Nan Li

Nan Li, Bo Kang, Tijl De Bie

We present SkillGPT, a tool for skill extraction and standardization (SES) from free-style job descriptions and user profiles with an open-source Large Language Model (LLM) as backbone. Most previous methods for similar tasks either need supervision or rely on heavy data-preprocessing and feature engineering. Directly prompting the latest conversational LLM for standard skills, however, is slow, costly and inaccurate. In contrast, SkillGPT utilizes a LLM to perform its tasks in steps via summarization and vector similarity search, to balance speed with precision. The backbone LLM of SkillGPT is based on Llama, free for academic use and thus useful for exploratory research and prototype development. Hence, our cost-free SkillGPT gives users the convenience of conversational SES, efficiently and reliably.

Rui Bao, Yaping Sun, Zhiyong Chen et al.

Edge deployments of generative inference increasingly face two practical realities: per-device per-model performance is often unknown at deployment time, and it is non-stationary due to user-driven semantic events, background load, and device churn. Consequently, a resource manager that is tuned offline under a fixed regime can become brittle and expensive to maintain. This paper presents $E^3$-Agent, an executable and evolving agent for edge artificial intelligence generated content (AIGC) resource management. $E^3$-Agent separates a fast-path router that makes millisecond-level dispatch decisions from a slow-path, event-driven large language model (LLM) meta-controller that mitigates regime shifts through a small, explicit control surface exposed via a tool interface, including risk gating, router configuration, and rapid performance calibration. The agent learns online from execution feedback and continuously adapts to unknown and time-varying service-time mappings. We evaluate $E^3$-Agent in a discrete-event simulator driven by MLPerf-derived device-model measurement priors, covering cold-start warmup and three dynamic regimes: semantic dynamics, device churn, and hidden drift. Across the dynamic scenarios, $E^3$-Agent reduces average latency by 65%-73% compared to the best static baseline, stays within 7%-10% of an online full-information Oracle used for evaluation, and effectively suppresses stutter rate under semantic degradation.

Nan Li, Alexandros Iosifidis, Qi Zhang

This paper studies the computational offloading of CNN inference in device-edge co-inference systems. Inspired by the emerging paradigm semantic communication, we propose a novel autoencoder-based CNN architecture (AECNN), for effective feature extraction at end-device. We design a feature compression module based on the channel attention method in CNN, to compress the intermediate data by selecting the most important features. To further reduce communication overhead, we can use entropy encoding to remove the statistical redundancy in the compressed data. At the receiver, we design a lightweight decoder to reconstruct the intermediate data through learning from the received compressed data to improve accuracy. To fasten the convergence, we use a step-by-step approach to train the neural networks obtained based on ResNet-50 architecture. Experimental results show that AECNN can compress the intermediate data by more than 256x with only about 4% accuracy loss, which outperforms the state-of-the-art work, BottleNet++. Compared to offloading inference task directly to edge server, AECNN can complete inference task earlier, in particular, under poor wireless channel condition, which highlights the effectiveness of AECNN in guaranteeing higher accuracy within time constraint.

Nan Li, Alexandros Iosifidis, Qi Zhang

This paper studies the computational offloading of CNN inference in dynamic multi-access edge computing (MEC) networks. To address the uncertainties in communication time and Edge servers' available capacity, we use early-exit mechanism to terminate the computation earlier to meet the deadline of inference tasks. We design a reward function to trade off the communication, computation and inference accuracy, and formulate the offloading problem of CNN inference as a maximization problem with the goal of maximizing the average inference accuracy and throughput in long term. To solve the maximization problem, we propose a graph reinforcement learning-based early-exit mechanism (GRLE), which outperforms the state-of-the-art work, deep reinforcement learning-based online offloading (DROO) and its enhanced method, DROO with early-exit mechanism (DROOE), under different dynamic scenarios. The experimental results show that GRLE achieves the average accuracy up to 3.41x over graph reinforcement learning (GRL) and 1.45x over DROOE, which shows the advantages of GRLE for offloading decision-making in dynamic MEC.

Nan Li, Alexandros Iosifidis, Qi Zhang

This paper studies inference acceleration using distributed convolutional neural networks (CNNs) in collaborative edge computing. To ensure inference accuracy in inference task partitioning, we consider the receptive-field when performing segment-based partitioning. To maximize the parallelization between the communication and computing processes, thereby minimizing the total inference time of an inference task, we design a novel task collaboration scheme in which the overlapping zone of the sub-tasks on secondary edge servers (ESs) is executed on the host ES, named as HALP. We further extend HALP to the scenario of multiple tasks. Experimental results show that HALP can accelerate CNN inference in VGG-16 by 1.7-2.0x for a single task and 1.7-1.8x for 4 tasks per batch on GTX 1080TI and JETSON AGX Xavier, which outperforms the state-of-the-art work MoDNN. Moreover, we evaluate the service reliability under time-variant channel, which shows that HALP is an effective solution to ensure high service reliability with strict service deadline.

Yutong Li, Nan Li, H. Eric Tseng et al.

The action governor is an add-on scheme to a nominal control loop that monitors and adjusts the control actions to enforce safety specifications expressed as pointwise-in-time state and control constraints. In this paper, we introduce the Robust Action Governor (RAG) for systems the dynamics of which can be represented using discrete-time Piecewise Affine (PWA) models with both parametric and additive uncertainties and subject to non-convex constraints. We develop the theoretical properties and computational approaches for the RAG. After that, we introduce the use of the RAG for realizing safe Reinforcement Learning (RL), i.e., ensuring all-time constraint satisfaction during online RL exploration-and-exploitation process. This development enables safe real-time evolution of the control policy and adaptation to changes in the operating environment and system parameters (due to aging, damage, etc.). We illustrate the effectiveness of the RAG in constraint enforcement and safe RL using the RAG by considering their applications to a soft-landing problem of a mass-spring-damper system.

Nan Li, Lihong Zhi

In this paper, we generalize the algorithm described by Rump and Graillat, as well as our previous work on certifying breadth-one singular solutions of polynomial systems, to compute verified and narrow error bounds such that a slightly perturbed system is guaranteed to possess an isolated singular solution within the computed bounds. Our new verification method is based on deflation techniques using smoothing parameters. We demonstrate the performance of the algorithm for systems with singular solutions of multiplicity up to hundreds.

Nan Li, Lihong Zhi

In this paper we describe how to improve the performance of the symbolic-numeric method in (Li and Zhi,2009, 2011) for computing the multiplicity structure and refining approximate isolated singular solutions in the breadth one case. By introducing a parameterized and deflated system with smoothing parameters, we generalize the algorithm in (Rump and Graillat, 2009) to compute verified error bounds such that a slightly perturbed polynomial system is guaranteed to have a breadth-one multiple root within the computed bounds.

