Recent advances in multimodal large language models (MLLMs) have significantly enhanced video understanding capabilities, opening new possibilities for practical applications. Yet current video benchmarks focus largely on indoor scenes or short-range outdoor activities, leaving the challenges associated with long-distance travel largely unexplored. Mastering extended geospatial-temporal trajectories is critical for next-generation MLLMs, underpinning real-world tasks such as embodied-AI planning and navigation. To bridge this gap, we present VIR-Bench, a novel benchmark consisting of 200 travel videos that frames itinerary reconstruction as a challenging task designed to evaluate and push forward MLLMs' geospatial-temporal intelligence. Experimental results reveal that state-of-the-art MLLMs, including proprietary ones, struggle to achieve high scores, underscoring the difficulty of handling videos that span extended spatial and temporal scales. Moreover, we conduct an in-depth case study in which we develop a prototype travel-planning agent that leverages the insights gained from VIR-Bench. The agent's markedly improved itinerary recommendations verify that our evaluation protocol not only benchmarks models effectively but also translates into concrete performance gains in user-facing applications.

We introduce SWE-Bench Pro, a substantially more challenging benchmark that builds upon the best practices of SWE-BENCH [25], but is explicitly designed to capture realistic, complex, enterprise-level problems beyond the scope of SWE-BENCH. SWE-BENCH PRO contains 1,865 problems sourced from a diverse set of 41 actively maintained repositories spanning business applications, B2B services, and developer tools. The benchmark is partitioned into a public set with open access to problems sourced from 11 repositories, a held-out set of 12 repositories and a commercial set of 18 proprietary repositories where we have formal partnership agreements with early-stage startups. Problems in the held-out and the commercial set are not publicly accessible, but we release results on the commercial set. Our benchmark features long-horizon tasks that may require hours to days for a professional software engineer to complete, often involving patches across multiple files and substantial code modifications. All tasks are human-verified and augmented with sufficient context to ensure resolvability. To better understand these limitations, we cluster the failure modes observed in the collected agent trajectories for a clearer characterization of the error patterns exhibited by current models. Overall, SWE-BENCH PRO provides a contamination-resistant testbed that more faithfully captures the complexity and diversity of real-world software development, advancing the pursuit of truly autonomous software engineering agents at a professional level.

In this paper, we investigate the efficiency of Deep Neural Networks (DNNs) to approximate the solution of a nonlocal conservation law derived from the identical-oscillator Kuramoto model, focusing on the evaluation of an architectural choice and its impact on solution accuracy based on the energy norm and computation time. Through systematic experimentation, we demonstrate that network configuration parameters-specifically, activation function selection (tanh vs. sin vs. ReLU), network depth (4-8 hidden layers), width (64-256 neurons), and training methodology (collocation points, epoch count)-significantly influence convergence characteristics. We observe that tanh activation yields stable convergence across configurations, whereas sine activation can attain marginally lower errors and training times in isolated cases, but occasionally produce nonphysical artefacts. Our comparative analysis with traditional numerical methods shows that optimally configured DNNs offer competitive accuracy with notably different computational trade-offs. Furthermore, we identify fundamental limitations of standard feed-forward architectures when handling singular or piecewise-constant solutions, providing empirical evidence that such networks inherently oversmooth sharp features due to the natural function space limitations of standard activation functions. This work contributes to the growing body of research on neural network-based scientific computing by providing practitioners with empirical guidelines for DNN implementation while illuminating fundamental theoretical constraints that must be overcome to expand their applicability to more challenging physical systems with discontinuities.

Deploying Large Language Models (LLMs) in medical applications requires fact-checking capabilities to ensure patient safety and regulatory compliance. We introduce MedFact, a challenging Chinese medical fact-checking benchmark with 2,116 expert-annotated instances from diverse real-world texts, spanning 13 specialties, 8 error types, 4 writing styles, and 5 difficulty levels. Construction uses a hybrid AI-human framework where iterative expert feedback refines AI-driven, multi-criteria filtering to ensure high quality and difficulty. We evaluate 20 leading LLMs on veracity classification and error localization, and results show models often determine if text contains errors but struggle to localize them precisely, with top performers falling short of human performance. Our analysis reveals the "over-criticism" phenomenon, a tendency for models to misidentify correct information as erroneous, which can be exacerbated by advanced reasoning techniques such as multi-agent collaboration and inference-time scaling. MedFact highlights the challenges of deploying medical LLMs and provides resources to develop factually reliable medical AI systems.

Deep neural networks have exhibited impressive performance in image classification tasks but remain vulnerable to adversarial examples. Standard adversarial training enhances robustness but typically fails to explicitly address inter-class feature overlap, a significant contributor to adversarial susceptibility. In this work, we introduce a novel adversarial training framework that actively mitigates inter-class proximity by projecting out inter-class dependencies from adversarial and clean samples in the feature space. Specifically, our approach first identifies the nearest inter-class neighbors for each adversarial sample and subsequently removes projections onto these neighbors to enforce stronger feature separability. Theoretically, we demonstrate that our proposed logits correction reduces the Lipschitz constant of neural networks, thereby lowering the Rademacher complexity, which directly contributes to improved generalization and robustness. Extensive experiments across standard benchmarks including CIF AR-10, CIF AR-100, SVHN, and TinyImagenet show that our method demonstrates strong performance that is competitive with leading adversarial training techniques, highlighting significant achievements in both robust and clean accuracy. Our findings reveal the importance of addressing inter-class feature proximity explicitly to bolster adversarial robustness in DNNs. The code is available in the supplementary material.

