Organic reaction mechanisms are the stepwise elementary reactions by which reactants form intermediates and products, and are fundamental to understanding chemical reactivity and designing new molecules and reactions. Although large language models (LLMs) have shown promise in understanding chemical tasks such as synthesis design, it is unclear to what extent this reflects genuine chemical reasoning capabilities, i.e., the ability to generate valid intermediates, maintain chemical consistency, and follow logically coherent multi-step pathways. We address this by introducing oMeBench, the first large-scale, expert-curated benchmark for organic mechanism reasoning in organic chemistry. It comprises over 10,000 annotated mechanistic steps with intermediates, type labels, and difficulty ratings. Furthermore, to evaluate LLM capability more precisely and enable fine-grained scoring, we propose oMeS, a dynamic evaluation framework that combines step-level logic and chemical similarity. We analyze the performance of state-of-the-art LLMs, and our results show that although current models display promising chemical intuition, they struggle with correct and consistent multi-step reasoning. Notably, we find that using prompting strategy and fine-tuning a specialist model on our proposed dataset increases performance by 50% over the leading closed-source model. We hope that oMeBench will serve as a rigorous foundation for advancing AI systems toward genuine chemical reasoning.

Continual learning the ability of a neural network to learn multiple sequential tasks without catastrophic forgetting remains a central challenge in developing adaptive artificial intelligence systems. While deep learning models achieve state-of-the-art performance across domains, they remain limited by overfitting and forgetting. This paper introduces Cluster-Aware Replay (CAR), a hybrid continual learning framework that integrates a small, class-balanced replay buffer with a regularization term based on Inter-Cluster Fitness (ICF) in the feature space. The ICF loss penalizes overlapping feature representations between new and previously learned tasks, encouraging geometric separation in the latent space and reducing interference. Using the standard five-task Split CIFAR-10 benchmark with a ResNet-18 backbone, initial experiments demonstrate that CAR better preserves earlier task performance compared to fine-tuning alone. These findings are preliminary but highlight feature-space regularization as a promising direction for mitigating catastrophic forgetting.

HE rapid expansion of the Internet of Things (IoT) has seamlessly integrated consumer electronics (CE) devices--such as smartphones, smartwatches, and laptops--into our daily lives, enabling remote access and connectivity across diverse sectors like e-healthcare, smart cities, and intelligent transportation [1]. The CE market is projected to reach 2.873 billion users by 2025, driven by the capacity of nearly every device to generate and share data [2], [3]. CE networks, composed of heterogeneous devices from various manufacturers, present unique challenges due to large-scale deployment, high device diversity, and limited computational resources [1], [4]. Unlike traditional IT networks, CE devices such as smart home appliances and wearables require lightweight, secure, and low-latency communication [5]. Their traffic is often encrypted, intermittent, and follows irregular patterns, complicating the task of network anomaly detection (NAD) [6]. Security breaches in CE can have severe consequences, including privacy invasion, financial loss, and physical safety risks, and compromised devices can be conscripted into botnets for large-scale attacks like DDoS campaigns [7]-[9]. While existing machine learning (ML) and deep learning (DL) methods for NAD have shown promise, they often suffer from time-consuming feature extraction processes and require extensive manual configuration, making them ill-suited for the dynamic nature of CE networks [10]. To overcome these limitations, advanced architectures like Compute First Networking (CFN) and Software-Defined Networking (SDN) are gaining traction.