Andrea Cavagna, Nan Li, Alexandros Iosifidis et al.

This paper aims to design robust Edge Intelligence using semantic communication for time-critical IoT applications. We systematically analyze the effect of image DCT coefficients on inference accuracy and propose the channel-agnostic effectiveness encoding for offloading by transmitting the most meaningful task data first. This scheme can well utilize all available communication resource and strike a balance between transmission latency and inference accuracy. Then, we design an effectiveness decoding by implementing a novel image augmentation process for convolutional neural network (CNN) training, through which an original CNN model is transformed into a Robust CNN model. We use the proposed training method to generate Robust MobileNet-v2 and Robust ResNet-50. The proposed Edge Intelligence framework consists of the proposed effectiveness encoding and effectiveness decoding. The experimental results show that the effectiveness decoding using the Robust CNN models perform consistently better under various image distortions caused by channel errors or limited communication resource. The proposed Edge Intelligence framework using semantic communication significantly outperforms the conventional approach under latency and data rate constraints, in particular, under ultra stringent deadlines and low data rate.

Nan Li, Alexandros Iosifidis, Qi Zhang

This paper studies inference acceleration using distributed convolutional neural networks (CNNs) in collaborative edge computing network. To avoid inference accuracy loss in inference task partitioning, we propose receptive field-based segmentation (RFS). To reduce the computation time and communication overhead, we propose a novel collaborative edge computing using fused-layer parallelization to partition a CNN model into multiple blocks of convolutional layers. In this scheme, the collaborative edge servers (ESs) only need to exchange small fraction of the sub-outputs after computing each fused block. In addition, to find the optimal solution of partitioning a CNN model into multiple blocks, we use dynamic programming, named as dynamic programming for fused-layer parallelization (DPFP). The experimental results show that DPFP can accelerate inference of VGG-16 up to 73% compared with the pre-trained model, which outperforms the existing work MoDNN in all tested scenarios. Moreover, we evaluate the service reliability of DPFP under time-variant channel, which shows that DPFP is an effective solution to ensure high service reliability with strict service deadline.

Xiaohui Liu, Meng Liu, Lin Zhang et al.

The Audio Deep Synthesis Detection (ADD) Challenge has been held to detect generated human-like speech. With our submitted system, this paper provides an overall assessment of track 1 (Low-quality Fake Audio Detection) and track 2 (Partially Fake Audio Detection). In this paper, spectro-temporal artifacts were detected using raw temporal signals, spectral features, as well as deep embedding features. To address track 1, low-quality data augmentation, domain adaptation via finetuning, and various complementary feature information fusion were aggregated in our system. Furthermore, we analyzed the clustering characteristics of subsystems with different features by visualization method and explained the effectiveness of our proposed greedy fusion strategy. As for track 2, frame transition and smoothing were detected using self-supervised learning structure to capture the manipulation of PF attacks in the time domain. We ranked 4th and 5th in track 1 and track 2, respectively.

Nan Li, Lihong Zhi

We present a symbolic-numeric method to refine an approximate isolated singular solution $\hat{\mathbf{x}}=(\hat{x}_{1}, ..., \hat{x}_{n})$ of a polynomial system $F=\{f_1, ..., f_n\}$ when the Jacobian matrix of $F$ evaluated at $\hat{\mathbf{x}}$ has corank one approximately. Our new approach is based on the regularized Newton iteration and the computation of approximate Max Noether conditions satisfied at the approximate singular solution. The size of matrices involved in our algorithm is bounded by $n \times n$. The algorithm converges quadratically if $\hat{\xx}$ is close to the isolated exact singular solution.

Peiyuan Fang, Weiqi Zhang, Lu Xiong et al.

This paper introduces the Generalized Action Governor (AG), a supervisory scheme that augments a nominal closed-loop system with the capability to enforce state and input constraints through online action adjustment. We develop a generalized AG theory for discrete-time systems under bounded uncertainties, and relax the usual requirement of positive invariance to returnability of a safe set. Based on the theory, we present tailored AG design procedures for linear systems and for discrete systems with finite state and action spaces. We further study safe online learning enabled by the AG and present two safe learning strategies, namely safe Q-learning and safe data-driven Koopman operator-based control, both integrated with the AG to guarantee constraint satisfaction during learning. Numerical results illustrate the proposed methods.

Nan Li, Bo Kang, Tijl De Bie

Automated occupation extraction and standardization from free-text job postings and resumes are crucial for applications like job recommendation and labor market policy formation. This paper introduces LLM4Jobs, a novel unsupervised methodology that taps into the capabilities of large language models (LLMs) for occupation coding. LLM4Jobs uniquely harnesses both the natural language understanding and generation capacities of LLMs. Evaluated on rigorous experimentation on synthetic and real-world datasets, we demonstrate that LLM4Jobs consistently surpasses unsupervised state-of-the-art benchmarks, demonstrating its versatility across diverse datasets and granularities. As a side result of our work, we present both synthetic and real-world datasets, which may be instrumental for subsequent research in this domain. Overall, this investigation highlights the promise of contemporary LLMs for the intricate task of occupation extraction and standardization, laying the foundation for a robust and adaptable framework relevant to both research and industrial contexts.

Zhongtian Dong, Nan Li, Alexandros Iosifidis et al.

For time-critical IoT applications using deep learning, inference acceleration through distributed computing is a promising approach to meet a stringent deadline. In this paper, we implement a working prototype of a new distributed inference acceleration method HALP using three raspberry Pi 4. HALP accelerates inference by designing a seamless collaboration among edge devices (EDs) in Edge Computing. We maximize the parallelization between communication and computation among the collaborative EDs by optimizing the task partitioning ratio based on the segment-based partitioning. Experimental results show that the distributed inference HALP achieves 1.7x inference acceleration for VGG-16. Then, we combine distributed inference with conventional neural network model compression by setting up different shrinking hyperparameters for MobileNet-V1. In this way, we can further accelerate inference but at the cost of inference accuracy loss. To strike a balance between latency and accuracy, we propose dynamic model selection to select a model which provides the highest accuracy within the latency constraint. It is shown that the model selection with distributed inference HALP can significantly improve service reliability compared to the conventional stand-alone computation.

Di Huang, Rui Zhang, Xing Hu et al.