Text generating capabilities have undergone a substantial transformation with the introduction of large language models (LLMs). Electroencephalography (EEG)-based text production is still difficult, though, because it requires a lot of data and processing power. This paper introduces a new method that combines the use of the Gemma 2B LLM with a classifier-LLM architecture to incorporate a Recurrent Neural Network (RNN) encoder. Our approach drastically lowers the amount of data and compute power needed while achieving performance close to that of cutting-edge methods. Notably, compared to current methodologies, our methodology delivers an overall performance improvement of 10%. The suggested architecture demonstrates the possibility of effective transfer learning for EEG-based text production, remaining strong and functional even in the face of data limits. This work highlights the potential of integrating LLMs with EEG decoding to improve assistive technologies and improve independence and communication for those with severe motor limitations. Our method pushes the limits of present capabilities and opens new paths for research and application in brain-computer interfaces by efficiently using the strengths of pre-trained language models. This makes EEG-based text production more accessible and efficient.

Sebocytes are lipid - secreting cells whose differentiation is marked by the accumulation of intracellular lipid droplets, making their quantification a key readout in sebocyte biology. Manual counting is labor - intensive and subjective, motivating automated solutions. Here, we introduce a simple attention - based multiple instance learning (MIL) framework for sebocyte image analysis. Nile Red - stained sebocyte images were annotated into 14 classes according to droplet counts, expanded via data augmentation to ab out 50,000 cells. Two models were benchmarked: a baseline multi - layer perceptron (MLP) trained on aggregated patch - level counts, and an attention - based MIL model leveraging precomputed ResNet - 50 feature embeddings with trainable instance weighting. Experiments using five - fold cross - validation showed that the baseline MLP achieved more stable performance (mean MAE = 5.6) compared with the attention - based MIL, which was less consistent (mean MAE = 10.7) but occasionally superior in specific folds. The se findings indicate that simple bag - level aggregation provides a robust baseline for slide - level droplet counting, while attention - based MIL requires task - aligned pooling and regularization to fully realize its potential in sebocyte image analysis.

Euclidean geometry has historically played a central role in cultivating logical reasoning and abstract thinking within mathematics education, but has experienced waning emphasis in recent curricula. The resurgence of interest, driven by advances in artificial intelligence and educational technology, has highlighted geometry's potential to develop essential cognitive skills and inspired new approaches to automated problem solving and proof verification. This article presents an ontology-based framework for annotating and optimizing geometry problem sets, originally developed in the 1990s. The ontology systematically classifies geometric problems, solutions, and associated skills into interlinked facts, objects, and methods, supporting granular tracking of student abilities and facilitating curriculum design. The core concept of 'solution graphs'--directed acyclic graphs encoding multiple solution pathways and skill dependencies--enables alignment of problem selection with instructional objectives. We hypothesize that this framework also points toward automated solution validation via semantic parsing. We contend that our approach addresses longstanding challenges in representing dynamic, procedurally complex mathematical knowledge, paving the way for adaptive, feedback-rich educational tools. Our methodology offers a scalable, adaptable foundation for future advances in intelligent geometry education and automated reasoning.

Significant progress in reward modeling over recent years has been driven by a paradigm shift from task-specific designs towards generalist reward models. Despite this trend, developing effective reward models remains a fundamental challenge: the heavy reliance on large-scale labeled preference data. Pre-training on abundant unlabeled data offers a promising direction, but existing approaches fall short of instilling explicit reasoning into reward models. To bridge this gap, we propose a self-training approach that leverages unlabeled data to elicit reward reasoning in reward models. Based on this approach, we develop GRAM-R$^2$, a generative reward model trained to produce not only preference labels but also accompanying reward rationales. GRAM-R$^2$ can serve as a foundation model for reward reasoning and can be applied to a wide range of tasks with minimal or no additional fine-tuning. It can support downstream applications such as response ranking and task-specific reward tuning. Experiments on response ranking, task adaptation, and reinforcement learning from human feedback demonstrate that GRAM-R$^2$ consistently delivers strong performance, outperforming several strong discriminative and generative baselines.

We present NoLBERT, a lightweight, timestamped foundational language model for empirical research -- particularly for forecasting in economics, finance, and the social sciences. By pretraining exclusively on text from 1976 to 1995, NoLBERT avoids both lookback and lookahead biases (information leakage) that can undermine econometric inference. It exceeds domain-specific baselines on NLP benchmarks while maintaining temporal consistency. Applied to patent texts, NoLBERT enables the construction of firm-level innovation networks and shows that gains in innovation centrality predict higher long-run profit growth.