Node affinity prediction is a common task that is widely used in temporal graph learning with applications in social and financial networks, recommender systems, and more. Recent works have addressed this task by adapting state-of-the-art dynamic link property prediction models to node affinity prediction. However, simple heuristics, such as Persistent Forecast or Moving Average, outperform these models. In this work, we analyze the challenges in training current Temporal Graph Neural Networks for node affinity prediction and suggest appropriate solutions. Combining the solutions, we develop NAViS - Node Affinity prediction model using Virtual State, by exploiting the equivalence between heuristics and state space models. While promising, training NAViS is non-trivial. Therefore, we further introduce a novel loss function for node affinity prediction. We evaluate NAViS on TGB and show that it outperforms the state-of-the-art, including heuristics. Our source code is available at https://github.com/orfeld415/NAVIS

As large language models (LLMs) advance in role-playing (RP) tasks, existing benchmarks quickly become obsolete due to their narrow scope, outdated interaction paradigms, and limited adaptability across diverse application scenarios. To address this gap, we introduce FURINA-Builder, a novel multi-agent collaboration pipeline that automatically constructs fully customizable RP benchmarks at any scale. It enables evaluation of arbitrary characters across diverse scenarios and prompt formats, as the first benchmark builder in RP area for adaptable assessment. FURINA-Builder simulates dialogues between a test character and other characters drawn from a well-constructed character-scene pool, while an LLM judge selects fine-grained evaluation dimensions and adjusts the test character's responses into final test utterances. Using this pipeline, we build FURINA-Bench, a new comprehensive role-playing benchmark featuring both established and synthesized test characters, each assessed with dimension-specific evaluation criteria. Human evaluation and preliminary separability analysis justify our pipeline and benchmark design. We conduct extensive evaluations of cutting-edge LLMs and find that o3 and DeepSeek-R1 achieve the best performance on English and Chinese RP tasks, respectively. Across all models, established characters consistently outperform synthesized ones, with reasoning capabilities further amplifying this disparity. Interestingly, we observe that model scale does not monotonically reduce hallucinations. More critically, for reasoning LLMs, we uncover a novel trade-off: reasoning improves RP performance but simultaneously increases RP hallucinations. This trade-off extends to a broader Pareto frontier between RP performance and reliability for all LLMs. These findings demonstrate the effectiveness of FURINA-Builder and the challenge posed by FURINA-Bench.

Large language models (LLMs) are increasingly deployed on complex reasoning tasks, yet little is known about their ability to internally evaluate problem difficulty, which is an essential capability for adaptive reasoning and efficient resource allocation. In this work, we investigate whether LLMs implicitly encode problem difficulty in their internal representations. Using a linear probe on the final-token representations of LLMs, we demonstrate that the difficulty level of math problems can be linearly modeled. We further locate the specific attention heads of the final Transformer layer: these attention heads have opposite activation patterns for simple and difficult problems, thus achieving perception of difficulty. Our ablation experiments prove the accuracy of the location. Crucially, our experiments provide practical support for using LLMs as automatic difficulty annotators, potentially substantially reducing reliance on costly human labeling in benchmark construction and curriculum learning. We also uncover that there is a significant difference in entropy and difficulty perception at the token level. Our study reveals that difficulty perception in LLMs is not only present but also structurally organized, offering new theoretical insights and practical directions for future research. Our code is available at https://github.com/Aegis1863/Difficulty-Perception-of-LLMs.

In this paper, we propose AUREXA-SE (Audio-Visual Unified Representation Exchange Architecture with Cross-Attention and Squeezeformer for Speech Enhancement), a progressive bimodal framework tailored for audio-visual speech enhancement (AVSE). AUREXA-SE jointly leverages raw audio waveforms and visual cues by employing a U-Net-based 1D convolutional encoder for audio and a Swin Transformer V2 for efficient and expressive visual feature extraction. Central to the architecture is a novel bidirectional cross-attention mechanism, which facilitates deep contextual fusion between modalities, enabling rich and complementary representation learning. To capture temporal dependencies within the fused embeddings, a stack of lightweight Squeezeformer blocks combining convolutional and attention modules is introduced. The enhanced embeddings are then decoded via a U-Net-style decoder for direct waveform reconstruction, ensuring perceptually consistent and intelligible speech output. Experimental evaluations demonstrate the effectiveness of AUREXA-SE, achieving significant performance improvements over noisy baselines, with STOI of 0.516, PESQ of 1.323, and SI-SDR of -4.322 dB. The source code of AUREXA-SE is available at https://github.com/mtanveer1/AVSEC-4-Challenge-2025.