Aiming to find a program satisfying the user intent given input-output examples, program synthesis has attracted increasing interest in the area of machine learning. Despite the promising performance of existing methods, most of their success comes from the privileged information of well-designed input-output examples. However, providing such input-output examples is unrealistic because it requires the users to have the ability to describe the underlying program with a few input-output examples under the training distribution. In this work, we propose a query-based framework that trains a query neural network to generate informative input-output examples automatically and interactively from a large query space. The quality of the query depends on the amount of the mutual information between the query and the corresponding program, which can guide the optimization of the query framework. To estimate the mutual information more accurately, we introduce the functional space (F-space) which models the relevance between the input-output examples and the programs in a differentiable way. We evaluate the effectiveness and generalization of the proposed query-based framework on the Karel task and the list processing task. Experimental results show that the query-based framework can generate informative input-output examples which achieve and even outperform well-designed input-output examples.

Zhenghao Liu, Zulong Chen, Moufeng Zhang et al.

This paper proposes the User Viewing Flow Modeling (SINGLE) method for the article recommendation task, which models the user constant preference and instant interest from user-clicked articles. Specifically, we first employ a user constant viewing flow modeling method to summarize the user's general interest to recommend articles. In this case, we utilize Large Language Models (LLMs) to capture constant user preferences from previously clicked articles, such as skills and positions. Then we design the user instant viewing flow modeling method to build interactions between user-clicked article history and candidate articles. It attentively reads the representations of user-clicked articles and aims to learn the user's different interest views to match the candidate article. Our experimental results on the Alibaba Technology Association (ATA) website show the advantage of SINGLE, achieving a 2.4% improvement over previous baseline models in the online A/B test. Our further analyses illustrate that SINGLE has the ability to build a more tailored recommendation system by mimicking different article viewing behaviors of users and recommending more appropriate and diverse articles to match user interests.

Nan Li, Bo Kang, Jefrey Lijffijt et al.

In settings such as e-recruitment and online dating, recommendation involves distributing limited opportunities, calling for novel approaches to quantify and enforce fairness. We introduce \emph{inferiority}, a novel (un)fairness measure quantifying a user's competitive disadvantage for their recommended items. Inferiority complements \emph{envy}, a fairness notion measuring preference for others' recommendations. We combine inferiority and envy with \emph{utility}, an accuracy-related measure of aggregated relevancy scores. Since these measures are non-differentiable, we reformulate them using a probabilistic interpretation of recommender systems, yielding differentiable versions. We combine these loss functions in a multi-objective optimization problem called \texttt{FEIR} (Fairness through Envy and Inferiority Reduction), applied as post-processing for standard recommender systems. Experiments on synthetic and real-world data demonstrate that our approach improves trade-offs between inferiority, envy, and utility compared to naive recommendations and the baseline methods.

Peng Jia, Ruiqi Sun, Nan Li et al.

Strong lensing in galaxy clusters probes properties of dense cores of dark matter halos in mass, studies the distant universe at flux levels and spatial resolutions otherwise unavailable, and constrains cosmological models independently. The next-generation large scale sky imaging surveys are expected to discover thousands of cluster-scale strong lenses, which would lead to unprecedented opportunities for applying cluster-scale strong lenses to solve astrophysical and cosmological problems. However, the large dataset challenges astronomers to identify and extract strong lensing signals, particularly strongly lensed arcs, because of their complexity and variety. Hence, we propose a framework to detect cluster-scale strongly lensed arcs, which contains a transformer-based detection algorithm and an image simulation algorithm. We embed prior information of strongly lensed arcs at cluster-scale into the training data through simulation and then train the detection algorithm with simulated images. We use the trained transformer to detect strongly lensed arcs from simulated and real data. Results show that our approach could achieve 99.63 % accuracy rate, 90.32 % recall rate, 85.37 % precision rate and 0.23 % false positive rate in detection of strongly lensed arcs from simulated images and could detect almost all strongly lensed arcs in real observation images. Besides, with an interpretation method, we have shown that our method could identify important information embedded in simulated data. Next step, to test the reliability and usability of our approach, we will apply it to available observations (e.g., DESI Legacy Imaging Surveys) and simulated data of upcoming large-scale sky surveys, such as the Euclid and the CSST.

Haosu Zhou, Nan Li

As the complexity of modern manufacturing technologies increases, traditional trial-and-error design, which requires iterative and expensive simulations, becomes unreliable and time-consuming. This difficulty is especially significant for the design of hot-stamped safety-critical components, such as ultra-high-strength-steel (UHSS) B-pillars. To reduce design costs and ensure manufacturability, scalar-based Artificial-Intelligence-empowered surrogate modelling (SAISM) has been investigated and implemented, which can allow real-time manufacturability-constrained structural design optimisation. However, SAISM suffers from low accuracy and generalisability, and usually requires a high volume of training samples. To solve this problem, an image-based Artificial-intelligence-empowered surrogate modelling (IAISM) approach is developed in this research, in combination with an auto-decoder-based blank shape generator. The IAISM, which is based on a Mask-Res-SE-U-Net architecture, is trained to predict the full thinning field of the as-formed component given an arbitrary blank shape. Excellent prediction performance of IAISM is achieved with only 256 training samples, which indicates the small-data learning nature of engineering AI tasks using structured data representations. The trained auto-decoder, trained Mask-Res-SE-U-Net, and Adam optimiser are integrated to conduct blank optimisation by modifying the latent vector. The optimiser can rapidly find blank shapes that satisfy manufacturability criteria. As a high-accuracy and generalisable surrogate modelling and optimisation tool, the proposed pipeline is promising to be integrated into a full-chain digital twin to conduct real-time, multi-objective design optimisation.

Tao Wang, Nan Li

Open vocabulary object detection has been greatly advanced by the recent development of vision-language pretrained model, which helps recognize novel objects with only semantic categories. The prior works mainly focus on knowledge transferring to the object proposal classification and employ class-agnostic box and mask prediction. In this work, we propose CondHead, a principled dynamic network design to better generalize the box regression and mask segmentation for open vocabulary setting. The core idea is to conditionally parameterize the network heads on semantic embedding and thus the model is guided with class-specific knowledge to better detect novel categories. Specifically, CondHead is composed of two streams of network heads, the dynamically aggregated head and the dynamically generated head. The former is instantiated with a set of static heads that are conditionally aggregated, these heads are optimized as experts and are expected to learn sophisticated prediction. The latter is instantiated with dynamically generated parameters and encodes general class-specific information. With such a conditional design, the detection model is bridged by the semantic embedding to offer strongly generalizable class-wise box and mask prediction. Our method brings significant improvement to the state-of-the-art open vocabulary object detection methods with very minor overhead, e.g., it surpasses a RegionClip model by 3.0 detection AP on novel categories, with only 1.1% more computation.

