Knowledge editing aims to update the embedded knowledge within Large Language Models (LLMs). However, existing approaches, whether through parameter modification or external memory integration, often suffer from inconsistent evaluation objectives and experimental setups. To address this gap, we conduct a comprehensive benchmarking study. In addition to fact-level datasets, we introduce more complex event-based datasets and general-purpose datasets drawn from other tasks. Our evaluation covers both instruction-tuned and reasoning-oriented LLMs, under a realistic autoregressive inference setting rather than teacher-forced decoding. Beyond single-edit assessments, we also evaluate multi-edit scenarios to better reflect practical demands. We employ four evaluation dimensions, including portability, and compare all recent methods against a simple and straightforward baseline named Selective Contextual Reasoning (SCR). Empirical results reveal that parameter-based editing methods perform poorly under realistic conditions. In contrast, SCR consistently outperforms them across all settings. This study offers new insights into the limitations of current knowledge editing methods and highlights the potential of context-based reasoning as a more robust alternative.

Recent advances in Multimodal Large Language Models (MLLMs) have shown promising results in integrating diverse modalities such as texts and images. MLLMs are heavily influenced by modality bias, often relying on language while under-utilizing other modalities like visual inputs. This position paper argues that MLLMs are deeply affected by modality bias. Firstly, we diagnose the current state of modality bias, highlighting its manifestations across various tasks. Secondly, we propose a systematic research road-map related to modality bias in MLLMs. Thirdly, we identify key factors of modality bias in MLLMs and offer actionable suggestions for future research to mitigate it. To substantiate these findings, we conduct experiments that demonstrate the influence of each factor: 1. Data Characteristics: Language data is compact and abstract, while visual data is redundant and complex, creating an inherent imbalance in learning dynamics. 2. Imbalanced Backbone Capabilities: The dominance of pretrained language models in MLLMs leads to overreliance on language and neglect of visual information. 3. Training Objectives: Current objectives often fail to promote balanced cross-modal alignment, resulting in shortcut learning biased toward language. These findings highlight the need for balanced training strategies and model architectures to better integrate multiple modalities in MLLMs. We call for interdisciplinary efforts to tackle these challenges and drive innovation in MLLM research. Our work provides a fresh perspective on modality bias in MLLMs and offers insights for developing more robust and generalizable multimodal systems-advancing progress toward Artificial General Intelligence.

Detecting anomalies in radio astronomy is challenging due to the vast amounts of data and the rarity of labeled anomalous examples. Addressing this challenge requires efficient methods capable of identifying unusual radio galaxy morphologies without relying on extensive supervision. This work introduces an innovative approach to anomaly detection based on morphological characteristics of the radio sources using trainable COSFIRE (Combination of Shifted Filter Responses) filters as an efficient alternative to complex deep learning methods. The framework integrates COSFIRE descriptors with an unsupervised Local Outlier Factor (LOF) algorithm to identify unusual radio galaxy morphologies. Evaluations on a radio galaxy benchmark data set demonstrate strong performance, with the COSFIRE-based approach achieving a geometric mean (G-Mean) score of 79%, surpassing the 77% achieved by a computationally intensive deep learning autoencoder. By characterizing normal patterns and detecting deviations, this semi-supervised methodology overcomes the need for anomalous examples in the training set, a major limitation of traditional supervised methods. This approach shows promise for next-generation radio telescopes, where fast processing and the ability to discover unknown phenomena are crucial.

Ransomware remains a critical threat to cybersecurity, yet publicly available datasets for training machine learning-based ransomware detection models are scarce and often have limited sample size, diversity, and reproducibility. In this paper, we introduce MLRan, a behavioural ransomware dataset, comprising over 4,800 samples across 64 ransomware families and a balanced set of goodware samples. The samples span from 2006 to 2024 and encompass the four major types of ransomware: locker, crypto, ransomware-as-a-service, and modern variants. We also propose guidelines (GUIDE-MLRan), inspired by previous work, for constructing high-quality behavioural ransomware datasets, which informed the curation of our dataset. We evaluated the ransomware detection performance of several machine learning (ML) models using MLRan. For this purpose, we performed feature selection by conducting mutual information filtering to reduce the initial 6.4 million features to 24,162, followed by recursive feature elimination, yielding 483 highly informative features. The ML models achieved an accuracy, precision and recall of up to 98.7%, 98.9%, 98.5%, respectively. Using SHAP and LIME, we identified critical indicators of malicious behaviour, including registry tampering, strings, and API misuse. The dataset and source code for feature extraction, selection, ML training, and evaluation are available publicly to support replicability and encourage future research, which can be found at https://github.com/faithfulco/mlran.

The rapid advancement of Large Language Models (LLMs) has resulted in interest in their potential applications within manufacturing systems, particularly in the context of Industry 5.0. However, determining when to implement LLMs versus other Natural Language Processing (NLP) techniques, ontologies or knowledge graphs, remains an open question. This paper offers decision-making guidance for selecting the most suitable technique in various industrial contexts, emphasizing human-robot collaboration and resilience in manufacturing. We examine the origins and unique strengths of LLMs, ontologies, and knowledge graphs, assessing their effectiveness across different industrial scenarios based on the number of domains or disciplines required to bring a product from design to manufacture. Through this comparative framework, we explore specific use cases where LLMs could enhance robotics for human-robot collaboration, while underscoring the continued relevance of ontologies and knowledge graphs in low-dependency or resource-constrained sectors. Additionally, we address the practical challenges of deploying these technologies, such as computational cost and interpretability, providing a roadmap for manufacturers to navigate the evolving landscape of Language based AI tools in Industry 5.0. Our findings offer a foundation for informed decision-making, helping industry professionals optimize the use of Language Based models for sustainable, resilient, and human-centric manufacturing. We also propose a Large Knowledge Language Model architecture that offers the potential for transparency and configuration based on complexity of task and computing resources available.