Intraoperative pathology is pivotal to precision surgery, yet its clinical impact is constrained by diagnostic complexity and the limited availability of high-quality frozen-section data. While computational pathology has made significant strides, the lack of large-scale, prospective validation has impeded its routine adoption in surgical workflows. Here, we introduce CRISP, a clinical-grade foundation model developed on over 100,000 frozen sections from eight medical centers, specifically designed to provide Clinical-grade Robust Intraoperative Support for Pathology (CRISP). CRISP was comprehensively evaluated on more than 15,000 intraoperative slides across nearly 100 retrospective diagnostic tasks, including benign-malignant discrimination, key intraoperative decision-making, and pan-cancer detection, etc. The model demonstrated robust generalization across diverse institutions, tumor types, and anatomical sites-including previously unseen sites and rare cancers. In a prospective cohort of over 2,000 patients, CRISP sustained high diagnostic accuracy under real-world conditions, directly informing surgical decisions in 92.6% of cases. Human-AI collaboration further reduced diagnostic workload by 35%, avoided 105 ancillary tests and enhanced detection of micrometastases with 87.5% accuracy. Together, these findings position CRISP as a clinical-grade paradigm for AI-driven intraoperative pathology, bridging computational advances with surgical precision and accelerating the translation of artificial intelligence into routine clinical practice.

AI coding assistants are rapidly becoming integral to modern software development. A key challenge in this space is the continual need to migrate and modernize codebases in response to evolving software ecosystems. Traditionally, such migrations have relied on rule-based systems and human intervention. With the advent of powerful large language models (LLMs), AI-driven agentic frameworks offer a promising alternative-but their effectiveness has not been systematically evaluated. In this paper, we introduce FreshBrew, a novel benchmark for evaluating AI agents on project-level Java migrations, with a specific focus on measuring an agent's ability to preserve program semantics and avoid reward hacking, which we argue requires projects with high test coverage for a rigorous and reliable evaluation. We benchmark several state-of-the-art LLMs, and compare their performance against established rule-based tools. Our evaluation of AI agents on this benchmark of 228 repositories shows that the top-performing model, Gemini 2.5 Flash, can successfully migrate 52.3 percent of projects to JDK 17. Our empirical analysis reveals novel insights into the critical strengths and limitations of current agentic approaches, offering actionable insights into their real-world applicability. Our empirical study reveals failure modes of current AI agents in realistic Java modernization tasks, providing a foundation for evaluating trustworthy code-migration systems. By releasing FreshBrew, we aim to facilitate rigorous, reproducible evaluation and catalyze progress in AI-driven codebase modernization.

Mathematical reasoning is a primary indicator of large language models (LLMs) intelligence. This paper attributes these deficiencies to spurious reasoning--i.e., producing answers from superficial features. To address this challenge, we propose the AdaR framework to enable adaptive reasoning, wherein models rely on problem-solving logic to produce answers. AdaR synthesizes logically equivalent queries by varying variable values, and trains models with RL VR on these data to penalize spurious logic while encouraging adaptive logic. To improve data quality, we extract the problem-solving logic from the original query and generate the corresponding answer by code execution and then apply sanity check. Experimental results demonstrate that AdaR improves robustness and generalization, achieving substantial improvement in mathematical reasoning while maintaining high data efficiency. Analysis indicates that data synthesis and RL VR function in a coordinated manner to enable adaptive reasoning in LLMs. Subsequent analyses derive key design insights into the effect of critical factors and the applicability to instruct LLMs. Our project is available at https://github.com/NJUNLP/AdaR. Large Language Models (LLMs) have demonstrated strong performance across various reasoning tasks (Wei et al., 2022a; Huang & Chang, 2023). Among these, mathematical reasoning serves as a crucial cognitive skill that supports problem-solving across tasks (Huang & Chang, 2023).