Nan Li, Bo Kang, Tijl De Bie

When LLMs judge moral dilemmas, do they reach different conclusions in different languages, and if so, why? Two factors could drive such differences: the language of the dilemma itself, or the language in which the model reasons. Standard evaluation conflates these by testing only matched conditions (e.g., English dilemma with English reasoning). We introduce a methodology that separately manipulates each factor, covering also mismatched conditions (e.g., English dilemma with Chinese reasoning), enabling decomposition of their contributions. To study \emph{what} changes, we propose an approach to interpret the moral judgments in terms of Moral Foundations Theory. As a side result, we identify evidence for splitting the Authority dimension into a family-related and an institutional dimension. Applying this methodology to English-Chinese moral judgment with 13 LLMs, we demonstrate its diagnostic power: (1) the framework isolates reasoning-language effects as contributing twice the variance of input-language effects; (2) it detects context-dependency in nearly half of models that standard evaluation misses; and (3) a diagnostic taxonomy translates these patterns into deployment guidance. We release our code and datasets at https://anonymous.4open.science/r/CrossCulturalMoralJudgement.

Yingxue Zhao, Haoran Li, Haosu Zhou et al.

Mechanics-related problems often present unique challenges in achieving accurate geometric and physical representations, particularly for non-uniform structures. Graph neural networks (GNNs) have emerged as a promising tool to tackle these challenges by adeptly learning from graph data with irregular underlying structures. Consequently, recent years have witnessed a surge in complex mechanics-related applications inspired by the advancements of GNNs. Despite this process, there is a notable absence of a systematic review addressing the recent advancement of GNNs in solving mechanics-related problems. To bridge this gap, this review article aims to provide an in-depth overview of the GNN applications in mechanics-related domains while identifying key challenges and outlining potential future research directions. In this review article, we begin by introducing the fundamental algorithms of GNNs that are widely employed in mechanics-related applications. We provide a concise explanation of their underlying principles to establish a solid understanding that will serve as a basis for exploring the applications of GNNs in mechanics-related domains. The scope of this paper is intended to cover the categorisation of literature into solid mechanics, fluid mechanics, and interdisciplinary mechanics-related domains, providing a comprehensive summary of graph representation methodologies, GNN architectures, and further discussions in their respective subdomains. Additionally, open data and source codes relevant to these applications are summarised for the convenience of future researchers. This article promotes an interdisciplinary integration of GNNs and mechanics and provides a guide for researchers interested in applying GNNs to solve complex mechanics-related problems.

Yong He, Cheng Wang, Shun Zhang et al.

Medical text learning has recently emerged as a promising area to improve healthcare due to the wide adoption of electronic health record (EHR) systems. The complexity of the medical text such as diverse length, mixed text types, and full of medical jargon, poses a great challenge for developing effective deep learning models. BERT has presented state-of-the-art results in many NLP tasks, such as text classification and question answering. However, the standalone BERT model cannot deal with the complexity of the medical text, especially the lengthy clinical notes. Herein, we develop a new model called KG-MTT-BERT (Knowledge Graph Enhanced Multi-Type Text BERT) by extending the BERT model for long and multi-type text with the integration of the medical knowledge graph. Our model can outperform all baselines and other state-of-the-art models in diagnosis-related group (DRG) classification, which requires comprehensive medical text for accurate classification. We also demonstrated that our model can effectively handle multi-type text and the integration of medical knowledge graph can significantly improve the performance.

Nan Li, Jiming Ren, Haris Miller et al.

Multi-Agent Task Assignment and Planning (MATP) has attracted growing attention but remains challenging in terms of scalability, spatial reasoning, and adaptability in obstacle-rich environments. To address these challenges, we propose OATH - Adaptive Obstacle-Aware Task Assignment and Planning for Heterogeneous Robot Teaming - which advances MATP by introducing a novel obstacle-aware strategy for task assignment. First, we develop an adaptive Halton sequence map, the first known application of Halton sampling with obstacle-aware adaptation in MATP, which adjusts sampling density based on obstacle distribution. Second, we propose a cluster-auction-selection framework that integrates obstacle-aware clustering with weighted auctions and intra-cluster task selection. These mechanisms jointly enable effective coordination among heterogeneous robots while maintaining scalability and suboptimal allocation performance. In addition, our framework leverages an LLM to interpret human instructions and directly guide the planner in real time. We validate OATH in both NVIDIA Isaac Sim and real-world hardware experiments using TurtleBot platforms, demonstrating substantial improvements in task assignment quality, scalability, adaptability to dynamic changes, and overall execution performance compared to state-of-the-art MATP baselines. A project website is available at https://llm-oath.github.io/.

Tao Wang, Nan Li

Open vocabulary object detection has been greatly advanced by the recent development of vision-language pretrained model, which helps recognize novel objects with only semantic categories. The prior works mainly focus on knowledge transferring to the object proposal classification and employ class-agnostic box and mask prediction. In this work, we propose CondHead, a principled dynamic network design to better generalize the box regression and mask segmentation for open vocabulary setting. The core idea is to conditionally parameterize the network heads on semantic embedding and thus the model is guided with class-specific knowledge to better detect novel categories. Specifically, CondHead is composed of two streams of network heads, the dynamically aggregated head and the dynamically generated head. The former is instantiated with a set of static heads that are conditionally aggregated, these heads are optimized as experts and are expected to learn sophisticated prediction. The latter is instantiated with dynamically generated parameters and encodes general class-specific information. With such a conditional design, the detection model is bridged by the semantic embedding to offer strongly generalizable class-wise box and mask prediction. Our method brings significant improvement to the state-of-the-art open vocabulary object detection methods with very minor overhead, e.g., it surpasses a RegionClip model by 3.0 detection AP on novel categories, with only 1.1% more computation.

Peng Jia, Jiameng Lv, Runyu Ning et al.

Large-scale astronomical surveys can capture numerous images of celestial objects, including galaxies and nebulae. Analysing and processing these images can reveal intricate internal structures of these objects, allowing researchers to conduct comprehensive studies on their morphology, evolution, and physical properties. However, varying noise levels and point spread functions can hamper the accuracy and efficiency of information extraction from these images. To mitigate these effects, we propose a novel image restoration algorithm that connects a deep learning-based restoration algorithm with a high-fidelity telescope simulator. During the training stage, the simulator generates images with different levels of blur and noise to train the neural network based on the quality of restored images. After training, the neural network can directly restore images obtained by the telescope, as represented by the simulator. We have tested the algorithm using real and simulated observation data and have found that it effectively enhances fine structures in blurry images and increases the quality of observation images. This algorithm can be applied to large-scale sky survey data, such as data obtained by LSST, Euclid, and CSST, to further improve the accuracy and efficiency of information extraction, promoting advances in the field of astronomical research.

