon guidance and recover geometric waypoints through scene reconstruction, leaving longer-horizon planning and precise video-to-action translation less explored. We present CueNav, a video model-based navigation framework combining visual cue guided video planning with an embodiment-specific Inverse-Dynamics Model (IDM). As visual cues, we use a Bird's-Eye View (BEV) map to convey global task context and retain part of the robot body in the egocentric observation to expose embodiment context. These cues guide the video planner, while the IDM translates dense flow fields extracted from the video plan into robot actions. With the visual cue encoding global task context, CueNav achieves nearly 2x higher success in maze navigation than planning without the cue. The body-aware view with the IDM enables precise navigation with 70% success in a narrow passage where comparison methods largely fail to complete the task. We further demonstrate zero-shot semantic-conditioned navigation and deployment of the same video planner across different robot platforms. Our results show that visual cue-guided video planning with embodiment-specific action grounding paves the way toward a generalizable navigation framework for longer-horizon planning and embodiment-aware control. Additional results and code are available on our project website: https://cuenav.github.io.