Reinforcement learning (RL) algorithms have been used recently to align diffusion models with downstream objectives such as aesthetic quality and text-image consistency by fine-tuning them to maximize a single reward function under a fixed KL regularization. However, this approach is inherently restrictive in practice, where alignment must balance multiple, often conflicting objectives. Moreover, user preferences vary across prompts, individuals, and deployment contexts, with varying tolerances for deviation from a pre-trained base model. We address the problem of inference-time multi-preference alignment: given a set of basis reward functions and a reference KL regularization strength, can we design a fine-tuning procedure so that, at inference time, it can generate images aligned with any user-specified linear combination of rewards and regularization, without requiring additional fine-tuning? We propose Diffusion Blend, a novel approach to solve inference-time multi-preference alignment by blending backward diffusion processes associated with fine-tuned models, and we instantiate this approach with two algorithms: DB-MPA for multi-reward alignment and DB-KLA for KL regularization control. Extensive experiments show that Diffusion Blend algorithms consistently outperform relevant baselines and closely match or exceed the performance of individually fine-tuned models, enabling efficient, user-driven alignment at inference-time. The code is available at https://github.com/bluewoods127/DB-2025}{github.com/bluewoods127/DB-2025.

Standing in 2025, at a critical juncture in the pursuit of Artificial General Intelligence (AGI), reinforcement fine-tuning (RFT) has demonstrated significant potential in enhancing the reasoning capability of large language models (LLMs) and has led to the development of cutting-edge AI models such as OpenAI-o1 and DeepSeek-R1. Moreover, the efficient application of RFT to enhance the reasoning capability of multimodal large language models (MLLMs) has attracted widespread attention from the community. In this position paper, we argue that reinforcement fine-tuning powers the reasoning capability of multimodal large language models. To begin with, we provide a detailed introduction to the fundamental background knowledge that researchers interested in this field should be familiar with. Furthermore, we meticulously summarize the improvements of RFT in powering reasoning capability of MLLMs into five key points: diverse modalities, diverse tasks and domains, better training algorithms, abundant benchmarks and thriving engineering frameworks. Finally, we propose five promising directions for future research that the community might consider. We hope that this position paper will provide valuable insights to the community at this pivotal stage in the advancement toward AGI. Summary of works done on RFT for MLLMs is available at https://github.com/Sun-Haoyuan23/Awesome-RL-based-Reasoning-MLLMs.

As connected and automated transportation systems evolve, there is a growing need for federal and state authorities to revise existing laws and develop new statutes to address emerging cybersecurity and data privacy challenges. This study introduces a Retrieval-Augmented Generation (RAG) based Large Language Model (LLM) framework designed to support policymakers by extracting relevant legal content and generating accurate, inquiry-specific responses. The framework focuses on reducing hallucinations in LLMs by using a curated set of domain-specific questions to guide response generation. By incorporating retrieval mechanisms, the system enhances the factual grounding and specificity of its outputs. Our analysis shows that the proposed RAG-based LLM outperforms leading commercial LLMs across four evaluation metrics: AlignScore, ParaScore, BERTScore, and ROUGE, demonstrating its effectiveness in producing reliable and context-aware legal insights. This approach offers a scalable, AI-driven method for legislative analysis, supporting efforts to update legal frameworks in line with advancements in transportation technologies.

Recent advances in LegalAI have primarily focused on individual case judgment analysis, often overlooking the critical appellate process within the judicial system. Appeals serve as a core mechanism for error correction and ensuring fair trials, making them highly significant both in practice and in research. To address this gap, we present the AppealCase dataset, consisting of 10,000 pairs of real-world, matched first-instance and second-instance documents across 91 categories of civil cases. The dataset also includes detailed annotations along five dimensions central to appellate review: judgment reversals, reversal reasons, cited legal provisions, claim-level decisions, and whether there is new information in the second instance. Based on these annotations, we propose five novel LegalAI tasks and conduct a comprehensive evaluation across 20 mainstream models. Experimental results reveal that all current models achieve less than 50% F1 scores on the judgment reversal prediction task, highlighting the complexity and challenge of the appeal scenario. We hope that the AppealCase dataset will spur further research in LegalAI for appellate case analysis and contribute to improving consistency in judicial decision-making.

While contemporary speech separation technologies adeptly process lengthy mixed audio waveforms, they are frequently challenged by the intricacies of real-world environments, including noisy and reverberant settings, which can result in artifacts or distortions in the separated speech. To overcome these limitations, we introduce SepALM, a pioneering approach that employs audio language models (ALMs) to rectify and re-synthesize speech within the text domain following preliminary separation. SepALM comprises four core components: a separator, a corrector, a synthesizer, and an aligner. By integrating an ALM-based end-to-end error correction mechanism, we mitigate the risk of error accumulation and circumvent the optimization hurdles typically encountered in conventional methods that amalgamate automatic speech recognition (ASR) with large language models (LLMs). Additionally, we have developed Chain-of-Thought (CoT) prompting and knowledge distillation techniques to facilitate the reasoning and training processes of the ALM. Our experiments substantiate that SepALM not only elevates the precision of speech separation but also markedly bolsters adaptability in novel acoustic environments.