Yang Han, Zhiqiang Zou, Nan Li et al.

Astronomical outliers, such as unusual, rare or unknown types of astronomical objects or phenomena, constantly lead to the discovery of genuinely unforeseen knowledge in astronomy. More unpredictable outliers will be uncovered in principle with the increment of the coverage and quality of upcoming survey data. However, it is a severe challenge to mine rare and unexpected targets from enormous data with human inspection due to a significant workload. Supervised learning is also unsuitable for this purpose since designing proper training sets for unanticipated signals is unworkable. Motivated by these challenges, we adopt unsupervised machine learning approaches to identify outliers in the data of galaxy images to explore the paths for detecting astronomical outliers. For comparison, we construct three methods, which are built upon the k-nearest neighbors (KNN), Convolutional Auto-Encoder (CAE)+ KNN, and CAE + KNN + Attention Mechanism (attCAE KNN) separately. Testing sets are created based on the Galaxy Zoo image data published online to evaluate the performance of the above methods. Results show that attCAE KNN achieves the best recall (78%), which is 53% higher than the classical KNN method and 22% higher than CAE+KNN. The efficiency of attCAE KNN (10 minutes) is also superior to KNN (4 hours) and equal to CAE+KNN(10 minutes) for accomplishing the same task. Thus, we believe it is feasible to detect astronomical outliers in the data of galaxy images in an unsupervised manner. Next, we will apply attCAE KNN to available survey datasets to assess its applicability and reliability.

Jun Wang, Xijuan Zeng, Chunyu Qiang et al.

We propose Kling-Foley, a large-scale multimodal Video-to-Audio generation model that synthesizes high-quality audio synchronized with video content. In Kling-Foley, we introduce multimodal diffusion transformers to model the interactions between video, audio, and text modalities, and combine it with a visual semantic representation module and an audio-visual synchronization module to enhance alignment capabilities. Specifically, these modules align video conditions with latent audio elements at the frame level, thereby improving semantic alignment and audio-visual synchronization. Together with text conditions, this integrated approach enables precise generation of video-matching sound effects. In addition, we propose a universal latent audio codec that can achieve high-quality modeling in various scenarios such as sound effects, speech, singing, and music. We employ a stereo rendering method that imbues synthesized audio with a spatial presence. At the same time, in order to make up for the incomplete types and annotations of the open-source benchmark, we also open-source an industrial-level benchmark Kling-Audio-Eval. Our experiments show that Kling-Foley trained with the flow matching objective achieves new audio-visual SOTA performance among public models in terms of distribution matching, semantic alignment, temporal alignment and audio quality.

Xuan Xia, Weijie Lv, Xing He et al.

Although industrial anomaly detection (AD) technology has made significant progress in recent years, generating realistic anomalies and learning priors of normal remain challenging tasks. In this study, we propose an end-to-end industrial anomaly detection method called FractalAD. Training samples are obtained by synthesizing fractal images and patches from normal samples. This fractal anomaly generation method is designed to sample the full morphology of anomalies. Moreover, we designed a backbone knowledge distillation structure to extract prior knowledge contained in normal samples. The differences between a teacher and a student model are converted into anomaly attention using a cosine similarity attention module. The proposed method enables an end-to-end semantic segmentation network to be used for anomaly detection without adding any trainable parameters to the backbone and segmentation head, and has obvious advantages over other methods in training and inference speed.. The results of ablation studies confirmed the effectiveness of fractal anomaly generation and backbone knowledge distillation. The results of performance experiments showed that FractalAD achieved competitive results on the MVTec AD dataset and MVTec 3D-AD dataset compared with other state-of-the-art anomaly detection methods.

Jiahui Peng, Jing Gao, Xin Tong et al.

In the evolving field of corporate sustainability, analyzing unstructured Environmental, Social, and Governance (ESG) reports is a complex challenge due to their varied formats and intricate content. This study introduces an innovative methodology utilizing the "Unstructured Core Library", specifically tailored to address these challenges by transforming ESG reports into structured, analyzable formats. Our approach significantly advances the existing research by offering high-precision text cleaning, adept identification and extraction of text from images, and standardization of tables within these reports. Emphasizing its capability to handle diverse data types, including text, images, and tables, the method adeptly manages the nuances of differing page layouts and report styles across industries. This research marks a substantial contribution to the fields of industrial ecology and corporate sustainability assessment, paving the way for the application of advanced NLP technologies and large language models in the analysis of corporate governance and sustainability. Our code is available at https://github.com/linancn/TianGong-AI-Unstructure.git.

Yingxue Zhao, Haoran Li, Haosu Zhou et al.

Finite element simulations of large-deformation sheet material forming involve node-element coupling between nodal kinematics and element-level deformation measures. Machine-learning surrogates can accelerate such simulations, but most graph-based models use node-centred representations. This representation is indirect for element-level quantities, which are often recovered from nodal predictions by interpolation or post-processing. It may also obscure the node-element coupling structure that underlies the finite element update. This work proposes a cross-attention-based bipartite graph neural network (CAtt-BiGNN) for coupled prediction of nodal displacement increments and elemental thinning. The graph represents mesh nodes and elements as distinct but connected entities, linked by directed node-element edges, so that nodal and elemental fields are predicted on their native discretisation domains. An edge-aware cross-attention processor conditions adaptive node-element coupling weights on geometric edge features, enabling bidirectional message passing between nodal kinematic states and elemental deformation states. A hierarchical extension, CAtt-BiUGNN, combines the CAtt-BiGNN with graph downsampling-upsampling to improve information propagation on larger meshes. Adaptive Gaussian noise is further evaluated as an optional rollout-stabilisation strategy. The models are tested on two representative forming cases with different graph sizes. CAtt-BiGNN improves the balance between displacement and thinning prediction relative to node-centred baselines and bipartite ablation variants, while CAtt-BiUGNN gives the strongest overall performance in the larger-graph setting. The results indicate that the proposed model provides an effective surrogate framework for large-deformation sheet material forming.

Jiajie Luo, Mohamed Mohamed, Osama Hassan et al.

Traditional sheet metal forming relies on time-consuming and expensive Finite Element Analysis (FEA) for design validation, a process that significantly prolongs design cycles. While surrogate models offer faster iteration, current approaches have limitations: scalar-based methods cannot capture comprehensive field-based FEA results, while existing image-based models often ignore the critical role of material properties by focusing solely on geometry. To address this gap, we develop a physics-guided deep learning framework, namely StampFormer, which simultaneously uses component geometry and material stress-strain responses to predict FEA outcomes. The StampFormer framework uses three core components to process data. A Material-Augmented Geometric Network (MAGN) first fuses geometric and material data. This information is then integrated at various levels by a Hierarchical Material Embedding Injection Unit (HMEIU) before being processed by the primary network backbone, an adapted Swin-UNet. We evaluated our model on the stamping of a crossmember panel with two simulation datasets for steel and aluminium panels, and results demonstrate that StampFormer provides high-fidelity predictions of critical physical fields - including thinning, major strain, minor strain, plastic strain, and displacement - in under a second. Compared with ground truth FEA, our model achieved an average relative error of less than 8.5% on the four 2D fields and a mean squared error of less than 1.2 mm2 for the 3D displacement field. In summary, we introduce a practical and efficient framework that integrates multimodal information, namely geometry and material properties, to provide fast and accurate predictions, enabling designers to perform real-time manufacturability assessments.

Haoran Li, Tobias Lehrer, Yingxue Zhao et al.

Nonlinear finite element crash simulations are accurate but computationally expensive, limiting their use in iterative design optimisation. Machine-learning surrogate models based on graph neural networks (GNNs) offer a faster alternative. Message-passing GNNs are widely used for mesh simulation, and their shared node and edge update functions are relatively generalisable across varying graph structures. By contrast, non-shareable edge-specific aggregation layers can capture nonlinear relationships more accurately but usually require fixed graph connectivity, which limits generalisability. This paper presents Mask-Morph Graph U-Net (MMGUNet), a practical approach to addressing the limitation of hierarchical Graph U-Net architectures that use edge-specific downsampling and upsampling layers. Fixed coarse graph connectivity is required for edge-specific layers. To retain this while improving spatial correspondence, the proposed method morphs the coarsened graph hierarchy to each input mesh using feature-aligned barycentric parameterisation before constructing cross-graph edges. It further applies node masking during supervised pretraining, followed by parameter-efficient fine-tuning in which high-parameter edge-specific layers are frozen. The proposed approach is evaluated in in-distribution, out-of-distribution, and cross-component transfer settings using mean Euclidean distance and maximum intrusion percentage error. Results show that coarse-graph morphing improves test accuracy relative to a fixed-coarse-graph baseline, while masked supervised pretraining reduces the train-test discrepancy and improves data efficiency during transfer. The proposed model also achieves lower prediction error compared with external baselines. These results demonstrate a practical route toward reusable, data-efficient mesh-based surrogate modelling for crashworthiness design exploration.

Jing Guo, Nan Li, Ming Xu

Generative AI holds significant potential for ecological and environmental applications such as monitoring, data analysis, education, and policy support. However, its effectiveness is limited by the lack of a unified evaluation framework. To address this, we present the Environmental Large Language model Evaluation (ELLE) question answer (QA) dataset, the first benchmark designed to assess large language models and their applications in ecological and environmental sciences. The ELLE dataset includes 1,130 question answer pairs across 16 environmental topics, categorized by domain, difficulty, and type. This comprehensive dataset standardizes performance assessments in these fields, enabling consistent and objective comparisons of generative AI performance. By providing a dedicated evaluation tool, ELLE dataset promotes the development and application of generative AI technologies for sustainable environmental outcomes. The dataset and code are available at https://elle.ceeai.net/ and https://github.com/CEEAI/elle.

Qianshi Wang, Xilong Qu, Wenbin Pei et al.

Influence maximization (IM) is a fundamental problem in complex network analysis, with a wide range of real-world applications. To date, existing approaches to influential node identification in IM have predominantly relied on standard graphs, failing to capture higher-order intrinsic interactions embedded in many real-world systems. Hypergraphs can be employed to better capture higher-order interactions. However, using hypergraphs may lead to an excessively large search space and increased complexity in modeling cascading dynamics, making it challenging to accurately identify influential nodes. Therefore, in this study, we propose a new hypergraph-modeled IM method, based on the Discrete Particle Swarm Optimization algorithm and the threshold model. In the proposed method, a particle (i.e., a candidate solution) represents the selection information of seed nodes, and the fitness function is designed to accurately and efficiently evaluate the influence of seed nodes via a two-layer local influence approximation. We also propose a degree-based initialization strategy to improve the quality of initial solutions and develop rules for updating particles' velocity and position, incorporated with a local search to drive particles toward better solutions. Experimental results demonstrate that the proposed method outperforms baseline methods on both synthetic and real-world hypergraphs. In addition, ablation studies validate the effectiveness of both the local search and the initialization strategies.

Xingchen Zhou, Nan Li, Peng Jia et al.

Source extraction is crucial in analyzing data from next-generation, large-scale sky surveys in radio bands, such as the Square Kilometre Array (SKA). Several source extraction programs, including SoFiA and Aegean, have been developed to address this challenge. However, finding optimal parameter configurations when applying these programs to real observations is non-trivial. For example, the outcomes of SoFiA intensely depend on several key parameters across its preconditioning, source-finding, and reliability-filtering modules. To address this issue, we propose a framework to automatically optimize these parameters using an AI agent based on a state-of-the-art reinforcement learning (RL) algorithm, i.e., Soft Actor-Critic (SAC). The SKA Science Data Challenge 2 (SDC2) dataset is utilized to assess the feasibility and reliability of this framework. The AI agent interacts with the environment by adjusting parameters based on the feedback from the SDC2 score defined by the SDC2 Team, progressively learning to select parameter sets that yield improved performance. After sufficient training, the AI agent can automatically identify an optimal parameter configuration that outperform the benchmark set by Team SoFiA within only 100 evaluation steps and with reduced time consumption. Our approach could address similar problems requiring complex parameter tuning, beyond radio band surveys and source extraction. Yet, high-quality training sets containing representative observations and catalogs of ground truth are essential.

Ziqi Chen, Yi Ren, Yixuan Huang et al.

Integrating AI into the physical layer is a cornerstone of 6G networks. However, current data-driven approaches struggle to generalize across dynamic environments because they lack an intrinsic understanding of electromagnetic wave propagation. We introduce the Wireless World Model (WWM), a multi-modal foundation framework predicting the spatiotemporal evolution of wireless channels by internalizing the causal relationship between 3D geometry and signal dynamics. Pre-trained on a massive ray-traced multi-modal dataset, WWM overcomes the data authenticity gap, further validated under real-world measurement data. Using a joint-embedding predictive architecture with a multi-modal mixture-of-experts Transformer, WWM fuses channel state information, 3D point clouds, and user trajectories into a unified representation. Across the five key downstream tasks supported by WWM, it achieves remarkable performance in seen environments, unseen generalization scenarios, and real-world measurements, consistently outperforming SOTA uni-modal foundation models and task-specific models. This paves the way for physics-aware 6G intelligence that adapts to the physical world.

Chi Wan, Yixin Cui, Jiatong Du et al.

End-to-end autonomous driving requires adaptive and robust handling of complex and diverse traffic environments. However, prevalent single-mode planning methods attempt to learn an overall policy while struggling to acquire diversified driving skills to handle diverse scenarios. Therefore, this paper proposes GEMINUS, a Mixture-of-Experts end-to-end autonomous driving framework featuring a Global Expert and a Scene-Adaptive Experts Group, equipped with a Dual-aware Router. Specifically, the Global Expert is trained on the overall dataset, possessing robust performance. The Scene-Adaptive Experts are trained on corresponding scene subsets, achieving adaptive performance. The Dual-aware Router simultaneously considers scenario-level features and routing uncertainty to dynamically activate expert modules. Through the effective coupling of the Global Expert and the Scene-Adaptive Experts Group via the Dual-aware Router, GEMINUS achieves both adaptability and robustness across diverse scenarios. GEMINUS outperforms existing methods in the Bench2Drive closed-loop benchmark and achieves state-of-the-art performance in Driving Score and Success Rate, even with only monocular vision input. The code is available at https://github.com/newbrains1/GEMINUS.

Nan Li, Wei Feng, Qian Zhang

Active camera relocalization (ACR) is a new problem in computer vision that significantly reduces the false alarm caused by image distortions due to camera pose misalignment in fine-grained change detection (FGCD). Despite the fruitful achievements that ACR can support, it still remains a challenging problem caused by the unstable results of relative pose estimation, especially for outdoor scenes, where the lighting condition is out of control, i.e., the twice observations may have highly varied illuminations. This paper studies an illumination-invariant active camera relocalization method, it improves both in relative pose estimation and scale estimation. We use plane segments as an intermediate representation to facilitate feature matching, thus further boosting pose estimation robustness and reliability under lighting variances. Moreover, we construct a linear system to obtain the absolute scale in each ACR iteration by minimizing the image warping error, thus, significantly reduce the time consume of ACR process, it is nearly $1.6$ times faster than the state-of-the-art ACR strategy. Our work greatly expands the feasibility of real-world fine-grained change monitoring tasks for cultural heritages. Extensive experiments tests and real-world applications verify the effectiveness and robustness of the proposed pose estimation method using for ACR tasks.

Nan Li, Bo Kang, Tijl De Bie

Creating robust occupation taxonomies, vital for applications ranging from job recommendation to labor market intelligence, is challenging. Manual curation is slow, while existing automated methods are either not adaptive to dynamic regional markets (top-down) or struggle to build coherent hierarchies from noisy data (bottom-up). We introduce CLIMB (CLusterIng-based Multi-agent taxonomy Builder), a framework that fully automates the creation of high-quality, data-driven taxonomies from raw job postings. CLIMB uses global semantic clustering to distill core occupations, then employs a reflection-based multi-agent system to iteratively build a coherent hierarchy. On three diverse, real-world datasets, we show that CLIMB produces taxonomies that are more coherent and scalable than existing methods and successfully capture unique regional characteristics. We release our code and datasets at https://anonymous.4open.science/r/CLIMB.

Tianli Liao, Ce Wang, Lei Li et al.

Large parallax between images is an intractable issue in image stitching. Various warping-based methods are proposed to address it, yet the results are unsatisfactory. In this paper, we propose a novel image stitching method using multi-homography warping guided by image segmentation. Specifically, we leverage the Segment Anything Model to segment the target image into numerous contents and partition the feature points into multiple subsets via the energy-based multi-homography fitting algorithm. The multiple subsets of feature points are used to calculate the corresponding multiple homographies. For each segmented content in the overlapping region, we select its best-fitting homography with the lowest photometric error. For each segmented content in the non-overlapping region, we calculate a weighted combination of the linearized homographies. Finally, the target image is warped via the best-fitting homographies to align with the reference image, and the final panorama is generated via linear blending. Comprehensive experimental results on the public datasets demonstrate that our method provides the best alignment accuracy by a large margin, compared with the state-of-the-art methods. The source code is available at https://github.com/tlliao/multi-homo-warp.

Haoyu Jiang, Haiyang Yu, Nan Li et al.

Deep neural networks (DNNs) have been found vulnerable to backdoor attacks, raising security concerns about their deployment in mission-critical applications. There are various approaches to detect backdoor attacks, however they all make certain assumptions about the target attack to be detected and require equal and huge numbers of clean and backdoor samples for training, which renders these detection methods quite limiting in real-world circumstances. This study proposes a novel one-class classification framework called One-class Graph Embedding Classification (OCGEC) that uses GNNs for model-level backdoor detection with only a little amount of clean data. First, we train thousands of tiny models as raw datasets from a small number of clean datasets. Following that, we design a ingenious model-to-graph method for converting the model's structural details and weight features into graph data. We then pre-train a generative self-supervised graph autoencoder (GAE) to better learn the features of benign models in order to detect backdoor models without knowing the attack strategy. After that, we dynamically combine the GAE and one-class classifier optimization goals to form classification boundaries that distinguish backdoor models from benign models. Our OCGEC combines the powerful representation capabilities of graph neural networks with the utility of one-class classification techniques in the field of anomaly detection. In comparison to other baselines, it achieves AUC scores of more than 98% on a number of tasks, which far exceeds existing methods for detection even when they rely on a huge number of positive and negative samples. Our pioneering application of graphic scenarios for generic backdoor detection can provide new insights that can be used to improve other backdoor defense tasks. Code is available at https://github.com/jhy549/OCGEC.

Tianli Liao, Nan Li

Natural image stitching aims to create a single, natural-looking mosaic from overlapped images that capture the same 3D scene from different viewing positions. Challenges inevitably arise when the scene is non-planar and captured by handheld cameras since parallax is non-negligible in such cases. In this paper, we propose a novel image stitching method using depth maps, which generates accurate alignment mosaics against parallax. Firstly, we construct a robust fitting method to filter out the outliers in feature matches and estimate the epipolar geometry between input images. Then, we utilize epipolar geometry to establish pixel-to-pixel correspondences between the input images and render the warped images using the proposed optimal warping. In the rendering stage, we introduce several modules to solve the mapping artifacts in the warping results and generate the final mosaic. Experimental results on three challenging datasets demonstrate that the depth maps of input images enable our method to provide much more accurate alignment in the overlapping region and view-consistent results in the non-overlapping region. We believe our method will continue to work under the rapid progress of monocular depth estimation. The source code is available at https://github.com/tlliao/NIS_depths.

Bo Chen, Alvaro Parra, Jiewei Cao et al.

Deep networks excel in learning patterns from large amounts of data. On the other hand, many geometric vision tasks are specified as optimization problems. To seamlessly combine deep learning and geometric vision, it is vital to perform learning and geometric optimization end-to-end. Towards this aim, we present BPnP, a novel network module that backpropagates gradients through a Perspective-n-Points (PnP) solver to guide parameter updates of a neural network. Based on implicit differentiation, we show that the gradients of a "self-contained" PnP solver can be derived accurately and efficiently, as if the optimizer block were a differentiable function. We validate BPnP by incorporating it in a deep model that can learn camera intrinsics, camera extrinsics (poses) and 3D structure from training datasets. Further, we develop an end-to-end trainable pipeline for object pose estimation, which achieves greater accuracy by combining feature-based heatmap losses with 2D-3D reprojection errors. Since our approach can be extended to other optimization problems, our work helps to pave the way to perform learnable geometric vision in a principled manner. Our PyTorch implementation of BPnP is available on http://github.com/BoChenYS/BPnP.

Zhiqing Zhang, Guojia Fan, Tianyong Liu et al.

Medical image segmentation is critical for diagnosing and treating spinal disorders. However, the presence of high noise, ambiguity, and uncertainty makes this task highly challenging. Factors such as unclear anatomical boundaries, inter-class similarities, and irrational annotations contribute to this challenge. Achieving both accurate and diverse segmentation templates is essential to support radiologists in clinical practice. In recent years, denoising diffusion probabilistic modeling (DDPM) has emerged as a prominent research topic in computer vision. It has demonstrated effectiveness in various vision tasks, including image deblurring, super-resolution, anomaly detection, and even semantic representation generation at the pixel level. Despite the robustness of existing diffusion models in visual generation tasks, they still struggle with discrete masks and their various effects. To address the need for accurate and diverse spine medical image segmentation templates, we propose an end-to-end framework called VerseDiff-UNet, which leverages the denoising diffusion probabilistic model (DDPM). Our approach integrates the diffusion model into a standard U-shaped architecture. At each step, we combine the noise-added image with the labeled mask to guide the diffusion direction accurately towards the target region. Furthermore, to capture specific anatomical a priori information in medical images, we incorporate a shape a priori module. This module efficiently extracts structural semantic information from the input spine images. We evaluate our method on a single dataset of spine images acquired through X-ray imaging. Our results demonstrate that VerseDiff-UNet significantly outperforms other state-of-the-art methods in terms of accuracy while preserving the natural features and variations of anatomy.

Xueying Sun, Zijia Li, Nan Li

This paper investigates singular configurations of the P3P problem. Using local dual space, a systematic algebraic-computational framework is proposed to give a complete geometric stratification for the P3P singular configurations with respect to the multiplicity $μ$ of the camera center $O$: for $μ\ge 2$, $O$ lies on the ``danger cylinder'', for $μ\ge 3$, $O$ lies on one of three generatrices of the danger cylinder associated with the first Morley triangle or the circumcircle, and for $μ\ge 4$, $O$ lies on the circumcircle which indeed corresponds to infinite P3P solutions. Furthermore, a geometric stratification for the complementary configuration $O^\prime$ associated with a singular configuration $O$ is studied as well: for $μ\ge 2$, $O^\prime$ lies on a deltoidal surface associated with the danger cylinder, and for $μ\ge 3$, $O^\prime$ lies on one of three cuspidal curves of the deltoidal surface.

Tingting Yang, Ping Zhang, Mengfan Zheng et al.

This paper introduces WirelessGPT, a pioneering foundation model specifically designed for multi-task learning in wireless communication and sensing. Specifically, WirelessGPT leverages large-scale wireless channel datasets for unsupervised pretraining and extracting universal channel representations, which captures complex spatiotemporal dependencies. In fact,this task-agnostic design adapts WirelessGPT seamlessly to a wide range of downstream tasks, using a unified representation with minimal fine-tuning. By unifying communication and sensing functionalities, WirelessGPT addresses the limitations of task-specific models, offering a scalable and efficient solution for integrated sensing and communication (ISAC). With an initial parameter size of around 80 million, WirelessGPT demonstrates significant improvements over conventional methods and smaller AI models, reducing reliance on large-scale labeled data. As the first foundation model capable of supporting diverse tasks across different domains, WirelessGPT establishes a new benchmark, paving the way for future advancements in multi-task wireless systems.

Jing Guo, Nan Li, Jianchuan Qi et al.

Large language models (LLMs) have achieved impressive linguistic capabilities. However, a key limitation persists in their lack of human-like memory faculties. LLMs exhibit constrained memory retention across sequential interactions, hindering complex reasoning. This paper explores the potential of applying cognitive psychology's working memory frameworks, to enhance LLM architecture. The limitations of traditional LLM memory designs are analyzed, including their isolation of distinct dialog episodes and lack of persistent memory links. To address this, an innovative model is proposed incorporating a centralized Working Memory Hub and Episodic Buffer access to retain memories across episodes. This architecture aims to provide greater continuity for nuanced contextual reasoning during intricate tasks and collaborative scenarios. While promising, further research is required into optimizing episodic memory encoding, storage, prioritization, retrieval, and security. Overall, this paper provides a strategic blueprint for developing LLM agents with more sophisticated, human-like memory capabilities, highlighting memory mechanisms as a vital frontier in artificial general intelligence.

Nan Li, Albert Gatt, Massimo Poesio

Collaborative dialogue relies on participants incrementally establishing common ground, yet in asymmetric settings they may believe they agree while referring to different entities. We introduce a perspectivist annotation scheme for the HCRC MapTask corpus (Anderson et al., 1991) that separately captures speaker and addressee grounded interpretations for each reference expression, enabling us to trace how understanding emerges, diverges, and repairs over time. Using a scheme-constrained LLM annotation pipeline, we obtain 13k annotated reference expressions with reliability estimates and analyze the resulting understanding states. The results show that full misunderstandings are rare once lexical variants are unified, but multiplicity discrepancies systematically induce divergences, revealing how apparent grounding can mask referential misalignment. Our framework provides both a resource and an analytic lens for studying grounded misunderstanding and for evaluating (V)LLMs' capacity to model perspective-dependent grounding in collaborative dialogue.